Special Session On Material Flow Accounting: OECD Working Group On Environmental Information and Outlooks (WGEIO)

OECD Working Group on Environmental Information
and Outlooks (WGEIO)
SPECIAL SESSION ON
MATERIAL FLOW ACCOUNTING
Paris, 24 October 2000
PAPERS AND PRESENTATIONS
The Special Session on Material Flow Accounting was organised following a joint proposal by the
United States and Japan at the 29th meeting of the Working Group on Environmental Information
and Outlooks (former Working Group on the State of the Environment) (October 1999, Paris).
The purpose was to review national and international experiences in material flow accounting
(MFA) as part of the WGEIO work programme, and as a contribution to the OECD work programme
on resource efficiency and to the OECD initiative on sustainable development. The objectives were
to:
take stock of the development and use of MFA at national and international level;
facilitate the exchange of experience on MFA, addressing in particular i) the policy-relevance of
MFA, and ii) selected methodological issues (frameworks, definitions, classifications, system
boundaries, etc.);
identify areas for further progress at national and international level.
Questions addressed included the following:

What are the most promising uses of MFA? Examples of the use of information derived from
MFA in decision making.
What are the main methodological and measurement issues?
What indicators could be derived from MFA and how should they be interpreted?
1
WGEIO Special Session on Material Flow Accounting
Working Group on Environmental Information

and Outlooks (WGEIO)
SPECIAL SESSION ON MATERIAL FLOW ACCOUNTING
TABLE OF CONTENTS
INTRODUCTION....................................................................................................................................... 5
CHAIR’S SUMMARY ............................................................................................................................... 9
PART I. DEVELOPMENT AND MOST PROMISING USES ................................................................... 11
HISTORY AND OVERVIEW OF MATERIAL FLOW ANALYSIS, .................................................... 13
1. Material flow analysis and accounting...................................................................................... 13
2. Concepts for sustaining the metabolism of economies............................................................. 14
3. Impacts of material flows and types of indicators .................................................................... 15
4. Types of analysis ...................................................................................................................... 15
5. Use of material flow related analyses ....................................................................................... 17
6. Economy-wide MFA ................................................................................................................ 18
7. MFA based indicators ............................................................................................................... 18
INTERNATIONAL COMPARISON OF INFLOWS AND OUTFLOWS .............................................. 25
MATERIAL FLOW ACCOUNTING: The experience of Canada........................................................... 29
1. Introduction............................................................................................................................... 29
2. Material and Energy Flow Accounts in Canada ....................................................................... 29
3. The MEFA accounting framework ........................................................................................... 31
4. Uses of the Material and Energy Flow Accounts ..................................................................... 32
5. Material and Energy Flow Indicators........................................................................................ 33
6. Data gaps................................................................................................................................... 34
7. Future directions for Material and Energy Flow Accounts....................................................... 34
ANALYSIS OF MATERIAL FLOWS FOR SUSTAINABILITY POLICY: The experience of Italy.... 37
1. Introduction............................................................................................................................... 37
2. Data and method of analysis ..................................................................................................... 38
3. Main results............................................................................................................................... 39
4. Conclusions............................................................................................................................... 44
MATERIAL FLOW ACCOUNTING AND ITS APPLICATION: The experience of Japan.................. 49
1. Background............................................................................................................................... 49
2. Overview of studies on MFA and inter-related tools................................................................ 49
3. Characterization of Japanese Material Flows ........................................................................... 51
4. Use of MFA for the development of environmental indicators ................................................ 53
5. Conclusion and future perspective............................................................................................ 55
MATERIAL FLOW ACCOUNTING: MATERIAL USE FOR NATIONAL CONSUMPTION AND
FOR EXPORT: The experience of Sweden .............................................................................................. 59
Abstract.................................................................................................................................................. 59
1. Introduction............................................................................................................................... 59
2. Method ...................................................................................................................................... 61
3. Results....................................................................................................................................... 61
4. Conclusions............................................................................................................................... 64
OECD 3 2003
PART II. METHODOLOGICAL ISSUES................................................................................................... 67

MATERIAL FLOW ACCOUNTING – FRAMEWORKS AND METHODS: Presentation by Eurostat
................................................................................................................................................................... 69
1. Introduction............................................................................................................................... 69
2. Mapping MFA activities with policy issues ............................................................................. 70
3. Frameworks for MFA ............................................................................................................... 71
4. A framework for economy-wide MFA ..................................................................................... 73
5. Indicators derivable from a material balance............................................................................ 74
6. A sequence of accounts............................................................................................................. 75
7. Basic data.................................................................................................................................. 75
8. Selection and interpretation of indicators ................................................................................. 76
9. The trade bias in material input indicators................................................................................ 76
10. Conclusions and outlook........................................................................................................... 77
MATERIAL FLOW ACCOUNTING – METHODOLOGY AND FRAMEWORKS: Presentation by the
United States ............................................................................................................................................. 79
1. Introduction............................................................................................................................... 79
2. The national economy............................................................................................................... 79
3. Frameworks .............................................................................................................................. 79
4. Boundary Conditions ................................................................................................................ 81
5. Approaches ............................................................................................................................... 82
6. Estimation ................................................................................................................................. 83
7. Indicators .................................................................................................................................. 84
8. Conclusion ................................................................................................................................ 84
PART III. LINKS WITH INDICATORS AND OTHER ACCOUNTING TOOLS.................................... 85
LINKS BETWEEN THE MICRO AND THE MACRO FLOWS: SUBSTANCE FLOW ANALYSIS:
Presentation by the Netherlands, Leiden University ................................................................................. 87
1. Introduction............................................................................................................................... 87
2. Material Flow Accounting (MFA)............................................................................................ 87
3. Bulk-MFA................................................................................................................................. 88
4. Substance Flow Analysis (SFA) ............................................................................................... 88
5. Links between bulk-MFA en SFA............................................................................................ 92
LINKS BETWEEN MACRO AND MICRO FLOWS: THE SECTORAL APPROACH: Presentation by
Austria, IFF ............................................................................................................................................... 97
1. Introduction............................................................................................................................... 97
2. Methodological preconditions for a sectoral approach ............................................................. 99
3. Problems witch mainly occur at the sectoral level.................................................................. 102
4. Discussion............................................................................................................................... 103
LINKS WITH NATURAL RESOURCE ACCOUNTS AND RELATED INCIDATORS: THE CASE
OF WATER RESOURCE & USES ACCOUNTS Presentation by France........................................... 105
1. Objectives of water Resource & Uses accounts...................................................................... 105
2. History .................................................................................................................................... 105
3. Selected recent developments in the continuation of the OECD exercise .............................. 105
4. The accounting framework of water R & U ........................................................................... 106
5. Examples of indicators of material flows derived from the water resource & uses accounts. 107
6. Next steps (very incomplete panorama).................................................................................. 108
LIST OF PARTICIPANTS OF THE SPECIAL SESSION ON MFA ....................................................... 113
OECD 4 2003
INTRODUCTION
Introductory speech by Ms. Joke Waller-Hunter

OECD Director, Environment Directorate
Mr. Chairman, Dear participants,

I am very pleased to introduce this Special session on Material Flows Accounting, organised by the
Working Group on Environmental Information and Outlooks following a proposal of the United States and
Japan last year.
It builds on earlier work carried out by the OECD and the WGEIO in particular on natural resource
accounts and on the use of environmental accounting in decision making. The issues that you will address
today also link closely to other areas of OECD work such as environmental indicators, resource use
efficiency, sustainable development, and will provide useful input into the work we are doing on the
environmental outlook and strategy.
LINK TO SUSTAINABLE DEVELOPMENT
Today, the goal of sustainable development has received wide international acceptance. For the OECD and
its member countries it is a key priority. You probably know that the OECD is carrying out a three-year
programme to help countries move towards sustainable development and to make the concept of
sustainable development operational for practical policies. This project will stretch until early 2001, with
the preparation of policy and analytical reports, including a chapter on measurement and indicators, and
leading to a ministerial dialogue in May 2001 when environment and finance ministers meet together.
Progress towards a more sustainable development path requires that the objectives of increasing economic
efficiency and material wealth be integrated with social and environmental objectives, and placed in an
inter-generational framework. It also implies that the capital base of our economies and societies should be
preserved for future generations.
OECD countries are collectively the biggest users of natural resources in the world and the environmental,
economic and social consequences of the production and consumption patterns of their population and
related economic activities extend far beyond their borders. From an environmental point of view, they
exert three major types of influences within and outside the OECD area. Firstly, on the rate of extraction of
non-renewable resources such as oil and minerals. Secondly, on the extent of harvest of renewable
resources such as forests, agriculture, and wildlife, and of withdrawal of water resources. And thirdly, on
the intensity of environmental stress associated with production, consumption and use of resources.
Consequently, OECD countries have an important role to play in achieving more sustainable production
and consumption patterns, and preserving the capital base of our economies.
The appropriate management of natural resources and their efficient use in our economies, is thus a key to
sd and it is part of the many cross-sectoral issues with which we are increasingly confronted.
OECD 5 2003
LINK TO INFORMATION AND DECISION MAKING
In this context, it is essential that effective policies are developed and implemented, and that environmental
and resource aspects are taken into account at an early stage and throughout the decision making chain. It
is further essential that these policies are based on appropriate factual information. Environmental
accounting contributes to the provision of such information. It is at the basis of information supply and is a
tool from which a number of useful indicators can be derived.
ENVIRONMENTAL ACCOUNTING
Over the past 10 to 20 years, general interest in environmental accounting, articulated by policy makers
and NGOs, has been increasing. A considerable amount of work has been carried out in academia and by
national administrations and international organisations to develop and refine methodologies for various
environmental accounting tools, including environmental satellite accounts, environmental expenditure
accounts, natural resource accounts, and material flows analysis. A lot of the work completed has shed light
on the supply side of environmental accounting. Practical applications have also progressed, mainly in areas
where the demand for accounting tools was clearly identified and linked to specific policy questions such
as the management and planning of natural resource use (e.g. water, forests, energy) and associated
indicator development.
USE OF ENVIRONMENTAL ACCOUNTING
Results from an OECD seminar held in 1994 to review the use of environmental accounting in decision
making, showed indeed that there are three uses of environmental and material resource accounts that are
relevant for decision-makers. They can be used as a tool for:
Resource management: To be suitable for this purpose, the accounts must provide extensive and
detailed information; this tends to require well- developed and large statistical bases and a relatively
sophisticated accounting system, often at the substance level.
Policy analysis: This use is less information-intensive than the use for resource management. Two
types of uses are distinguished:
• Direct use of physical accounts: where NRA trace the flow of natural resources from the
environment to the economy and within the economy, they provide information about the
impacts of sectoral economic activities on the resource flows and stock and vice versa.
• Indirect use of physical accounting information through linking it to economic information in the
context of integrated environmental and economic accounting or through introducing it
into environment-economy models.
Indicator development: the construction of selected and/or aggregated indicators of resource use and
pollution intensities is likely to be the least demanding application of resource accounts in terms of
information requirements.
Last year, the OECD organised a Conference in Rome to take stock of national and international initiatives
in the field of sustainable development indicators. The presentations made at this Conference showed that
the wide variety of indicators of sustainable development already in use, generally include indicators
drawn from accounting frameworks. Most countries that have developed a set of sustainable development
indicators include in their set one or several indicators dealing with material use and related intensities.
The OECD itself uses a number of indicators derived form its work on natural resource accounting. These
indicators are part of the OECD Core Set of environmental indicators and have proven to be useful tools in
OECD 6 2003
our policy analysis and evaluation work. They are also discussed as part of the OECD work on sd
indicators.
OBJECTIVES OF THIS SPECIAL SESSION
In this Special session we wish to review progress made and progress to be made with material flow
accounting at national and international level, and to provide a forum for a constructive exchange of
experience on how to best use the information that can be derived from these accounts to support decision-
making and policy development. As mentioned earlier, a lot of the accounting work that has been carried
out to date, focuses on the supply side. We would like this session to establish a link with the demand side,
and to help countries to make their MFA work operational for national policies.
One important question we have in mind is:
• What are the most promising policy uses of MFA? I.e. to what extent can MFA actually provide a
basis for policy relevant, analytically sound and measurable indicators that i) monitor the use and
management of resources in our economies, and the associated environmental stress; and ii) support
related decision making and policies. In our opinion, such indicators should complement other
indicators derived from natural resource accounting already in use such as the intensity of use of water
resources, the intensity of use of forest resources, or energy indicators such as energy intensities and
efficiencies.
• When answering this question, we need also to have a closer look at methodological and measurement
issues, at system boundaries, at levels of aggregation, and how they affect the interpretation and policy
relevance of derived indicators.
The roundtable at the end of the special session will give the opportunity to further broaden the debate. It
will help to identify what role the different actors can or should play in turning MFA into a useful decision-
making tool, and what role OECD countries as a group could play in refining the MF tool and fostering
progress at international level.
Today’s discussions will also provide a basis for tomorrow’s seminar that will more specifically dwell on
waste material flows. It will consider the potential role of MFA within waste policy generallly, and the use
of MFA as an input to the design and evaluation of waste management and prevention efforts particularly.
Mister Chair, this gives this session a challenging task. Many of the participants have prepared documents
for the meeting, for which we are extremely grateful. I am looking forward to a stimulating exchange of
views and discussions.
OECD 7 2003
OECD Working Group on Environmental Information

and Outlooks - OECD Initiative on Sustainable Development
SPECIAL SESSION ON MATERIAL FLOW ACCOUNTING
CHAIR’S SUMMARY
Following a joint proposal by the United States and Japan at the latest meeting of the Working Group on
Environmental Information and Outlooks (WGEIO), a special session on material flow accounting (MFA) was held
on 24 October 2000, back to back with i) the formal meeting of the WGEIO and ii) a seminar on waste material flows
and resource efficiency. The session was chaired by Mr. Yuichi Moriguchi (Japan) and was attended by more than 70
delegates and experts from OECD member countries and international organisations.
The aim of the special session was to take stock of the work carried out to date, to review progress made and progress
to be made with material flow accounting at national and international level, and to explore how to best use the
information that can be derived from these accounts to support decision-making and policy development.
PROGRESS TO DATE
Over the past 10 to 20 years, general interest in environmental accounting, articulated by policy makers and NGOs,
has been increasing, stimulated by concern with the sustainable use of natural resources and the inappropriate
treatment of the depletion of natural resources in economic accounts. A considerable amount of work has been carried
out in academia and by national administrations and international organisations to develop and refine methodologies
for various environmental accounting tools, including environmental satellite accounts, environmental expenditure
accounts, and natural resource accounts. In recent years, material flow studies have gained renewed interest as a way
of looking at the impacts of economic activity on the environment – the effects of mobilisation of materials from the
environment and the flow of materials to the environment.
The presentations made showed that MFA includes a variety of approaches covering different flows at different levels
of detail for different entities with different system boundaries. Reviewed experiences covered accounting of total
material flows (collaborative effort carried out by Austria, Germany, Japan, the Netherlands, and the United States1),
analysis of flows of specific substances (Netherlands) and accounting of natural resources (France). They were
mainly at nation-wide scale, and examples of disaggregation by sectors and regions were presented. The linkages of
MFA with economic accounting/modelling tools were also reviewed, in particular, applications of economic Input-
Output models (Canada, Italy) for the analysis of indirect flows, and .the integration of MFA with Physical Input-
Output Tables (Germany).
The discussion showed that there are several uses of MFA that are relevant for decision-makers. This includes
indicator development, data provision to modelling, and direct use in waste and resource management. MFA may also
contribute to improvements in environmental data quality (e.g. waste statistics, discharges to water) and vice-versa
better environmental data are essential to improve the quality of basic data for MFA.
LINKS WITH ENVIRONMENTAL AND SUSTAINABILITY INDICATORS

The most promising use that was identified is the derivation of selected and/or aggregated environmental and
sustainability indicators. This is also likely to be the least demanding application of MFA in terms of information
requirements.
Indicators derived from MFA need to be policy relevant, analytically sound and measurable, i.e. feasible. They should i)
monitor the use and management of resources in our economies, and the associated environmental stress; and ii) support
related decision making and policies. MFA indicators can provide an overall metric to measure the material requirements
of economies (resource use), or the absorptive capacity they demand of the environment (pollution). Material or resource
1. Joint research effort involving governmental and non-governmental institutions including the World Resource Institute; the Wuppertal Institute.
OECD 9 2003
efficiency and pollution intensity, can be calculated on a per capita or per GDP basis to enable tracking of environmental
progress and allow international comparisons. More detailed indicators might be used to track changes in material
efficiency in particular industrial sectors, or the pollution intensity for particular media (air, land, water). Such indicators
should be seen as a complement to other indicators derived from natural resource accounting already in use.
Most countries that have developed a set of sustainable development indicators include in their set one or several
indicators dealing with material use and related intensities. Examples are indicators included in the set supporting the
basic environmental plan of Japan, or indicators to monitor policy objectives in Sweden. The OECD itself uses a number
of indicators derived from its work on natural resource accounting. These indicators are part of the OECD Core Set of
environmental indicators and have proven to be useful tools in our policy analysis and evaluation work.
PROGRESS TO BE MADE
The discussions also showed that a lot of the work completed to date has been carried out by universities and research
institutes, and has shed light on the supply side of environmental accounting. The actual use in policy making is still
limited, and the potential of MFA as a policy making tool is often not sufficiently known.
This is also due to the fact that MFA covers many aspects within an integrated framework, and that the responsibilities
for related policy use and data gathering are often scattered and handled by different administrations. Roles of national
environmental and statistical institutions, research communities and international organisations should be further
examined in order to strengthen the interface between policy-relevant uses and methodological developments, to
improve data availability and quality, and to improve harmonisation of methodologies. Also, conceptual approaches still
vary and merit further clarification and convergence.
If MFA has to be turned into a useful decision-making tool progress will need to be made to:
• Further explore and explain the links between MFA and other tools and indicators, and in particular the
links and complementarities between indicators derived from MFA, other indicators describing natural
resource use and other environmental indicators;
• Further explore how methodological issues, data quality and availability affect the interpretation and
policy relevance of derived indicators;
• Further identify, within the broad scope of sustainable development, crucial issues to which MFA
could contribute, in order to attract more focused attention of policy makers to the potential of MFA;
• Further harmonise methodologies for constructing simple MFA and provide guidance to countries on
how best to use MF information to calculate indicators;
• Improve co-operation and communication between the various actors involved, including academia,
national statistical offices, and environmental policy administrations.
NEXT STEPS
The discussion showed that MFA is a potentially powerful tool, especially for Member countries of the OECD, which
still account for a major part of the world’s resource consumption and waste generation. OECD countries as a group
could thus play an important role in refining the MFA tool for use at national level and in fostering progress at
international level. Possible first actions include:
• To invite Member country Delegates to express their interest in contributing to expand work to other
countries
• To encourage voluntary contributions to construct simple MF indicators for member countries as an
input to ongoing work on environmental and sustainable development indicators at national and
international level;
• To set up opportunities for further exchanges of experiences and for tracking methodological progress,
e.g. by meetings, documentation and electronic communications.
OECD 10 2003
PART I. DEVELOPMENT AND MOST PROMISING USES
OECD 11 2003
HISTORY AND OVERVIEW OF MATERIAL FLOW ANALYSIS2,3
1. Material flow analysis and accounting
Material flow analysis (MFA) refers to the analysis of the throughput of process chains
comprising the extraction or harvest, chemical transformation, manufacturing, consumption, recycling, and
disposal of materials. It is based on accounts in physical units (usually in terms of tons) quantifying the
inputs and outputs of those processes. The subjects of the accounting are chemically defined substances
(e.g. carbon or carbon dioxide) on the one hand and natural or technical compounds or ‘bulk’ materials
(e.g. coal, wood) on the other hand.
MFA has become a fast growing field of research with increasing policy relevance. All studies
are based on the common paradigm of industrial metabolism and use the methodological principle of mass
balancing. However, there are various methodological approaches which are based on different goals,
concepts and target questions, although each study may claim to contribute to sustaining the industrial
metabolism. In 1996 the network ConAccount was established to provide a platform for information
exchange on MFA (www.conaccount.net). A first inventory on MFA projects and activities was provided
(Bringezu et al. 1998a). Several meetings took place (Bringezu et al. 1997a, 1998b, Kleijn et al. 1999) and
a research and development agenda was defined through an interactive process (Bringezu et al. 1998c).
MFA has often been linked to the systems perspective of the metabolism of society ("industrial
metabolism" coined by Ayres 1989). This paradigm perspective has been rooted in different scientific
disciplines (Fischer-Kowalski 1997). MFA has been used for the analysis of biogeochemical cycles and the
analysis of natural ecosystems. For the current debate on the necessities and possibilities to sustain the
metabolism of industrial economies the analytical approaches to the interactions of man and nature deserve
prior attention. To this respect historical traits relate to the report of the "President´s Materials Policy
Commission" ("Paley report" 1952), the Princeton conference on "Man´s role in changing the
environment" (Thomas 1956), and the Scientific American Issues of Sept. 1970 and Sept. 1971 on
substance flows and energetic metabolism.
Pioneering work was performed by Wolman (1965) on the average metabolism of a city.
Boulding (1996) stressed the necessary shift from an "cowboy economy" towards a "spaceman economy."
The first analyses of material exchange between the US economy and environment had been performed by
Ayres and Kneese (1969). Later in the 1980ies the analysis of various substance flows was performed to
reduce the risk of exposure to hazardous chemicals such as heavy metals. In the 1990ies the analysis of the
total material throughput of economies again attracted attention before the background of the sustainability
debate and due to the rising demand for adequate indicators for policy support.
2 Paper prepared by Mr. Stefan Bringezu, Wuppertal Institute.

3. A more detailed description of the methodological issues of MFA together with extended review of the literature is presented in the
contribution of S. Bringezu and Y. Moriguchi: Material Flow Analysis. In: R.U. Ayres and L. Ayres: Handbook of Industrial Ecology. Edwar
Elgar, forthcoming.
OECD 13 2003
2. Concepts for sustaining the metabolism of economies
The diversity of MFA approaches derives from different conceptual background. The basic
concept common to many studies is that the industrial system together with its societal interactions is
embedded into the biogeosphere system, thus being dependant upon factors critical for the coexistence of
both systems (Ayres and Simonis 1994, Baccini and Brunner 1991) (Figure 1). The paradigm vision of a
sustainable industrial system is characterised by minimised and consistent physical exchanges between
human society and the environment with the internal material loops being driven by renewable energy
flows (e.g. Richards et al. 1994). However, different strategies contribute to develop the industrial
metabolism towards a sustainable fashion.
Figure 1. Figure 1. Physical exchange between economy/society and environment
Environment
materials materials
Economy/Society
energy energy
One basic strategy may be described as detoxification of the industrial metabolism. This refers to
the mitigation of the releases of critical substances to the environment by pollution reduction. In a wider
sense, this relates to any specific environmental impact such as toxicity to human beings and other
organisms, eutrophication, acidification, ozone depletion, global warming etc.. Regulatory governmental
actions in terms of substance bans and restrictions of use represented the first measures of environmental
policy in the 1970ies and 1980ies. The introduction of cleaner technology aimed primarily at the mitigation
of critical releases to the environment. As a consequence of the effectiveness of such measures pollution
problems in spatial-temporal short range could be solved. However, transregional and global problems and
the problem shifting to future generations as well as the complexity of the industrial metabolism made it
necessary to analyse the flows of hazardous substances, selected materials or products in a systems-wide
approach, i.e. from cradle-to-grave, and with respect to the interlinkage of different flows.
Another complementary strategy may be regarded as dematerialisation of the industrial
metabolism. Considering the current quantity of primary resource use by industrial economies an increase
of resource efficiency by a factor of 4 to 10 had been proposed (Schmidt-Bleek 1994, Weizsäcker et al.
1995). Meanwhile this goal has been adopted by a variety of international organisations and national
governments4. The factor concept aims at the provision of increased services and value added with reduced
resource requirements5. The concept of eco-efficiency includes not only the major inputs (materials,
energy, water, area) but also the major outputs to the environment (emissions to air, water, waste) and
relates them to the products, services or benefits produced (EEA 1999a, OECD 1998, Verfaillie and
4 On the programme level the factor 4/10 concept was adopted by the special session of the United Nations (UNGASS 1997) and the World
Business Council for Sustainable Development (WBCSD 1998). The environmental ministers of OECD (1996) expected progress towards this
end. Several countries included the aim in political programmes (e.g. Austria, Netherlands, Finland, Sweden; see also Gardener and Sampat
1998). In scandinavian countries research was launched to test the broad scale feasibility of factor 4/10 (Nordic Council of Ministers 1999). In
Germany the draft for an environmental policy programme (BMU 1998) refers to a factor of 2.5 increase in productivity of non-renewable raw
materials (1993 to 2020). An increase in eco-efficiency is also being regarded as essential by the environmental ministers of the European
Union (1999). The review of the Fifth (environmental) Action Programme (Decision No 2179/98(EC) emphasises resource use and efficiency.
5. Dematerialisation of the economy may imply a mitigation of all hardware products and thus the throughput of the economy as a whole,
comprising the use of primary and secondary materials. However, dematerialisation may also be directed more specifically to the reduction of
the primary inputs and/or final waste disposal.
OECD 14 2003
Bidwell 2000). However, for the environment the reduction of the absolute impacts through material flows
is essential. Thus, also the quantity of human induced material flows through the industrial system has to
be adjusted to adequate levels of exchange between economy and environment.
3. Impacts of material flows and types of indicators
The impacts of human induced material flows on the environment can be short-term or long-
term, direct or indirect, local to global, predictable or unknown. In general, the impacts are determined by
the volume (weight flow per time period) and the specific impacts per unit of volume flow.
Correspondingly, the impact potentials of material flows can be indicated by volume based and impact
based indicators (Table 1).
Figure 2. Table 1. Indicators for impact potentials of human induced material flows
Volume based indicators Impact based indicators

e.g. e.g.
- Energy Requirements - Global Warming Potential (GWP)
- Material Requirements - Ozone Depletion Potential (ODP)
- Water Consumption - Acidification Potential
Impact based indicators exist only for selected specific effects for which quantitative measures
are available. Most of them relate to substance releases to the environment (output of the
economy/society). Volume based indicators may not be used to indicate specific effects rather than a
generic environmental pressure associated with resource consumption. Usually these indicators relate to
resource extraction (input of the economy/society).
Both types of indicators are complementary in supporting policy for controlling specific hazards
as well as generic impacts of human induced material flows.
4. Types of analysis
Before that background, two basic types of material flow related analyses may be distinguished
according to the primary interest, although in practice a continuum of different approaches exists (Table 2).
Both types I and II are not strictly coincident with the above mentioned two paradigmatic strategies.
However, the importance of the detoxification concept seems highest in Ia and lowest in IIc. Vice versa,
the intention to support dematerialisation seems highest in analyses of IIc and lowest in Ia. Nevertheless
both complementary strategies are increasingly being combined, especially in Ic and IIa. Whereas type I
analyses are often performed from a technical engineering perspective, type II analyses are more directed
to the socio-economic relationship.
Type Ia
Substance flow analysis (SFA) has been used to determine the main entrance routes to the
environment, the processes associated with these emissions, the stocks and flows within the industrial
system as well as the trans-media flows, chemical, physical, biological transformations and resulting
concentrations in the environment. Spacio-temporal distribution is of high concern in SFA. Results from
these analyses are often used as inputs to further analyses for quantitatively assessing risks to substance-
specific endpoints. A variety of studies has been conducted on heavy metals, nutrients, carbon, and
halogenated compounds.
OECD 15 2003
Type Ib
Selected bulk material flows have been studied for various reasons. The resource extraction by
mining and quarrying was studied to assess the geomorphic and hydrological impacts. The flow of biomass
from human production was analysed to evaluate the pressure to species diversity.
Table 1. Table 2. Types of material flow related analysis (after Bringezu and Kleijn 1997)
Type of analysis I II
a b c a b c
Objects of Specific environmental problems related Problems of environmental concern related to
primary interest to certain impacts the throughput
per unit of flow of of
Substances Materials Products Firms Sectors Regions
e.g. e.g. e.g e.g. e.g. e.g.
Cd, Cl, Pb, wooden diapers, single plants, production total or main
Zn, Hg, N, products, batteries, medium and sectors, throughput,
P, C, CO2, energy cars big chemical mass flow
CFC carriers, companies industry, balance, total
excavation, construction material
biomass, requirement
plastics
within certain associated with
firms, sectors, regions substances, materials, products
On the one hand metals like aluminium, timber products like pulp and paper, construction
aggregates represent important base materials for industrial purposes. On the other hand these flows –
although per se rather harmless - may be linked with other flows significantly burdening the environment.
For instance, the ‘red mud’ problem with alumina production and the energy intensive production of
aluminium. Base materials such as plastics have been subject to various studies on the potentials and
environmental consequences of recycling and cascading use.
Type Ic
When the environmental impacts of certain products and services is the primary interest in the
approach is normally addressed as Life Cycle Assessment (LCA). In general, the system boundary of LCA
(`cradle-to-grave´) corresponds with the systems perspective of the anthroposphere, technosphere or
physical economy.
Type IIa
Following the sequence of Table 2 the primary interest may lie in the metabolic performance of a
firm or household, a sector or a region. In this case, there may be no or insufficient information about
specific environmental problems. Often the main task is to evaluate the throughput of those entities in
order to find the major problems, support priority setting, check the possibilities for improvement measures
and provide tools for monitoring their effectiveness.
Accounting for the physical throughput of a firm is becoming a more and more regular feature at
least for bigger companies, as it is sometimes found in corporate environmental reporting. Materials
accounts are used for environmental management. Attention is paid to indicate eco-efficiency on the firm
OECD 16 2003
level6. Flow analyses of materials have been applied for optimisation within companies. However, the
limited scope of firm accounts calls for complementary analyses with a wider systems perspective, either
through LCA type analyses for infrastructures and main products or by analyses of higher aggregates of
production and consumption, i.e. analyses of total production sectors or whole economies.
Type IIb
When the primary interest is devoted to certain industrial sectors or fields of activity, MFA may
be used to identify the most critical fluxes in terms of quality and/or quantity. For instance, different
industrial sectors may be compared with regard to various inputs and outputs either from other sectors or
the environment. Certain sectors or activities may be analysed in detail, e.g. the construction sector or
activities such as nutrition, cleaning, dwelling and working, transport and communication.
Type IIc
A major field of MFA represents the analysis of the metabolism of cities, regions, and national or
supranational economies. The accounting may be directed to selected substances and materials or to total
material input, output and throughput. "Economy-wide MFA" on the national level has attracted special
attention (see below). The main interest lies in the overall characterisation of the metabolic performance of
the studied entities, in order to understand the volume, structure and quality of the throughput and to assess
the status and trend with regard to sustainability.
The term MFA has usually been referred to analyses of type Ia, Ib, IIb, and IIc. Studies of type Ic
are being dealt with under the heading of LCA. Accounting of type IIa is mainly related to environmental
management. There are also combinations of regional and product oriented analyses. Nevertheless, all of
those analyses use the accounting of material inputs and outputs of processes in a quantitative manner, and
many of them apply a systems or chain perspective.
5. Use of material flow related analyses

In general MFA provides a system-analytical view of various interlinked processes and flows to
support the strategic and priority oriented design of management measures. Following the history of
environmental protection since the sixties, Type Ia analyses were applied to control the flow of hazardous
substances. The results contributed to public policy in different ways (Bovenkerk 1998 and Hansen 1998):
• The analyses assisted in finding a consensus on the data which is an important pre-requisite
for policy measures.
• MFA lead to new insights and to changes in environmental policy7.
• The analyses discovered new problems (e.g. the mercury stocks in chlorine plants).
• They also contributed to find new solutions (e.g. source oriented input reduction in the case
of non degradable substances).
The use and policy relevance of Type II analyses has been increased in recent years in the following
aspects (Bringezu 2000):
− the support for policy debate on goals and targets, especially with regard to the resource and
eco-efficiency debate and the integration of environmental and economic policies,
− the number of companies performing firm and product accounts,
− the provision of economy-wide material flow accounts for regular use by official statistics,
6. See WBCSD publications, for method overview and pilot study results Verfaillie and Bidwell (2000), for program activities Executive Brief
Jan. 99’Measuring eco-efficiency with cross-comparable indicators’, Geneva, www.wbcsd.ch.
7. e.g. abandoning the aim of closed chlorine cycling in favour of controlling the most hazardous emissions.
OECD 17 2003
− the derivation of indicators for progress towards sustainability.
6. Economy-wide MFA
Material flow accounts may quantify the physical exchange of national economies with the
environment8. Aggregated material flow balances comprise domestic resource extraction and imports
(inputs) and domestic releases to the environment and exports (outputs) (FIGURE 2). Upstream or
downstream flows associated with imports and exports (resource requirements or emissions) may also be
accounted. A sectoral disaggregation can be provided by physical input-output tables.
MFA was applied by official statistics within the framework of integrated environmental and
economic accounting (SEEA) and a methodological guide has been prepared by EUROSTAT (2000).
National material accounts exist for Austria, Denmark, Germany, Finland, Italy, Japan, the Netherlands,
Sweden, United Kingdom and USA. Work is ongoing for China, Egypt, and Amazonia.
Figure 3. Figure 1. Economy-wide material flows (prepared by IFF for Matthews et al. 2000)
Add. Air Water

and Vapour
Water
Imports Exports
Foreign Economic
Hidden Flows
DMI
Processing Domestic
Processed
Domestic
TMR Extraction Outputs DPO
(to Air, Land,
TDO and Water)
Stock
Domestic Domestic
Hidden Flows Hidden Flows
Domestic Environment
7. MFA based indicators

Material flow accounts provide an important basis for the derivation of environmental indicators
and indicators for sustainability (Berkhout 1999, Jimenez-Beltran 1998, Ministry of the Environment
1999). In order to monitor and assess the environmental performance of national and regional economies a
variety of indicator systems has been proposed (Moldan et al. 1997). As a framework the Driving Force-
Pressure-State-Impact-Response9 (DPSIR) scheme has been established (EEA 1999, 1999a, OECD 1998a).
The extraction of resources on the input side and the release of emissions and waste on the output side
relate to environmental pressures, (sectoral) activities represent driving forces, the flows may change the
state of environment which give rise to various impacts and the societal or political response may influence
the metabolic situation towards sustainability.
8. After the first comprehensive approaches of Ayres and Kneese (1969), domestic MFA had been established independently for Austria (Steurer
1992), Japan (Japanese Environmental Agency 1992) and Germany (Schuetz and Bringezu 1993).
9. Since the early nineties as PSR used especially by OECD.
OECD 18 2003
MFA based indicators have been introduced to official reports to provide an overview on the
headline issues of resource use, waste disposal and emissions to air and water as well as eco-efficiency
(EEA 2000, DETR 1999, Hoffren 1999).
On the one hand, economy-wide material flow accounts provide a more comprehensive picture of
the industrial metabolism than single indicators. On the other hand, they can be used to derive several
parameters which - when taken in time series and for international comparison - provide certain aggregated
information on the metabolic performance of national or regional economies (Figure 2). First international
comparisons have been provided on input and resource efficiency indicators by Adriaanse et al. (1997), on
output and balance indicators by Matthews et al. (2000).
Input indicators
Direct Material Input (DMI) measures the input of used materials into the economy, i.e. all
materials which are of economic value and are used in production and consumption activities; DMI equals
domestic (used) extraction plus imports. Materials which are extracted by economic activities but that do
not normally serve as input for production or consumption activities (mining overburden, etc.) have been
termed ‘hidden flows’ or ‘ecological rucksacks’10. They are not used for further processing and are usually
without economic value. DMI plus unused domestic extraction comprises Total (domestic) Material Input.
Total Material Requirement (TMR)11 includes, in addition to TMI, the upstream hidden material
flows which are associated with imports and predominantly burden the environment in other countries. It
measures the total ‘material base’ of an economy, i.e. the total primary resource requirements of the
production activities. Adding these upstream flows converts imports into their ‘primary resource extraction
equivalent’.
Data for TMR and DMI (incl. composition, i.e. input structure of the industrial metabolism) have
been provided for China, Germany, Netherlands, Japan, USA, Poland, Finland and the European Union.
DMI is available for Sweden. Work is ongoing for Italy and Amazonia. TMI was accounted for Australia.
Output indicators
Domestic Processed Output (DPO) represents the total mass of materials which have been used
in the domestic economy, before flowing into the environment. These flows occur at the processing,
manufacturing, use, and final disposal stages of the economic production-consumption chain. Exported
materials are excluded because their wastes occur in other countries. Included in DPO are emissions to air
from commercial energy combustion and other industrial processes, industrial and household wastes
deposited in landfills, material loads in wastewater, materials dispersed into the environment as a result of
product use (dissipative flows), and emissions from incineration plants. Material flows recycled in industry
are not included in DPO.
Total Domestic Output (TDO): The sum of DPO and disposal of unused domestic extraction.
This indicator represents the total quantity of material outputs to the environment released on the domestic
territory by economic activity. Direct Material Output (DMO): The sum of DPO and exports. This
parameter represents the total quantity of direct material outputs leaving the economy after use either
towards the environment or towards the rest of the world. Total Material Output (TMO) includes also
exports and therefore measures the total of material that leaves the economy; TMO equals TDO plus
exports.
10. Hidden flows (Adriaanse et al. 1997) or rucksack flows (Schmidt-Bleek et al. 1998, Bringezu et al. 1996) comprise the primary resource
requirement not entering the product itself; hidden flows of primary production = unused domestic extraction; hidden flows of imports =
unused and used predominantly foreign extraction associated with the production and delivery of the imports.
11. In studies before Adriaanse et al. (1997), TMR had been defined as TMI (Total Material Input) (Bringezu 1997).
OECD 19 2003
Consumption indicators
Domestic material consumption (DMC) measures the total amount of material directly used in an
economy, excluding hidden flows. DMC equals DMI minus exports.
Total material consumption (TMC) measures the total primary material requirement associated
with domestic consumption activities. TMC equals TMR minus exports and their hidden flows.
Balance indicators
Net Additions to Stock (NAS) measures the physical growth rate of an economy. New materials
are added to the economy’s stock each year (gross additions) in buildings and other infrastructure, and
materials incorporated into new durable goods such as cars, industrial machinery, and household
appliances, while old materials are removed from stock as buildings are demolished, and durable goods
disposed of.
Physical Trade Balance (PTB) – measures the physical trade surplus or deficit of an economy.
PTB equals imports minus exports. Physical trade balances may also be defined including hidden flows
associated with imports and exports (e.g. on the basis of TMC accounts).
Efficiency indicators
Services provided or economic performance (in terms of value added or GDP) may be related to
either input or output indicators to provide efficiency measures. For instance, GDP per DMI indicates the
Direct Materials Productivitiy. GDP per TDO measures the economic performance in relation to material
losses to the environment. Setting the value added in relation to the most important inputs and outputs
provides information on the eco-efficiency of an economy. The interpretation of those relative measures
should always consider the trends of the absolute parameters. The latter are usually also provided on a per
capita basis to support international comparison.
MFA and its indicators are going to provide the basis for political measures and the evaluation of
their effectiveness. For that purpose bulk material flow analyses and substance flow analyses can be
combined, and the monitoring of progress towards sustainability could be gradually improved in a stepwise
approach.
OECD 20 2003
REFERENCES
Adriaanse, Albert, S. Bringezu, A. Hammond, Y. Moriguchi, E. Rodenburg, D. Rogich, H. Schütz (1997),

Resource Flows - The Material Basis of Industrial Economies. Ed by: World Resources Institute,
Wuppertal Institute, Netherlands Ministry of Housing, Spatial Planning, and Environment, Japans
National Institute for Environmental Studies. Washington: World Resources Institute Report.
Ayres, Robert U. (1989), ‚Industrial Metabolism.‘, in J. Ausubel, H. Sladovich (Eds.): Technology and
Environment. Washington D.C.
Ayres, Robert U., A.V. Kneese, (1969), ‚Production, Consumption & Externalities.‘ American Economic
Review, 59, 282-296.
Ayres, Robert U., U. E. Simonis (Eds.) (1994), Industrial Metabolism: Restructuring for sustainable
development, Tokyo: UNU Press.
Baccini, Peter, P. H. Brunner (1991), Metabolism of the Anthroposphere, Berlin, New York, Tokyo.
Berkhout, F. (1999): The concept of industrial metabolism and its implications for statistics. Eurostat
Working Papers 2/1999/B/2, Luxembourg.
BMU - Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit (Ed.) (1998), Nachhaltige
Entwicklung in Deutschland - Entwurf eines umweltpolitischen Schwerpunktprogramms. Bonn:
BMU.
Boulding, K. (1966): The Economics of the Coming Spaceship Earth. In: Boulding, K. et al.:
Environmental Quality in a Growing Economy: 3-14
Bovenkerk, Michiel (1998), The Use of Material Flow Accounting in Environmental Policy Making in the
Netherlands. In: S. Bringezu, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.): Proceedings of the
ConAccount Conference, 11-12 September 1997, Wuppertal Special 6, Wuppertal: Wuppertal
Institute, 28-37.
Bringezu, Stefan (1997), ‚Accounting for the physical basis of national economies: material flow
indicators‘, in: B. Moldan, S. Billharz, R. Matravers (Eds.) (1997), Sustainability Indicators: A
Report on the Project on Indicators of Sustainable Development. SCOPE 58. Chichester: John Wiley
& Sons Ltd.
Bringezu, Stefan (2000), Industrial ecology and material flow analysis, Proceedings of the international
symposium ‚Industrial ecology and sustainability‘ 22-25 Sept 1999, Troyes, France, Technical
University of Troyes
Bringezu, Stefan, René Kleijn (1997), Short review of the work presented, in: S. Bringezu et al. (1997a),
306-308
Bringezu, Stefan, H. Schütz (2000), Indicator study: Total Material Requirement of the European Union,
European Environment Agency Technical Report, Copenhagen.
Bringezu, Stefan, H. Stiller, F. Schmidt-Bleek (1996), ‚Material Intensity Analysis – a screening Step for
LCA, Concept, method and applications‘, Proceedings of the Second International Conference on
EcoBalance, November 18-20, 1996, Tsukuba, Japan, 147-152.
Bringezu, Stefan, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.) (1997a), Regional and National Material
Flow Accounting: From Practice to Paradigm of Sustainability. Proceedings of the ConAccount
Workshop 21 – 23 January 1997, Leiden, Wuppertal Special 4, Wuppertal: Wuppertal Institute.
OECD 21 2003
Bringezu, Stefan, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.) (1998a), The ConAccount Inventory: A
Reference List for MFA Activities and Institutions, Wuppertal Special 9, Wuppertal: Wuppertal
Institute.
Bringezu, Stefan, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.) (1998b), Analysis for Action: Support
for Policy towards Sustainability by Regional and National Material Flow Accounting. Proceedings
of the ConAccount Conference 11 – 12 September 1997, Wuppertal Special 6, Wuppertal:
Wuppertal Institute.
Bringezu, Stefan, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.) (1998c), The ConAccount Agenda: The
Concerted Action on Material Flow Analysis and its Research & Development Agenda, Wuppertal
Special 8, Wuppertal: Wuppertal Institute.
DETR – Department of Environment, Transport and the Regions (1999), Quality of life counts – indicators
for a strategy for sustainable development for the UK: a baseline assessment, Government Statistical
Service, London.
EEA – European Environment Agency (1999), Environmental indicators: typology and overview,
Technical report No 25, Copenhagen
EEA – European Environment Agency (1999a), Environment in the European Union at the turn of the
century, Copenhagen.
EEA – European Environment Agency (2000), Environmental signals 2000, European Environment
Agency regular indicator report, Copenhagen.
European Union (1999), Presidency summary of the informal meeting of the EU Environment Ministers
and Environment Ministers of the Candidate Countries of Central and Eastern Europe and of Cyprus,
July 23-25, in Helsinki, paragraph 3
Eurostat - Statistical Office of the European Communities (Ed.) (2000), Economy-wide material flow
accounts and balances with derived resource use indicators. A methodological Guide. Luxembourg.
Fischer-Kowalski, M. (1997), Society’s Metabolism. On the Development of Concepts and Methodology
of Material Flow Analysis. A Review of the Literature. Schriftenreihe Soziale Ökologie, Band 46.
Institut für Interdisziplinäre Forschung und Fortbildung der Universitäten Innsbruck, Klagenfurt und
Wien, Wien
Gardener, G., P. Sampat (1998), Mind over matters: recasting the role of materials in our lives. World
Watch Paper 144.
Hansen, Erik (1998), ‚Experiences with SFAs on the national level for hazardous substances in Denmark‘,
in: Bringezu, Stefan, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.): Proceedings of the
ConAccount Conference, 11-12 September, 1997, Wuppertal Special 6, Wuppertal: Wuppertal
Institute, 115-118.
Hoffrén, J. (1999): Measuring the eco-efficiency of the Finnish economy. Research Reports 229, Statistics
Finland, Helsinki.
Japanese Environmental Agency (1992), Quality of the Environment in Japan 1992, Tokyo.
Jimenez-Beltran, Domingo (1998), ‚A Possible Role of Material Flow Analysis within a European
Environmental Reporting System – Changing Course in Environmental Information‘, in: Bringezu,
Stefan, M. Fischer-Kowalski, R. Kleijn, V. Palm (Eds.): Proceedings of the ConAccount
Conference, 11-12 September, 1997, Wuppertal Special 6, Wuppertal: Wuppertal Institute, 16-27
Kleijn, René, S. Bringezu, M. Fischer-Kowalski, V. Palm, (Eds.) (1999), Ecologizing societal metabolism.
Designing scenarios for sustainable materials management. Proceedings of the ConAccount
workshop Nov. 21 1998, Amsterdam. CML report 148, Leiden University.
OECD 22 2003
Matthews, Emily, C. Amann, S. Bringezu, M. Fischer-Kowalski, W. Hüttler, R. Kleijn, Y. Moriguchi, C.

Ottke, E. Rodenburg, D. Rogich, H. Schandl, H. Schütz, E. van der Voet, H. Weisz (2000), The
weight of nations: Material outflows from industrial economies, World Resources Institute Report,
Washington D. C.
Ministry of the Environment (1999), Material flow accounting as a measure of the total consumption of
natural resources, The Finnish Environment 287, Helsinki.
Moldan, Bedrich, S. Billharz, R. Matravers (Eds.) (1997), Sustainability Indicators: A Report on the
Project on Indicators of Sustainable Development. SCOPE 58. Chichester: John Wiley & Sons Ltd.
Nordic Council of Ministers (1999), Factor 4 and 10 in the Nordic Countries. The transport sector - the
forest sector - the building and real estate sector - the food supply chain. TemaNord 1999:528.
Copenhagen.
OECD - Organisation for Economic Co-operation and Development (Ed.) (1996), Meeting of OECD
environment policy committee at ministerial level. Paris, 19-20 February 1996. OECD
Communications Division, Paris.
OECD (1998), Eco-Efficiency, Paris
OECD (1998a), Towards sustainable development: Environmental Indicators, Paris.
Paley Report, The President´s Materials Policy Commission (1952), Resources for Freedom, 5 vols.
Washington D.C., Government Printing Office
Richards, D.J., B.R. Allenby, R.A. Frosch (1994), ‚The Greening of Industrial Ecosystems: Overview and
Perspective.‘, in Allenby, B.R., D.J. Richards (Eds.), The Greening of Industrial Ecosystems, 1-22.
Schmidt-Bleek, Friedrich (1994), Wieviel Umwelt braucht der Mensch? MIPS - Das Maß für ökologisches
Wirtschaften. Berlin, Basel, Boston.
Schmidt-Bleek, Friedrich, S. Bringezu, F. Hinterberger, C. Liedtke, J. Spangenberg, H. Stiller, M.J.
Welfens (1998), MAIA – Einführung in die Material-Intensitäts-Analyse nach dem MIPS-Konzept,
Berlin, Basel, Boston: Birkhäuser Verlag.
Schütz, Helmut, S. Bringezu (1993): Major Material Flows in Germany. Fres. Env. Bull., 2: 443-448.
Steurer, Anton (1992): Stoffstrombilanz Östereich 1988. Schriftenreihe Soziale Ökologie, Band 26. Institut
für Interdisziplinäre Forschung und Fortbildung der Universitäten Innsbruck, Klagenfurt und Wien,
Wien.
Thomas, W.L. Jr. (ed.) (1956): Man’s Role in Changing the Face of the Earth. Chicago
UNGASS - United Nations General Assembly Special Session (1997), Programme for the Further
Implementation of Agenda 21. Adopted by the Special Session of the General Assembly 23-27 June
1997, Paragraph 28, New York.
Verfaillie, Hendrick A., R. Bidwell (2000), Measuring eco-efficiency – a guide to reporting company
performance, World Business Council for Sustainable Development, Geneva.
WBCSD - World Business Council for Sustainable Development (1998), WBCSD Project on Eco-
Efficiency Metrics & Reporting. State-of-Play Report by M. Lehni. Geneva.
Weizsäcker, Ernst. U., A.B. Lovins, L.H. Lovins (1995), Factor Four. Doubling Wealth Halving Resource
Use, London.
Wolman, Abel (1965), The Metabolism of Cities‘. Scientific American, vol.213, no.3: 178-193.
OECD 23 2003
INTERNATIONAL COMPARISON OF INFLOWS AND OUTFLOWS12
The examples below are based on the report “The Weight of Nations: Material Outflows from Industrial Economies”
published in September 20, 2000. They present the results of an international research project whose goal is to
establish a complete set of physical accounts for each participating country, documenting resource inputs and waste
outputs in tonnes and paralleling monetary accounts. “The Weight of Nations” documents waste flows and hidden
flows and is a follow up to the report “Resource Flows: The Material Basis of Industrial Economies”, published in
1997 and focusing on inputs. It builds on an international collaboration between the WRI and other research institutes
in Europe and Japan including: the Wuppertal Institute, Germany; the University of Vienna, Department of Social
Ecology, Austria; the University of Leiden, Centre of Environmental Science, the Netherlands; the National Institute
for Environmental Studies, Japan.
THE MATERIAL FLOWS CYCLE
DMI (Direct Material Input)
Add. Air
= Domestic Extraction -+ Imports
Foreign Water
and Water Vapor
TMR (Total Material Requirement)
Hidden
Flows = DMI + Domestic Hidden Flows + Foreign Hidden
Flows
Imports Exports
DPO (Domestic Processed Output)
= DMI – Net Additions to Stock - Exports
ECONOMIC
TDO (Total Domestic Output)
DMI PROCESSING Domestic = DPO + Domestic Hidden Flows
TMR Processed NAS (Net Additions to Stock)
Domestic Outputs DPO
Extraction = DMI – DPO - Exports
(to Air, Land,
TDO and Water)
STOCKS Absolute size of flows (conventional and hidden)
Use vs. waste
Domestic Domestic Composition of flows (dominance of energy
system, hazardous materials)
DOMESTIC ENVIRONMENT Links to economy and population (concept of
decoupling)
MATERIAL FLOWS TRACKED FOR THE STUDY

Non-Renewable Industrial Minerals Metals Fossil Fuels
Material Flows Asbestos Aluminum Liquid
Bromine Arsenic Natural gas
Clay Cadmium Solid
Fluorspar (fluorine) Chromium Coal combustion products
Gypsum Copper
Nitrogen (ammonia) Gold
Phosphate Iron metal (steel)
Potqash Iron and steel slag
Salt Lead
Chlorine Manganese
Caustic soda Mercury
Sand and Gravel, Industrial Molybdenum
Soda ash Nickel
Sulfur Tin
Zinc
Construction Movements Nonrenewable
Materials of Earth Organic Material
Crushed stone From soil erosion Petroleum
Cement For highway construction Asphalt
Lime For general construction Petrochemicals
Sand and gravel From dredging operations Plastics and resins
Medical chemicals
Synthetic rubber
Petroleum coke
Renewable Agriculture Forestry
Material Flows Animal biomass and manures Natural rubber
Crops biomass Wood products
Human wastes and manures Paper and pulp products
12 Presentation prepared by Ms. Emily Matthews, World Resources Institute (WRI). See also: http://wri.igc.org/materials/ and Matthews, Emily,
C. Amann, S. Bringezu, M. Fischer-Kowalski, W. Hüttler, R. Kleijn, Y. Moriguchi, C. Ottke, E. Rodenburg, D. Rogich, H. Schandl, H.
Schütz, E. van der Voet, H. Weisz (2000), The weight of nations: Material outflows from industrial economies, WRI, Washington D.C.
OECD 25 2003
International Comparison of Material Flows
Total Domestic Output, 1996 Comparing countries by using per capita data shows the following:
100
100
The U.S., the biggest economy, generates the biggest flows with the
DPO mining, construction and agriculture sectors generating large hidden
80
80 Hidden Flows flows.
capita
per capita
Conventional wastes (emissions, discharges, solid wastes, system

60
60 losses). i.e. Domestic Processed Output (DPO) are more directly
tons per
comparable across countries.

Metric tons
40
40 N.B. The size of flows is not a measure of environmental impact. Macro-
Metric
indicators of flows are physical descriptors of an economy, just as GDP

20
20
and other economic indicators are monetary descriptors of the economy.
The size of flows, however, and in particular changes over time (growing
or shrinking) are a good proxy indicator of an economy’s demands on the
00 resource and assimilative capacity of the environment.
Austria
Austria Germany
Germany Japan
Japan Netherlands
Netherlands USA
USA
Net Additions to Stock, and Domestic Processed Output, 1996 The second chart shows the following.
30
30 The proportion of material that is retained in the economy vs. material
that is lost as waste outputs: the ratio of net additions to stock to
25
25
NAS
NAS DPO outflows. In the U.S. the ratio is about 1:4 reflecting a high throughput
20
20 rate, and a relatively low recycling rate
capita
per capita
The actual quantities of material going to stock appear to be higher in

tons per
15
15 Europe on a per capita basis. Interpretation of these figures is
Metric tons
Metric
however not easy; they may reflect differences in building practices

10
10
(bricks and stone in Europe, versus wood frame construction
55 favoured in much of US).
00
Austria
Austria Germany
Germany Japan
Japan Netherlands
Netherlands USA
USA
CO2 Output from Fossil Fuel Combustion and Industrial Processes, CO2 Output as a Percentage of DPO, 1996
Index, 1975-96
USA
USA 1996
1996
1975
1975
Netherlands
Netherlands
Austria
Japan
Japan
Germany
Japan
Netherlands Germany
Germany
USA
Austria
Austria
0
0 20
20 40
40 60
60 80
80 100
100
Percent
Percent
Carbon dioxide from fossil fuel combustion and other industrial processes accounts for over 80% by weight of material outflows from
industrial economies (excluding hidden flows). This proportion has not changed much since 1975. Austria is a partial exception,
because the country has implemented policies to encourage the use of biomass and energy efficiency. Carbon emissions resulting
from biomass combustion are not included in the accounting of outflows in our project, because they are assumed to be carbon
neutral (compensated by planting biomass crops).
OECD 26 2003
MATERIAL FLOWS AND DECOUPLING

A big issue in documenting resource throughput is that of decoupling. Are industrial economies becoming more efficient; do they use
fewer resources and generate less waste to produce the same amount of wealth?
Domestic Processed Output, Index, 1975-96 Domestic
1,400 Processed Output per Unit of GDP, USD, 1975, 1996
1,200
1,200 1975
1975
Austria 1996
1996
1,000
1,000
$million
per $million
Germany
800
800
tonsper
Japan
Metric tons
600
600
Metric
Netherlands
400
400
USA
200
200
00
Austria
Austria Germany
Germany Japan
Japan Netherlands
Netherlands USA
USA
The data reflect an important decline in tonnes of wastes generated per unit of GDP (million constant USD) between 1975 and 1996,
though DPO in absolute terms has increased over the same period.
Decoupling between Material Outputs and GDP, 1975-96 Decoupling between Material Outputs and GDP in the U.S., 1975-96
GDP
GDP has risen 74%;
outflows by about 30%
Austria reflecting some
Germany DPO
decoupling between
Japan
outputs and economic
Netherlands
growth (DPO/GDP).
USA
DPO/GDP
Decoupling between Material Outputs and Population, Decoupling between Material Outputs and Population in the U.S.,
1975-96 1975-96
20
Population grew by
10 about 25%; waste DPO
00 outputs by about 30%. Pop
Austria
No decoupling from
90 Germany
population growth; DPO/Pop
Japan
80 Netherlands
waste generation per
USA capita today is slightly
70
higher than in 1975.
60
50
Detailed Composition of Flows: the U.S. example

Material flow data can be disaggregated in a number of ways as shown in the examples below based on data for the U.S. .
A. Stock Materials and Associated Hidden Flows, 1975-96 B. Composition of Total Domestic Output, 1996
Flows in the U.S., 1975–
1975–1996
1975–1996
Other
Other
7,000
7,000
Agriculture,
Agriculture, Forestry
Forestry and
and Human
Human Waste
Waste
6,000
6,000
Hidden
Hidden Flows
Flows CO2,
CO2, SOx
SOx and
and NOx
NOx
5,000
5,000
Dredged
Dredged materials
materials
tons
metric tons
4,000
4,000 Durable
Durable Goods
Goods
Million metric
Soil
Soil Erosion
Erosion
3,000
Million
3,000 Earth
Earth Moving
Moving
Construction
Construction
2,000
2,000 Materials
Materials Other
Other Mining
Mining Overburden
Overburden
1,000
1,000 Coal
Coal Mining
Mining Overburden
Overburden
00 00 2,000
2,000 4,000
4,000 6,000
6,000 8,000
8,000
1975
1975 1980
1980 1985
1985 1990
1990 1995
1995 Million
Million metric
metric tons
tons
The more detailed composition of U.S. flows shows the importance of hidden flows, and the dominance of energy related flows, coal
mining wastes and CO2 emissions, that together account for about half of the total material outflows in U.S.
OECD 27 2003
A disaggregation of outputs by environmental media (air, land, water) shows that.
Outputs to Air, Land, and water in the U.S., 1975-96 Outputs to air are dominated by fossil fuel combustion and rise with
8,000
8,000 increased consumption.
Uncertain Outputs to land are relatively stable.
6,000
6,000
Outputs to water are poorly documented.
tons
metric tons
Air N.B.: “Uncertain” refers to materials for which fate in environment cannot be
Million metric
determined.
4,000
4,000
Million
2,000
2,000
Water
00
Land
1975
1975 1995
1995
Selected individual flows in the U.S.

Potentially Hazardous Outflows to the U.S. Environment, 1975-96 This chart shows not just substances officially listed as hazardous -
500
500
regulated or listed under the Toxic Release Inventory (TRI) - but all
Other
Other
materials that are potentially harmful in the environment.
400
400 Chlorine
Chlorine
tons
Many of these substances occur not at the processing or

metric tons
Heavy
Heavy Metals
Metals
300 manufacturing stage, but further up or downstream. Many of them
Million metric
300 Asbestos
Asbestos
Salt
Salt are embodied in products.
Million
200
200 Synthetic
Synthetic Organic
Organic Chemicals
Chemicals
Fuel-related
Fuel-related Contaminants
Contaminants NB: the category of fuel-related contaminants contains over 200
100
100 flows, but is dominated by coal fly ash, methane leaks, and oil spills.
0
0
1975
1975 1995
1995
Lead Outputs to the U.S. Environment, Arsenic Use in the U.S., 1975-96 Medical Chemical Outputs to the U.S.
1975-96 Environment, 1975-96
Other 30
30 Other
Other
700
700 Other
Oxides
Oxides and
and Chemicals
Chemicals Glass
Glass Other
Other
Transportation
Transportation Wood
Wood Preservatives
Preservatives 200
200 Vitamins
Vitamins
600
600 25
25 Coal
Coal Fly
Fly Ash
Ash
Gasoline
Gasoline Additives
Additives Gastrointestinal
Gastrointestinal Agents
Agents
Ammunition
Ammunition
Agricultural
Agricultural Chemicals
Chemicals Anti-depressants
Anti-depressants and
and Stimulants
Stimulants
500 Antibiotics
Antibiotics
tons
500
metric tons
tons
metric tons
20
tons
20
metric tons
150
150
Thousand metric
Thousand metric
Thousand metric
400
400
15
15
Thousand
Thousand
Thousand
100
100
300
300
10
10
200
200
50
50
55
100
100
00 00 00
1975 1995 1975 1995 1975
1975 1995
1995
1975 1995 1975 1995
This chart shows the influence of Arsenic use in US: largely regulated out of Medical chemicals are a “watch this
regulation on airborne lead emissions agricultural pesticides but increasingly space” set of flows. US consumption is
from gasoline use that were largely used in pressure treated wood products. increasing. Many medical chemicals that
eliminated. Overall outflows have not Currently this material held in stock but at are highly biologically active end up in the
declined. A major new application in the end of useful life (15-30 years) wood is soil and in water.
“other” category is electrical and chipped, or burned, or landfilled. Arsenic
electronic goods, notably computer will be going up in the air, or leaching out
monitors. into soil and groundwater.
N.B.: The 1994 spike in transportation use
(lead acid batteries) is a data anomaly.
OECD 28 2003
MATERIAL FLOW ACCOUNTING:

The experience of Canada13
1. Introduction
The Canadian initiative to develop environmental and resource accounts had its origins with
exploratory work carried out during the late 1970s and early 1980s (Friend and Rapport, 1979; Friend,
1981). At that time, data representing the interactions between human activity and the environment were
recorded using the Stress-Response Environmental Statistical System. This framework had as its focus the
physical measurement of the environment's response to various human stresses. Although useful as a
means of organising physical data, the framework did not attempt to incorporate monetary data or to
provide links to the economic variables most often employed in policy development; namely, the variables
of the Canadian System of National Accounts (CSNA).14
Building on this early work, the Canadian System of Environmental and Resource Accounts
(CSERA) has been developed with the specific objective of organising physical and monetary statistics
related to natural resources and the environment using classifications, concepts and methods that are
compatible with those of the CSNA. Thus, the statistics of the CERA can in large part be directly
integrated with those of the CSNA. The integration of these two data sets—one environmental, the other
economic—represents a significant milestone in assessing economic activity and its dependence upon the
natural environment. This capacity is particularly relevant today, given the increasing focus of
governments, businesses and individual Canadians on the objectives of sustainable development.
The main objective of this paper is to present one component of the CSERA: the Material and
Energy Flow Accounts. These accounts record the flows of materials and energy—in the form of natural
resources and wastes—between the economy and the environment. The two other components of the
System, the Natural Resource Stock Accounts and the Environmental Protection Expenditure Accounts,
are described in detail in Statistics Canada (1997).
2. Material and Energy Flow Accounts in Canada

The Material and Energy Flow Accounts (MEFA) record in substantial detail the annual flows of
materials and energy—in the form of resources and wastes—between the Canadian economy and the
environment. The accounts record the quantities of natural resources produced (that is, harvested or
extracted) by industries, households and governments, and show how these resources are consumed by
these same agents. Likewise for wastes, the accounts show the quantities produced by each agent and how
these wastes are “consumed,” either as recycled materials or as flows into waste disposal sites or to the
environment. The MEFA share their classifications of industries, households and governments with
Statistics Canada’s Input-Output Accounts (Statistics Canada, 1999). This allows the environmental data in
the MEFA to be linked directly and easily with the economic data found in the Input-Output Accounts,
adding value to both data sets.
The true strength of the MEFA is that detailed data on resource and waste flows are directly
linked with the rich body of economic statistics available. The analytical power that this linkage offers,
13. Paper prepared by Mr. Martin Lemire, Statistics Canada, and based on previous work done by Statistics Canada (1997). Consequently,
opinions expressed here reflect those of Statistics Canada.
14. The CSNA, which has a history of more than 40 years, is the source of a number of Statistics Canada's most important indicators of economic
activity, including Gross Domestic Product.
OECD 29 2003
within the well-established and widely-used framework of the CSNA, contributes substantially to our
ability to study the Canadian economy and the demands it places on the environment. Such understanding
has an important role to play in the informed management of the economy toward the simultaneous
realisation of our economic and environmental goals. This linkage allows the calculation of important
indicators of the resource and waste intensiveness of economic activity.
The MEFA are compiled using the sectoral classifications of the Input-Output Accounts. Likewise, the
geographical scope of the MEFA is national and the frequency with which they are compiled is annual,
matching the scope and frequency of Input-Output Accounts.
2.1 Why account for materials and energy?

The rationale for developing the MEFA rests on arguments that economic use of the environment
has exceeded (or is approaching) critical thresholds. Although human degradation of the environment is
not new, there are both qualitative and quantitative differences between the environmental impacts of
economic activity in the past and those of today. A major difference is that of scale. While the
environmental degradation that occurred in ancient times was mainly localised and attributable to a few
activities, today it is wide spread and associated with myriad activities. Another important distinction is the
fact that the environment has essentially no assimilative capacity for many of the waste materials released
from modern economic activities. For example, the family of chemicals known as the halocarbons best
known for their role in depletion of the ozone layer have atmospheric lifetimes that range from a few
thousand to tens of thousands of years.
Consequently, it is clear that economic activity has long since passed the point where the
environment can be taken for granted as a source of resources and a dump for wastes. Both local and global
environmental capacities to absorb wastes are being pressed upon today in unprecedented ways. Likewise,
the environment’s capacity to supply the resources needed to meet the economy’s growing material
demands is increasingly being exceeded. The measurement of the quantity of material and energy flows in
the Canadian economy and the “intensity” with which we use the environment has thus become essential.15
It is perhaps easier to make a case for the importance of the MEFA in terms of what they reveal
about our waste flows than what they reveal about the production and consumption of resources. Most
people are familiar with the impacts of waste from first hand experience of air, water and land pollution;
fewer directly feel the environmental impacts of excessive resource use. A reasonable person might
therefore ask, “Is there as compelling a reason to account for resources as there is for wastes?” Two
arguments suggest there is.
First, although Canada is not in immediate danger of running out of most of its natural resources,
there are certain instances where our resource stocks have been depleted almost, or completely, to
extinction. The disappearance of commercially viable stocks of northern cod off the East Coast is perhaps
the best recent example of this. One could cite as well the near disappearance of old-growth forests across
the country and the slow, but continual loss of prime agricultural land to urban development. In such cases,
it is important to measure both how much of the remaining resources we are consuming and how they are
being consumed. Even for resources that are abundant in supply, it is sensible to monitor use, particularly
for non-renewable resources that will, by definition, run out one day. The MEFA are designed to provide
measures of resource use that can facilitate monitoring of this sort.
The second argument in favour of measuring resource flows is related to the concerns already
raised with respect to waste production. This is quite simply that the quantity of waste produced by
economic activity is directly related to the quantity of raw materials and energy consumed in the first place.
The basic law of conservation of mass and energy demands that all material and energy entering the econ-
15. In this context, intensity is taken to be the degree to which the environment is used as a source of raw materials or a sink for wastes per unit of
economic output.
OECD 30 2003
omy must leave it again at some point (or be permanently stored within it). Thus, any problem of excessive
waste output is in the first place one of excessive material and energy input.
3. The MEFA accounting framework

Flows of produced goods and services are well articulated in monetary terms in the existing
national accounts. The Input-Output Accounts provide annual estimates of the production and consumption
of 476 commodities by 167 industries and 122 categories of final demand (Statistics Canada, 1999). The
Material and Energy Flow Accounts build on this substantial detail by incorporating physical estimates of
natural resource and waste flows into the accounting framework of Statistics Canada’s Input-Output
Accounts. In modifying the framework to make it suitable for the MEFA,16 several objectives were sought.
− The framework should be structured in such a way that it is suitable for recording all material
and energy flows related to economic activity, regardless of the nature of this relationship.17
− The framework should represent both the production and consumption of materials and
energy. In doing so, the traditional accounting identity between production and consumption
should be respected.
− All sectors of the economy should be covered in the framework. The definitions of these
sectors should conform with those of the CSNA.
− The framework should facilitate the integration of material and energy flow data with the
economic statistics in the Input-Output Accounts.
3.1 Benefits of the input-output framework
The benefits of using input-output accounting frameworks to analyse material and energy flows
have long been recognised. Corresponding with the environmental movement of the late 1960s, several
economists suggested the use of input-output techniques for environment-economy analysis (Cumberland,
1966; Daly, 1968; Isard, 1969; Ayres and Kneese, 1969; Leontief, 1970; and Victor, 1972). The work of
Victor in particular has influenced the approach to material and energy flow accounting taken in the
MEFA. His work represents the most comprehensive of the original frameworks and, importantly, he
employed Statistics Canada’s input-output accounting framework in his study.
Despite the relatively long-standing interest in the use of input-output techniques for
environment-economy analysis, there was not a great deal of empirical development in the field during the
1970s and 1980s. Indeed, one of the authors cited above noted for his contributions to the field of material
and energy analysis over the last three decades found cause to note that the “approach deserves much
greater attention that it has received to date” (Ayres, 1996). There are several reasons why the approach is
worthy of attention.
To begin with, the Input-Output Accounts are very detailed. At their most elaborate, they present
production and consumption statistics for 243 industry groups, 679 commodity groups and 162 categories
of final demand, in both current and constant dollar measures.18 This detail is of great benefit when
analysing material and energy flows. It allows one to move beyond highly aggregated measures, such as
total energy use per unit of GDP, that are sometimes proposed as environmental indicators. Changes from
year to year in such “economy-wide” indicators can be difficult to interpret because so many factors are at
play. For example, a decrease in energy use per unit of GDP might result from a real increase in the energy
efficiency of production processes, or it might result from a reduction in the contribution of energy
16. Full details of the modify framework are available in Chapter 4 of Statistics Canada, 1997, Concepts, Sources and Methods of the Canadian
System of Environmental and Resource Accounts, Catalogue No. 16-505-GPE, Ottawa.
17. Economic activity in this context includes all human activities associated directly or indirectly with the production or consumption of
commodities.
18. The accounts are actually produced at three different levels of aggregation. Many data at the most detailed level are not publicly available, as
the legal requirement to protect respondents’ confidentiality prevents their release. Thus, the accounts are released to the public at the so-called
“medium” level of aggregation.
OECD 31 2003
intensive industries to the total output of the economy. Determining which factor is the actual cause is
difficult. In contrast, when energy use is measured for individual industries, the effect of the changing
structure of the economy is largely eliminated. A decrease in energy use per unit of GDP for the electric
power industry, for example, can be interpreted unambiguously as an increase in the energy efficiency of
electric power production processes.
Another reason for choosing the Input-Output Accounts as the basis for the MEFA is that they
are flow accounts. This means that many of the concepts already well defined in the Input-Output
Accounts are easily transferred to the MEFA (which, as their name implies, are also flow accounts).
Production in the Input-Output Accounts, for example, is defined as the fabrication of goods and services
for sale in the open market. It is not a significant leap to adapt this concept for use in the MEFA; three
extensions suffice:
− the measurement of production must be allowed in physical units as well as monetary units;
− non-traded resource flows for self-consumption purposes must be treated as production; and
− the production of unwanted “bads” (wastes) along with the production of “goods” must be
recognised.
The fact that many of the concepts of the Input-Output Accounts can be transferred to the MEFA
enhances the degree to which the data from the two sets of accounts can be integrated. This, in turn,
enhances the analytical power of the MEFA.
Another important benefit of linking to the Input-Output Accounts is the fact that these accounts
have been produced annually by Statistics Canada for over 30 years. Moreover, since they form a
cornerstone of the CSNA, a great deal of effort is put into ensuring that they are as reliable, consistent and
comparable as possible. Thus, the Input-Output Accounts represent a rich data set to draw on for the
development of historical material and energy flow accounts. Equally importantly, Statistics Canada is
committed to continued production of the Input-Output Accounts for the foreseeable future. Indeed, current
plans call for the expansion of their scope to include annual provincial input-output accounts in addition to
the current national versions.
Despite these strengths, one limitation of the Input-Output Accounts should be mentioned. This is
the fact that the accounts are only released four years after the reference year. This delay is explained by
two factors. One is simply their complexity—the enormous detail they present demands a long time to
amass. The other is the aforementioned fact that the Input-Output Accounts are used as Statistics Canada’s
benchmark estimate of GDP. Because they represent the “final word” on GDP, the Input-Output Accounts
cannot be completed until the most detailed source data are available in final form. In the case of some
sources (income tax files for example), this can take several years. The delay in releasing the Input-Output
Accounts means that the MEFA cannot be used for analysis of material and energy flows within the last
four years. Of course, it is possible to compile material and energy flow data for more recent years, just not
to combine these data with input-output data.
4. Uses of the Material and Energy Flow Accounts

Considerable use has already been made of MEFA data in university research projects. Other
researchers accustomed to using Statistics Canada’s economic data and concepts also find the MEFA
useful, as their classifications allow easy incorporation of material and energy flow data into existing
models built around these concepts.
Journalists as well may find the MEFA of interest, particularly for the new context that they
provide for reporting on the economy. In addition to the economic indicators that journalists are
accustomed to using from Statistics Canada (GDP, unemployment rates, inflation rates and so on), the
MEFA provide important environment-economy indicators that shed additional light on the nature of
economic development in Canada. These are quantitative measures that define the extent to which the
OECD 32 2003
economy places demands on the environment as a source of raw materials and as a sink for waste
materials.
As for policy use of material and energy flow indicators, it is still not clear how these indicators
are used in Canada for setting national policy. At the micro-level, the work has focused on eco-efficiency
and material intensities in enterprises.
5. Material and Energy Flow Indicators

The indicators that could be developed from the MEFA include:
− resource intensity of industrial output
− resource intensity of household consumption
− resource intensity of net exports
− waste intensity of industrial output
− waste intensity of household consumption
− waste intensity of net exports
− renewable energy as a proportion of total energy production
− recycled proportion of total resource use (in progress)
The rationale underlying all of these indicators is, as argued above, that current material and
energy flows approach or exceed those that are environmentally sustainable in the long-term. The
indicators have been selected as key variables to monitor in this regard as the economy develops over time.
Each one considers an important aspect of the economy’s use of the environment as a source of material or
as a sink for wastes. By uniting the environmental data in the MEFA with the economic data in the Input-
Output Accounts, these indicators tell us how our economy is developing with respect to its demands on
the environment. While they cannot themselves answer the question “What is a sustainable level of
material and energy flows?”, they can demonstrate broadly whether the economy is heading toward or
away from environmental sustainability. Other things equal, if fewer demands are placed on the
environment (in terms of resource and waste flows) per unit of economic output over time, the
development path is leading toward sustainability. As more scientific information regarding the
environment’s capacities to provide resources and absorb wastes becomes available, these indicators will
allow more concrete statements about the absolute environmental sustainability of the economic activity to
be made.
A core set of material and energy flow indicators based on the MEFA is published annually along
with other economy-environment indicators developed by Statistics Canada. These environment-economy
indicators provide important counterparts to the long-standing economic indicators published by Statistics
Canada. Their creation makes it possible now to consider the development of the economy vis à vis our
economic goals while at the same time considering the accompanying growth (or decline) in resource
consumption and waste production. The environment-economy indicators allow this by answering
questions of the following sort:
− what raw materials and energy are consumed by the economy, in what quantities and by
whom;
− what is the “intensity” of our resource use; that is, how much raw material and energy is
required to produce one unit of economic output;
− what waste products are emitted from the economy, in what quantities and by whom;
− what is the “intensity” of our use of the environment for waste absorption; that is, how much
waste is released per unit of economic output;
OECD 33 2003
− is resource use/waste output increasing or decreasing over time, both in absolute terms and
per unit of output;
− what wastes are recycled, in what quantities and by whom; how much raw material and
energy input is avoided by the use of recycled wastes?
6. Data gaps
Despite the substantial progress made in compiling material and energy flow data for the MEFA,
significant gaps remain in the coverage of the accounts, particularly for waste flows. The following areas in
which existing data are weak or there is a complete lack of data can be noted.
− Reliable, detailed data on the quantities of non-toxic solid waste (e.g. household solid waste)
generated and recycled in Canada do not exist. Those data that do exist are not highly detailed
with respect to material composition. More detrimental from the perspective of the MEFA,
the data do not classify solid waste flows according to producer. Instead, they generally
report just a single measure dubbed “municipal solid waste”, which includes all solid waste
collected by, or on behalf of, local municipalities and disposed of in local landfill sites. This
aggregate comprises part or all of the solid waste generated from households, light industrial
and commercial establishments, office buildings, public institutions and government
operations. In the absence of information with which to disaggregate these flows according to
producer, aggregate solid waste data are not suitable for use in the MEFA.
− Waterborne waste data are weak for all sectors of the economy, particularly for sewage flows.
Although some data exist on the quantities and composition of sewage treated in municipal
sewage treatment plants, these data suffer from the same problem as the solid waste data
described above. That is, the data do not detail the composition of the sewage, and they do
not distinguish sewage flows according to producer.
− Data on the production of durable-good wastes are almost entirely lacking. Although the
possibility exists to model these waste flows using historical data on purchases of durable
goods, this approach is currently unproven.
− Data on leakages from waste inventories are inadequate. Only some rather crude estimates
have been made for methane emissions from landfill sites. Estimates of other gaseous
emissions and liquid leachates from landfill sites are unavailable.
− Data on public sector waste production are almost entirely lacking. This includes data
describing the wastes generated in government office buildings, those associated with the
operation of public institutions (hospitals, prisons, military bases, schools) and with the
provision of government services (road building and maintenance, for example).
− A general shortcoming in the waste data that are available, even those that are detailed and
reliable, is that they do not exist as long time-series. It is rare to find waste data that extend
further back in time than the mid 1980s; most begin in the 1990s.
− With respect to resource flows, the major shortcoming is the lack of data representing the
flows of recycled materials. Although some data are available on the quantities of municipal
solid wastes collected for recycling, again these data do not classify the wastes by producer.
Data representing the quantities of waste materials collected for recycling outside of
municipal recycling programs are not readily available at all.
7. Future directions for Material and Energy Flow Accounts
Although the conceptual framework of the MEFA is well-established, their current state of
empirical development represents only a small fraction of the material and energy flows that would be
ideally covered. With respect to waste flows, estimates of greenhouse gas emissions are the sole statistics
OECD 34 2003
to have been compiled to date (see section 6). With respect to natural resource flows, current plans call for
the development of estimates for timber, metallic and non- metallic minerals and possibly land. Statistics
Canada collects many of the raw data required to incorporate these flows into the MEFA framework.
Integrating these data is a high priority for the future development of the MEFA. Estimating the quantities
of recycled wastes that are used in place of virgin resources is also a high priority.
REFERENCES
Ayres, R.U., 1996, "Statistical Measures of Uncertainty," Ecological Economics, Vol. 16, No. 3, p. 239-
255.
Ayres, R.U., and A.V. Kneese, 1969, "Production Consumption and Externalities," American Economic
Review, Vol. 59 (June), p. 282-97.
Cumberland, J.H., 1966, "A Regional Inter-Industry Model for Analysis of Development Objectives,"
Regional Science Association Papers, Vol. 17, p. 65-95.
Daly, H.E., 1968, "On Economics as a Life Science," The Journal of Political Economy, Vol. 76 (May-
June), p. 393-406.
Friend, A., 1981, "Conceptual Frameworks and a Unified Approach to Environmental Statistics,"
Canadian Statistical Review, Statistics Canada, Catalogue No. 11-003E, Ottawa.
Friend, A. and D. Rapport, 1979, Towards a Comprehensive Framework for Environmental Statistics: A
Stress-Response Approach, Statistics Canada, Catalogue No. 11-510, Ottawa.
Houghton, J.T., L.G. Meira Filho, B.A. Callander, N. Harris, A. Kattenberg and K. Maskel (eds.), 1996,
Climate Change 1995: The Science of Climate Change, Cambridge University Press, Cambridge.
Isard, W., 1969, "Some Notes on the Linkage of the Ecologic and Economic Systems," paper delivered to
the Regional Science and Landscape Analysis Project, Harvard University, June 27.
Leontief, W., 1970, "Environmental Repercussions and the Economic Structure: An Input-Output
Approach," Review of Economics and Statistics, Vol. 52 (August), p. 262-71.
Statistics Canada, 1997, Concepts, Sources and Methods of the Canadian System of Environmental and
Resource Accounts, Catalogue No. 16-505-GPE, Ottawa.
Statistics Canada 1999, The Input-Output Structure of the Canadian Economy, 1993-1995, Catalogue No.
15-201-XPB, Ottawa.
Victor, P.A., 1972, Pollution: Economy and Environment, University of Toronto Press, Toronto.
OECD 35 2003
ANALYSIS OF MATERIAL FLOWS FOR SUSTAINABILITY POLICY:

The experience of Italy19
1. Introduction
In the course of time, and especially in the last years, the scientific debate concerning the
environmental effects of material flows has been more and more influenced by authors who shifted
attention from direct and well-known effects to systemic interaction and uncertainty and from
environmental emergencies to precaution and long-term sustainability. This has brought about a shift in
emphasis from the emissions of specific substances to the whole of material flows induced by human
activities. While in the traditional approach an ex-post aggregation of a myriad of indicators of
environmental pressures is needed in order to come to a “policy-friendly” synthetic index of total pressure,
the “modern” material flows approach tries to tackle the problem directly at the aggregate level. The claim
at its basis is that it is not possible, both for practical and for theoretical reasons, to measure and put under
control the overall environmental disruption potential of human activities by using hundreds of indicators,
each of which is very precise in its meaning, but also very limited in scope. Moreover, ex-post aggregation
does not grant unambiguous results as far as the meaning of the index is concerned (weights have to be
chosen for the different pressures). In order to overcome the impasse, paradoxically given by “too much
detailed information”, a more synthetic catch-all – though necessarily fuzzy in its meaning – accounting, in
terms of tons, of all the material flows induced by man has been proposed, from which indicators suited to
monitor systemic progress towards ecological sustainability can be derived.
This enlargement of the perspective to the overall dimension of human activities with respect to
nature, has been accompanied by the growth of the activities of production and elaboration of data referred
to the whole of the material flows caused by human activities, and especially to the inputs to the economy
(“what comes in will eventually come out”: therefore, a reduction of inputs is needed in the first place).
The patterns of materials’ use through economic growth revealed by the evidence collected so far show no
strong tendency to “dematerialisation” of the economies, even the most advanced ones, or show such a
tendency only in relative terms (i.e. in terms of intensity of use of materials per GDP unit)20.
We will not discuss here the issue of the significance of these aggregate measures of material
flows, which will be assumed as indeed very meaningful when long-term ecological sustainability is
concerned, but try to make a step forward from sheer accounting into analysis, showing how the
determinants of material flows in terms of prevailing technology and structure of the economy can be
identified and measured, and highlighting the way they interact in determining the observed material flow
aggregates. These determinants can be seen as the intermediate objectives of an ecological-economic
policy. The trends recalled above, indeed, highlight the inadequacy of the “efficiency” strategies followed
so far, based mainly on technology improvements, and the necessity of a “sufficiency” policy, based on
improvements of societal organisation and profound structural change of the economies and lifestyles.
A policy that does not want to be confined to the ecological dimension but aims at taking into
consideration several aspects of sustainability at once, needs to be based not just on environmental
indicators but also on indicators which are specific of the other dimensions considered, and on an
equivalent analysis of their determinants. The main problem, in this analytical setting, is then how to link
19 Paper prepared by Mr. Aldo Femia, ISTAT, Italy.
20. See Femia-Hinteberger-Renn, “Economic Growth with less Material Input? De-Composing the Empirical Trend of De-Linking”, paper
presented at the World Congress of Environmental and Resource Economists, Venice, June 25-27, 1998.
OECD 37 2003
indicators concerning the social, economic and ecological dimensions, in a way that the trade-offs between
the respective objectives are correctly highlighted. In this paper we attempt to show how this can be done,
drawing from an application to the Italian case of Leontievian input-output analysis of material inputs and
of value added. The application is limited in many respects, but its main aim is to show how material flow
accounting, fed into a research program based on the application of input-output logic, can contribute to
the formulation of a sustainability policy attentive to the (socio-)economic dimension of development as
well as to the ecological one.
2. Data and method of analysis
The data used in this application are referred to the material content of the direct current material
inputs to the Italian productive activities carried out in 1988. Due to the limitations of the data, which do
not allow very meaningful further aggregations, we could not derive the “total material requirement”
indicator, the one that mostly characterises the approach, but had to keep 8 different categories of
materials, to be separately analysed21.
In order to perform the analysis described below, the data have been organised in a very simple
NAMEA-type environmental accounting scheme22. In particular the matrix of direct material inputs
reported in Table 1 has been linked to the monetary input-output table of the Italian economy of 1988. The
material input matrix collects the data on the direct intake of the 8 types of materials by the 44 NACE
branches and the 5 final demand categories of the input-output table. From this scheme, direct input
coefficients are derived, expressing – branch by branch – intermediate and primary inputs per unit of total
product.
The application of the input-output technique is based on recursive use of these unitary
coefficients in order to reconstruct (backwards) the whole production chain that “stays behind” the final
production of any given branch. This “disentangling” of the intermediate exchanges among industries, can
be seen as a virtual separation of the production chains; the activities are collected in fictitious branches,
which we will call “vertically integrated” after Pasinetti, according to the final purpose for which
production is carried out. These fictitious branches, therefore, produce only for final demand directly from
primary inputs. The application of this technique thus allows to calculate, for any given branch of the
economy the total (i.e. direct plus indirect) inputs needed to satisfy the final demand for the products of the
branch, or – in other words – the total inputs “activated” by this final demand. This applies to material
inputs as well as to any other input of the economy (employment, services of capital…) and also to the
generation of value added, which is what we consider in this application in order to represent the economic
dimension of sustainability. For any given primary resource, the sum of its total input in all the vertically
integrated branches gives the quantity of that input used in the whole economy.
The absolute dimension of the use of inputs by vertically integrated branch is by itself not
sufficient to design a policy framework, because it is heavily dependent on the sheer dimension of the
branch itself, i.e. on statistical aggregation. It is therefore necessary to adopt some standardisation method,
as for instance the division of the quantities of the inputs of the vertically integrated branches by the values
of the respective deliveries to final demand. The measures derived this way express the intensities of use of
the inputs, which can be considered inverse indicators of the efficiency with which material resources are
transformed into products suited to satisfy final demand. Summarising, the total quantity of a given input
21. No attempt has been made to reconstruct the flows of non-used materials accompanying those of used inputs, nor to assess the indirect material
inputs of imported goods and services. As a consequence, domestic and imported direct material inputs cannot be summed together due to the
different nature of the materials to which the data refer: imported products are not a direct intake from nature, and represent much bigger flows
of materials than the quantities they actually embody. The materials contained in some of the imported goods, moreover, could not be
identified, and have therefore been collected in the residual category of “undetermined” materials.
22. The NAMEA (National Accounting Matrix with Environmental Accounts) is an extension of the matrix presentation of national accounts
(NAM), in which indicators of environmental pressures are reported in additional modules using the same aggregation keys and classifications
as those used in the national accounts.
OECD 38 2003
used in the economy will be expressed as the scalar product of two vectors, one expressing the intensities
of use of that input in the vertically integrated branches, the other one the level and composition of final
demand23.
The first vector is actually a vector of “vertically integrated input coefficients”, and can therefore
be seen as the one describing the role of technology in the determination of the use of inputs and put in
relation to the “efficiency” side of the problem of reducing material inputs to the economy. The latter
vector can be put in relation to the issue of “sufficiency” mentioned in the introduction, and is important
for its level (ceteris paribus, less consumption leads to less material inputs) as well as for its composition.
Concerning composition, it can be observed that in many cases a given need can be satisfied with different
mixes of products (think for instance of different transportation means, or also of communication as a
substitute of mobility), which have greatly different implications in terms of activated material flows, due
to the different efficiency in the use of resources of the different branches24.
Moreover, for any given vertically integrated branch, it is possible to compare the results on the
use of different inputs (in our case, materials and value added). In this way we can obtain measures such
as, for instance, the “use of biotic inputs per unit of value added” by vertically integrated branch, which
expresses the trade-off between the environmental and the economic objectives, useful to answer questions
such as for instance: “how much value added must be given up in order to reduce biotic inputs of the given
vertically integrated branch of a predetermined quantity, assuming that technology in that vertically
integrated branch will become more efficient in the use of biotic resources by x% and all other things being
equal?”. Since this indicator is available for all branches, it can help identifying the mix of incentives and
disincentives that achieves the desired reduction of use of biotic input with the minimum possible loss of
income generated.
This kind of analysis can be extended – if data are available for a period of time – in order to
identify the causes of the past observed changes in the total intake of an input, to assess to which extent has
technology progressed and/or structural change influenced materials’ use, and to project in the future the
effects of past trends of technology improvement and structural change; moreover, simulation exercises
can be performed, hypothesising evolution patterns for intensities of use and final demand, helping to
figure the effects of alternative sustainability policy.
3. Main results
3.1 Absolute material use by vertically integrated branch
The figures in the annex report the total quantities of materials used by the vertically integrated
branches.
Due to the high aggregation of the 44-branches input-output table and to the kind of materials
available in the Italian territory, which is poor of metals and energy carriers, an important share of the use
of materials of domestic origin is concentrated in the same branches in which their removal from nature
occurs. As a matter of fact, often the activities that further transform th e materials belong to the same
23. In the presence of economies of scale, the intensities of use cannot be considered fully independent from the absolute dimension of demand for
the products of the branch. This is relevant when simulations are performed, but does not subtract anything from the descriptive power of the
model. The intensities of use, moreover, can be split into an “extraction technology” component and another one referring to interindustry
exchanges (the former is the vector of direct input coefficients, the latter the Leontiev inverse).
24. The way in which this vector depends (in turn) from income generation, the constraints in its composition given by complementarity and
substitutability and therefore the room for policy to change it in order to influence the use of primary inputs are all parts of the research
program mentioned above which we cannot discuss here, but should be kept in mind for their crucial importance in the derivation of policy
prescriptions.
OECD 39 2003
Table 1. Direct material inputs of productive branches and final demand categories
Italy, 1988 (tons)
Materials of domestic origin Imported materials

I/O (NACE) Biomass Non-energy Energy Endogenous steam Biomass Non-energy Energy minerals Undetermined
branches (excl. fodder) minerals minerals minerals
01 77.462.041 0 0 0 2.249.039 94.758 104.299 386.484
03 0 0 1.648.771 0 0 683 67 0
05 0 0 0 0 0 5.628 960.319 0
07 0 56.907 17.716.085 0 24.456 18.692 24.147.766 60.781
09 0 0 0 31.458.439 4.011 42.269 3.534.071 10.443
11 0 0 0 0 0 0 27 0
13 0 172.652 0 0 611.698 13.521.619 464.310 63.401
15 0 343.928.030 45.638 0 102.523 10.339.253 711.748 118.732
17 0 0 0 0 1.524.356 3.238.120 2.271.418 3.093.391
19 0 0 0 0 38.836 5.299.193 238.489 89.291
21 0 0 0 0 6.081 3.530.436 243.398 17.578
23 0 0 0 0 1.371 713.244 13.265 21.629
25 0 0 0 0 54.661 2.605.536 77.165 77.244
27 0 0 0 0 27.094 1.984.938 536.930 108.323
29 0 0 0 0 5.031 758.564 67.675 5.721
31 0 0 0 0 3.179.016 15.107 19.803 1.792
33 0 0 0 0 168.758 5.518 21.679 434
35 0 0 0 0 6.819.675 1.018.944 303.589 41.539
37 0 0 0 0 243.555 606.729 29.851 13.824
39 0 0 0 0 349.684 1.778 1.229 44
41 0 0 0 0 2.965.712 127.072 163.112 675.471
43 0 0 0 0 1.228.649 79.029 34.419 74.390
45 0 0 0 0 4.394.972 273.519 157.797 39.550
47 0 0 0 0 3.326.607 86.153 126.241 121.483
49 0 0 0 0 811.438 64.393 1.155.186 1.193.433
51 0 0 0 0 59.633 2.831.253 44.778 39.024
53 0 0 0 0 412.934 11.719.725 272.101 160.326
55 0 0 0 0 14.080 696.867 32.981 32.173
57 0 0 0 0 107.103 55.809 577.368 3.702
59 0 0 0 0 1.259.633 345.062 351.588 301
61 0 0 0 0 14.040 75.405 394.311 4.257
63 0 0 0 0 132.142 284.583 1.132.691 11.814
65 0 0 0 0 3.922 20.094 39.080 58
67 0 0 0 0 546 11.057 14.535 313
69 0 0 0 0 0 13.153 15.790 178
71 0 0 0 0 40.082 33.045 86.466 16.296
73 0 0 0 0 0 20 0 27
75 0 0 0 0 2.552 1.667 25.082 1.653
77 0 0 0 0 7.253 7.142 48.029 9.241
79 0 0 0 0 45.026 108.614 57.882 75.592
81 0 0 0 0 225.964 229.686 309.435 22.426
85 0 0 0 0 2.123 65.996 53.416 2.116
89 0 0 0 0 29.882 155.039 143.468 37.271
93 0 0 0 0 4.172 0 10.392 12
Households 0 0 0 0 9.212.574 7.108.788 8.110.710 824.662
P.A. Cons. 0 0 0 0 0 0 0 0
Investments 0 0 0 0 621.709 2.348.713 5.646 19.806
Inventories 0 0 0 0 12.663 -10.043 -287.826 22.930
Export. 0 0 0 0 669 0 136.288 0
Total 77.462.041 344.157.589 19.410.494 31.458.439 40.345.922 70.532.851 46.958.063 7.499.156
Source: Author's processing of ISTAT and Foreign Trade Statistics data.
OECD 40 2003
branches to which the activities that take them from nature agriculture belong: and forestry for biomass
(branch 01); oil, refineries and gas for energy carriers (branch 07, which comprises both the extraction of
these materials and their refinement); electricity generation for endogenous steam (branch 09 – in this case
the activities of extraction and transformation coincide); and cement, glass, ceramics for non-energy
minerals (branch 15, which also comprises their extraction). Exceptions to this rule are the activation of
extraction of non-energy minerals by the final demand of the products of the construction branch (53),
which is the and the use of biomass in the vertically integrated branches of food industries (31-35) and
catering (59).
The materials embodied in imported goods show a more complex pattern:
− undetermined materials are mainly demanded for the production of final chemical products
(17), textiles (41), rubber (49) and for the construction and maintenance of buildings (53);
− energy materials are imported for the delivery to final users of refined petroleum products
(07), of chemical products (17), buildings (53), and of some services like commerce (57),
hotels, restaurants and bars (59), transport (61 and 63) and Public Administration (81);
− non-energy minerals from abroad are demanded for the purpose of construction (53) as well
as for the final sales of the industries of metal and non-metal minerals (13 and 15), of all
kinds of metal products, machinery and vehicles (19-29), and of chemicals (17); among
services, those provided by the P.A. (81) require substantial inputs of these materials;
− imports of biomass are activated by the final demand of agricultural products (01), of some
products of food industries (meat and cereals mainly, branches 31 and 35), of chemicals and
of some typical “made in Italy” products such as textiles, clothes, furniture… branches 41-
47) and also by services of hotels, restaurants and bars (59).
3.2 Intensity of use of materials by vertically integrated branch
As observed before, the absolute quantities of materials used by the vertically integrated branches
are per se not very significant: to a certain extent they simply reflect the importance of the branches in final
demand. Therefore it is useful to consider coefficients expressing the ratios between the quantities of
materials used and value of the products delivered to final demand by branch (intensities of use). Figure 1
gives the “vertically integrated material input coefficients”, i.e. the average “intensities of use”, of the eight
types of materials, in a highly aggregated form (each of the sectors reported in the figure groups together
several NACE branches). It is important to bear in mind that the ij-th bar represents the quantity of material
i directly and indirectly (i.e. through other sectors) used (activated) by sector j in order to deliver one unit
of its product to final demand (expressed in the figure in tonnes per LIT million of product).
For what concerns materials of domestic origin, it is also interesting to consider more specifically
the indirect activation of materials (indirect intensity of use):
− in general, this is relatively high in the same branches that directly take the materials from the
environment: endogenous steam in the production of electricity (branch 09), energy minerals
in that of carbon (03) and oil (07), non-energy minerals in the processing of metals (13) and
in production of cement, glass and ceramics (15), biomass in agricultural production (01);
this shows that a significant part of the use of these materials is activated by the final demand
for products of the same branch that removes them from the national environment, rather than
by the intermediate demand of the other branches; again, this can be partly attributed to the
high aggregation of the input-output table, partly to the type of materials available in Italy;
− since endogenous steam is immediately used in electricity production, its indirect activation
by the vertically integrated branches mostly reflects the energy intensity of their deliveries to
final demand; this activation is substantially higher than average in minerals’ extraction and
OECD 41 2003
first processing (03, 13 and 15) and in coke production (05), while all other branches (except
tobacco products) are not far from the average of 25 kg. per LIT million of final products;
− similarly to the previous case, the presence of energy generation in the productive chains of
all vertically integrated branches implies that all branches indirectly activate the extraction of
energy minerals from the Italian territory; the only branches that show particularly high
values of indirect activation are the production of manufactured gas and electricity itself (09)
and transport activities (61 and 63), which are the main users of refined petroleum products;
− not surprisingly, indirect activation of non-energy minerals is very high in the construction
branch; it is also high in the beverages branch, probably due to the use of glass. Most of the
other industrial sectors are not far from the average of about 280 kg. per LIT million of final
product;
− biomass shows a very clear pattern of indirect activation, with all food industries (31-37) and
services of hotels, restaurants and bars (59) far above the average (which is 6.3 kg. per LIT
million), and all other branches far below it.
Figure 1. Average Intensity of use by vertically integrated sector
2,5
tons per LIT million
1,5 Agriculture, breeding, forest, fishery

Extraction, energy, cement...
Chemistry, rubber and others
1 Metals, machines, transport vehicles
Food and drinks
Textiles, wood, paper
Construction
0,5 Commerce and restoration
Transport and communications
0 Other services for sale
Non-energy minerals- domestic
Biomass- domestic
PA and non market services

Endogenous steam- domestic
Non-energy minerals- imported

Energy minerals- domestic
Biomass- imported
Undetermined materials- imported
Energy carriers- imported
As shown above, the branches through which imported materials enter the economy are much
more numerous than those extracting materials from the domestic territory. This is reflected in the much
smoother profiles of the total (direct and indirect) intensities of use of imported materials with respect to
those of the materials of domestic origin. As a matter of fact, imported intermediate inputs are demanded
by almost all vertically integrated branches. The main feature of the resulting picture is a strong
differentiation between service activities on the one hand and all other activities on the other hand, with the
OECD 42 2003
only noticeable exception of the textiles-wood-paper compound, which shows unitary activation
coefficients that are almost always the lowest among industrial activities.
3.3 Intensities of use of materials by type of final demand
Intensities of use can be calculated not just by vertically integrated branches, but also, using a
different aggregation of the data, by category of final demand, basically corresponding to the different
institutional sectors to which final goods and services are delivered. In Table 2 we show an example of this
type of analysis, that allows to identify the most material-intensive final users (the numbers in brackets
give the share of total final demand going to the respective uses). Attention can be drawn, for instance, on
the particularly high values of some material intensities of households’ consumption. To show this in a
different way, also the ratio between the share of final demand and that of material input can be
considered: if this higher than unity, the corresponding intensity of use will be higher than average. Thus,
households consumption accounts for “only” 52% of final aggregate demand, it activates 68% of
endogenous steam and of energy minerals of domestic origin, 86% of domestic biomass, 67% of imported
energy minerals and 73% of imported biomass. Since expenses for buildings are investments (inventories if
not finished), the values present in the row of domestic non-energy minerals are not surprising.
Table 2. Table 2. Materials’ intensities of use of final demand by institutional sector

Italy 1988 (kilograms per LIT million of final demand)
Households P.A. Investments Inventories Export Overall

(52%) (15%) (16%) (1%) (16%) average
Endogenous steam 34 12 17 28 22 26
Domestic energy minerals 21 7 9 6 16 16
Domestic non-energy minerals 92 75 831 1.590 436 281
Domestic biomass 104 11 3 78 51 63
Undetermined imported materials 6 2 4 17 10 6
Imported energy minerals 50 15 23 12 42 38
Imported non energy minerals 34 16 142 138 84 58
Imported biomass 47 7 16 31 30 33
3.4 An indicator of the trade-off between ecological and economic objectives
As explained above, the same operation of “vertical integration” can be (and traditionally is)
performed on variables other than material inputs (or outputs), such as value added, profits, employment,
taxation revenue and so on. Comparing the results for material flows with the results for these other
variables gives interesting measures of the sector-specific trade-offs faced by an economic policy for
dematerialisation. Table 3 summarises the results of this comparison.
The results are quite similar to those concerning the intensities of use shown in Figure 1. The gap
between service activities and the rest of the economy is confirmed: one unit of value added is generated
with the need of much less materials in services than elsewhere. This is not as obvious as it may seem: it
must be kept in mind that we are considering all the upstream material flows, and not just direct inputs, to
the final production of services. Only in the textiles-wood-paper vertically integrated sector the shares
between material inputs and value added are not much different from those of service activities, while all
other vertically integrated sectors display a strong specialisation in the use of some specific materials,
clearly of great importance for their production activity, like for instance in the case of non-energy
minerals for constructions, and of biomass for food industries.
OECD 43 2003
Table 3. Table 3. Ratio between activated material input and value added by groups of NACE branches
Italy 1988 (Kilograms per LIT million of domestic product)
Intra-regional Inputs Imported Materials

Non- Non-
Endogenous Energy Undeter- Energy
energy Biomass energy Biomass
Steam minerals mined minerals
minerals minerals
Agriculture, breeding, forests,
34 1721 19 10 10 66 11 20
hunting and fishing
Extraction, energy, cement... 2541 3 283 217 156 8 4 323
Chemistry, rubber and others 397 21 37 17 153 43 59 76
Metals, machines, vehicles 185 4 26 9 149 10 6 31
Food and drinks 127 613 29 12 35 186 6 28
Textiles, wood, paper 92 20 28 9 19 103 14 22
Construction 1478 3 16 9 179 13 5 24
Commerce and catering 54 47 17 10 9 19 1 19
Transport and communications 56 7 13 35 20 8 2 78
Other services for sale 62 3 11 4 10 3 2 8
PA and non-market services 76 11 13 7 16 8 2 15
4. Conclusions
Accounting of the material inputs of an economy is not of much use if we are not able to analyse
these flows, i.e. to say something about their socio-economic determinants, on which policies have to act in
order to cause our way of life to evolve in overall terms towards a less material well-being
(“dematerialisation”). One of the most important instruments for the analysis of determinants is input-
output analysis, via which technological (interlinkages between economic activities) and structural (relative
shares of the branches of the economy in final demand) components of the overall demand for materials
from nature can to some extent be disentangled.
With this kind of analysis it is possible, as we have shown, to identify the kinds of final demand
(according to the products demanded and to the subjects from which it originates) which are “responsible”
for the removal of materials from nature at the start of the production chains, as well as for the creation of
value in the economy. In turn this allows to quantify the reductions of economic activity and/or gains in
efficiency needed to achieve environmental goals expressed in terms of material inputs, and to simulate the
repercussions on economic and social variables of the different changes of demand or technology that can
lead to the achievement of those goals. This is crucial for the purpose of identifying sustainability policies
that are at the same time effective on the environmental side an not unbearable on the economic side.
OECD 44 2003
Figure 4. Annex 1. Material inputs by vertically integrated branch, Italy 1988 (tons)
En d o g e n o u s ste a m - d o m e stic En e rg y m in e ra ls - d o m e stic
0 2000000 4000000 6000000 8000000 10000000 12000000 0 1000000 2000000 3000000 4000000 5000000 6000000 7000000 8000000 9000000
01 01
05 05
09 09
13 13
17 17
21 21
25 25
29 29
33 33
37 37
41 41
45 45
49 49
53 53
57 57
V e rtica lly inte gra te d bra nche s

61 61
65 65
69 69
73 73
77 77
81 81
89 89
45
Figure 5. Annex 1 (continued). Material inputs by vertically integrated branch, Italy 1988 (tons)
Non-energy minerals - domestic Biomass - domestic
0 30000000 60000000 90000000 120000000 150000000 0 5000000 10000000 15000000 20000000 25000000 30000000 35000000
01 01
05 05
09 09
13 13
17 17
21 21
25 25
29 29
33 33
37 37
41 41
45 45
49 49
53 53
57 57
Vertically integrated branches

61 61
65 65
69 69
73 73
77 77
81 81
89 89
46
Non-energy minerals - imported Biomass - imported
0 4.000.000 8.000.000 12.000.000 16.000.000 20.000.000 0 1.000.000 2.000.000 3.000.000 4.000.000 5.000.000 6.000.000
01 01
05 05
09 09
13 13
17 17
21 21
25 25
29 29
33 33
37 37
41 41
45 45
49 49
53 53
57 57

61 61
65 65
69 69
73 73
77 77
81 81
89 89
47
Energy minerals - imported Undetermined - imported
0 3.000.000 6.000.000 9.000.000 12.000.000 0 500.000 1.000.000 1.500.000 2.000.000
01 01
05 05
09 09
13 13
17 17
21 21
25 25
.
29 29
33 33
37 37
41 41
45 45
49 49
53 53
57 57
61 61

65 65
69 69
73 73
77 77
81 81
89 89
48
MATERIAL FLOW ACCOUNTING AND ITS APPLICATION:

The experience of Japan25
1. Background
Japan experienced severe environmental pollution involving serious health damages in 1960s to
early 1970s, the era of rapid industrialization. End-of-pipe technologies for large point sources, such as
desulfurization and denitrofication have successfully contributed to improve such traditional environmental
pollution problems. Can we continue to rely on such end-of-pipe approach to solve all of emerging
environmental problems at the end of huge energy and material flows of the industrialized economy?
The answer seems to be rather negative. Many of present environmental issues have their roots in
the basic structure of our society, characterized by ‘mass-production, mass-consumption and mass-
disposal’, and we need to transform our production and consumption patterns to more sustainable manner.
Such recognition is clearly stated in recent Japanese national environmental policy documents such as the
Basic Environment Plan. Based on this standpoint, a basic law for promoting recycle-based society is
enacted in 2000 towards more sustainable material management. It seems that ‘looking upstream’ is being
built in environmental policies.
Such paradigm shift in Japanese environmental policy may be interpreted differently in global
and domestic contexts. Increasing attention is being paid to global environmental problems, and the
concept of sustainable development is being spread. Recognition that the environment is finite as a source
of resources supply and as a recipient of residuals is the most essential standpoint to discuss the sustainable
development. At local level, on the other hand, limitation of the end-of-pipe approach becomes evident
(Moriguchi, 1999). It has to be kept in mind that such recognition in Japan can not be prevailed without
urgent visible problems on municipal solid wastes and industrial wastes. We are suffering from the
shortage of final disposal site capacity, but the development of new dumping site faces to difficulty
because of potential negative impacts on the environment. Cost for dumping industrial solid wastes is
expensive enough to call industries’ attention for waste minimization. Incineration of solid waste has been
effective in reducing the size of final disposal, but recently revealed problem of dioxins from waste
incineration works as another driving force to recall people’s attention to negative aspects of mass-
disposal-society.
2. Overview of studies on MFA and inter-related tools
2. MFA and similar physical accounting

As will be later described more in detail, Japan highly depends on imported natural resources,
which often have environmental relevance. Harvest of timbers from tropical rainforest is a typical issue.
Japan’s participation to the OECD’s pilot study on natural resource accounting (NRA) in the beginning of
1990s (OECD 1994) was driven by such situation. This experience of the forest resource account was later
applied to Asian countries (Koike 1999).
25 Paper prepared by Mr. Yuichi Moriguchi, Social and Environmental Systems Division, National Institute for Environmental Studies,
Environment Agency of Japan.
49
On the other hand, material flow accounting (MFA) was studied mainly to respond to the
domestic issues of increasing solid wastes. A flow chart describing Japan’s macroscopic material flow
balance has been published on the annual ‘Quality of the Environment’ report, what we call ‘White Paper’
since its 1992 edition. Dissemination of an English edition of the report created the opportunities for
European experts on MFA to involve Japan to international collaborative efforts in this field. In 1995, the
SCOPE (Scientific Committee for Problems on Environment) organized a scientific workshop for
indicators of sustainable development at the Wuppertal Institute in Germany. Participants to this workshop
from four industrialized nations, Germany, the United States, the Netherlands and Japan agreed to launch
an international collaborative study to compare their overall material flows at national level. The results
will be shown later in this paper.
Inter-industrial flows of some individual materials such as non-ferrous metals have been studied
mainly from the viewpoint of material recycling (Clean Japan Center 1997). Substance flow analysis
(SFA), which captures the flow of specific elements of environmental concern, was applied to some case
studies such as an analysis of nitrogen flow and its impacts on eutrophication. The SFA framework for
toxic substances has yet been explicitly adopted until recently.
Inventories of pollutants’ emissions, what we call ‘emission inventories’ may be categorized as
one specific form of MFA in broader sense. Official inventories are compiled for Greenhouse gases
(GHGs) along with an international convention, whereas those for others, even for traditional air
pollutants, have not been made available until recently. This is mainly because institutional basis for
environmental statistics is rather weak compared to other countries. PRTR (Pollutant Release and Transfer
Register) system has been tested by pilot studies in late 1990s, and a nationwide system by the newly
enacted law will start in the fiscal year 2001.
2.2 Input-output analysis with environmental extension
Application of economic input-output analysis (IOA) including case studies for Japan was
undertaken by Leontief himself, the pioneer of IOA, in early 1970s. This might be one reason why
environmental application of IOA has been and is being active in Japan. Another reason is that the
Japanese IO table consists of some 400 sectors, and is thought to be one of the most detailed and qualified
ones in the world.
IOA was already applied for what we call ‘energy analysis’, in late 1970s. Energy consumption
by sectors is indicated in physical unit tables, which is officially accompanied with national IO tables.
Other official statistics on energy consumption are often used in order to supplement the data in physical
unit tables, of which data coverage and accuracy are not complete. Once sectoral direct energy
consumption per unit output is quantified, one can easily calculate overall sectoral energy intensity, which
includes indirect energy consumption in upstream industries by applying Leontief’s inverse matrix. This
calculation process has been applied to energy consumption, CO2 emission (Kondo et al. 1996), traditional
air pollutants (SOx, NOx) (Hondo et al. 1997), water pollutants (BOD, N, P), solid wastes, and so on. They
have been often used for life cycle inventory analysis, and other examples of application include an
analysis of structural changes of CO2 emissions from the viewpoint of final demand of the economy and an
analysis of influences by international trade (Kondo et al. 1998).
More recently, application of IOA to the issues of waste management and recycling is getting
active. A Waste Input Output (WIO) model was proposed (Nakamura 2000) to describe an
interdependence of goods producing sectors and waste management sectors, in which both monetary and
physical flows were dealt with. Description of whole material flows within the economy and their
interaction with the environment is also attempted, to link sectoral IO studies and macroscopic MFA
studies. Learning from the German pioneering experiences (Stahmer et al. 1998) in PIOT (Physical Input-
Output Tables), a framework of 3DPIOT (3 dimensional PIOT) is being proposed, and case studies are
50
being undertaken. These environmentally extended IO studies have many common features with MFA
studies.
2.3 Monetary environmental accounting
Monetary environmental accounting in micro economic level, i.e., corporate environmental

accounting is being widely spread, since the Environment Agency of Japan published a guideline in 2000.
Many leading companies disclosed their trial of environmental accounts around this period. In many cases,
physical accounts are not accompanied, but some companies published overview of physical material
flows around their activities as a part of environmental reporting.
Environmental accounting for macro-economy has been studied along with international
discussion on Integrated System of Environmental and Economic Accounting (SEEA). Worldwide experts
in this field gathered at Tokyo in 1996 (Uno and Bartelmus 1999), which stimulated the international
exchange of experiences. Economic Planning Agency of Japan published its trial calculation of
environmentally adjusted net domestic product (EDP). Though this monetary accounting study and the
physical material flow accounting study led by the author have been undertaken within the same research
project, they have not been fully integrated.
3. Characterization of Japanese Material Flows
3.1 Geographical and historical background
About 125 million people densely inhabit in Japan, whereas domestic stock of natural resources
is not sufficient and its exploitation is costly for sustaining these populations. Therefore, Japanese material
flows should not be discussed without international trade flows. The history of international trades of Japan
has an interesting profile. The Edo era when Japan closed the country is often referred to as a model of
environmentally sustainable society because of its self-sufficiency of resources. On the contrary, present
Japanese economy heavily depends on international trade, both imports and exports. Without tremendous
amount of imported natural resources, such as fossil fuels and metal ores, Japanese economy can not be
sustained. Growing export of products by raw material industries and assembly industries have been a
major driving force of rapid economic growth.
In the following sections, Japanese material flows will be characterized, based on a database
recently compiled through participation to the international joint project on MFA (Adriaanse et al., 1997,
Matthews et al. 2000). By the joint study, in addition to direct material inflows and outflows, hidden flows
(originally named ecological rucksacks in Germany) were quantified. They refer to ancillary material and
excavated and/or disturbed material flow along with the desired material.
3.2 Material inflows
An overview of Japanese material flows is shown in Fig. 1. Material input flows that support the
Japanese economy are characterized by high dependency on the import of natural resources. Imported
commodities account for about one-third of the mass of direct inputs (DMI) to the economy, and account
for one-half of the total material requirement (TMR), which is the a sum of the DMI and hidden flows.
Imports provide the Japanese economy with essential materials, including fossil fuels, metal ores,
agricultural and forestry products. Import dependency is particularly high for metal ores and fossil fuels, as
we have only poor stocks of these resource categories in domestic territory. Dependency is also high for
timbers, in spite of the fact that the two third of our land is covered by forest. Timber imports to Japan
contribute significant portion to whole international trade in the world. Recent trends revealed that
51
commodities increasingly tend to be imported in more manufactured form, for example, refined metals
rather than metal ores, or plywood rather than round wood. Imports of semi-manufactures and final
products have also been increasing. Large hidden flows are associated with metals (particularly with
copper and iron), coal, as well as agricultural and forestry products.
Domestic material flows, both commodity mass and rucksacks, are dominated by construction
activities, nearly 90 percent in early 1990s. Domestic construction minerals including limestone, crushed
stone and sand and gravel are used to create and improve buildings, roadways, water reservoirs, and other
infrastructure.
3.3 Material outflows
The largest output flow from the economy to the nature is emission of carbon dioxide. Apart
from the fact that CO2 is famous as a greenhouse gas, which is also of our important concern, we have to
recognize another fact that CO2 is the heaviest waste from industrialized economies.
After CO2, waste disposal to controlled landfill sites is the next major component of Direct
Processed Output (DPO). This is of greater environmental significance than the nominal weight implies.
Japan has a shortage of landfill sites for waste disposal. Reclaiming coastal areas for this purpose has
sometimes caused the decrease of habitats for wildlife. The weight of waste disposal to landfill is much
smaller than that of waste generated. The difference between the amount generated and the amount sent to
landfill is the amount recycled or reduced by incineration and drying. Three quarters of MSW is
incinerated to reduce waste volumes. The amount of landfill wastes was almost constant until 1990, but is
now decreasing trend, thanks to waste minimization and recycling efforts.
Input of food and feed are balanced by return flows of CO2 and water after digestion. Dissipative
use is another important category of output flows. Dissipative flows are dominated by applications of
animal manure to fields. Fertilizers and pesticides are intensively used in Japanese agriculture to enhance
productivity and compensate for the limited area of available farmland.
However, total of these output flows is still much less than input flows. This is because more than
half of direct material inputs is added to the stock, including consumer durables, capitals of industries and
public infrastructures. They can be deemed as potential sources of wastes in the future.
3.4 Domestic Hidden Flows
Excavated soil during construction activities dominates domestic hidden flows. Only surplus soil,
which means the soil excavated then moved out of the construction site to landfill or other sites for
application, is quantified by official surveys. The total quantity of soil excavation by construction activities
is much greater, because excavation work is usually designed to balance ‘cut and fill’, to use excavated soil
on site, and minimize the generation of surplus soil.
Hidden flows associated with mining activities are trivial in quantity, because of the limited
resources of fossil fuels and metal ores in Japan. Consequently, the contribution of domestic hidden flows
to Total Domestic Output (TDO) is relatively small, compared with more resource-rich countries. It should
be borne in mind that the small size of domestic hidden flows is counterbalanced by imported hidden flows
associated with imported metals and energy carriers; this represents the transfer of Japan’s environmental
burden to its trade partners, which the first joint report emphasized.
3.5 International comparison of MFA-based indicators
Japanese TMR is about 45 tons per capita, which is much lower than other three countries
(around 85 tons per capita). This is mainly because smaller energy consumption per capita and lower
52
dependency on coal. In terms of DMI per capita, Japanese figure is only slightly smaller than Germany and
the United States. DMI per capita for the Netherlands is also in the similar level, if huge transit import
flows to other European countries are excluded
Dependency on imported material flows varies largely among countries, from less than 10 % for
the United States to 70 % for the Netherlands. Ecological rucksacks accompanied by imports imply various
environmental impacts in trade partners. More specific analysis will be necessary to identify individual
problems behind ecological rucksacks.
The absolute level of DPO per capita in Japan is about 4 metric tons without oxygen and 11
metric tons with oxygen. These values are relatively small among the countries studied.
3.6 Recent trends
According to the analysis of historical trends of indicators, Japanese TMR per capita has shown
upward trends. This coincides with the increase of waste flows. DPO and TDO in Japan grew 20 percent
during the period 1975-1996. These growths occurred mainly after the late 1980s. Before then, DPO was
almost constant and TDO decreased slightly.
On per capita basis, there was a downward trend in TDO from the late 1970s to the mid 1980s.
DPO per capita also decreased slightly in this period. Growth in DPO per capita and TDO per capita were
particularly evident in the late 1980s, when the country experienced the so-called ‘bubble-economy’.
Material output intensity, that is, DPO or TDO per constant unit of GDP declined until 1990
because of larger growth in the monetary economy than in physical throughput (the physical economy).
However, since 1990, de-coupling between economic growth and material throughput has not been
improved, because DPO and TDO have continued to increase, while economic growth has slowed down.
This recent trend can be explained by structural changes in energy consumption, i.e., thanks to relatively
cheap oil prices. Household energy consumption including gasoline consumption by private cars has
increased and contributed to larger CO2 missions, but this trend has contributed little to GDP growth.
Net additions of materials to the stock (NAS) in the Japanese technosphere have fluctuated in
accordance with patterns of governmental and private investment. NAS increased significantly in the late
1980s, then stabilized at a lower level in 1990. Because Japan has a shorter history of industrialization than
other Western countries, construction work is still active and significantly contributes to the country's
overall picture of material flows. As much as 60 percent of direct material input (DMI) is added to the
stock. This figure also has a close relation with inputs of construction materials as well as with soil
excavation. Increasing quantities of stock imply that demolition wastes will also increase in the future.
Currently, the government is attempting to encourage recycling of demolition wastes.
4. Use of MFA for the development of environmental indicators
4.1 Japanese Basic Environmental Plan and indicators development
In 1993, the law named "Basic Environment Law" was enacted as a new basic law for
environmental policy in Japan. In 1994, the first "Basic Environmental Plan" was established by cabinet
decision based on the Law. The plan mandated the government to develop comprehensive indicators for
implementing the Plan and for reviewing policy performance towards four long-term policy goals set by
the Plan, namely ‘sound material cycle’, ‘harmonious co-existence’ with nature, ‘participation’, and
‘international activities’. These indicators are also expected to be utilized for revising the Basic
Environment Plan. An expert advisory committee was established in the Environment Agency to elaborate
the concept and show options of indicators, drawing upon related activities in international organizations
53
such as OECD and UNCSD, as well as those in other countries. The author has been involved in this
process as an expert to support the secretariat of the committee. After seven times of plenary meetings of
the committee since November 1995 as well as many meetings of subsidiary working groups, the
committee published "A draft set of comprehensive environmental indicators" in July 1997. The draft set
was disseminated to the related Ministries within the national government, local governments, academic
societies, industries, as well as NGOs to invite their comments on the draft set and to improve it to
establish final set of indicators. In November 1999, the committee submitted a revised report to the Central
Environmental Council, to be used for the planning process of new Basic Environment Plan that is
scheduled to be decided by the Cabinet in FY2000.
4.2 Outline of indicators set and relevance to the MFA
Summary of the set of indicators for four policy goals is tabulated in Table 1 and Table 2. The
set is not based on a single framework, but incorporate multiple different frameworks. Out of the four
policy goals of the Plan, the first one, "sound material cycle" has direct relevance to the MFA. Up-to-date
information about international research activities for the MFA was introduced in the early stage of
discussions at the advisory committee. The set of indicators for sound material cycle is subdivided into
three subsets. The first is a subset of indicators for material and energy flows, which directly deals with
material flows. The second is a subset for sound water cycle, which is also relevant to the MFA in broader
sense. The third is a subset of indicators using D-S-R framework for environmental issues associated with
material and energy flows.
The first subset indicators for material and energy flows stands for the recognition that the
prevailing socioeconomic activities and lifestyles marked by mass-production, mass-consumption, and
mass-disposal are essential sources of the burdens on the environment. This subset is further subdivided
into indicators for material flows and those for energy flows. Indicators for material flows include those for
inputs of resources, outputs of wastes as well as recycling of materials. Needless to say, indicators in this
field are strongly influenced by recent progress in international research activities for the MFA. For
example, not only direct material input (DMI) but also the concept of hidden flows (ecological rucksacks)
is included in indicators.
4.3 Pending issues of the MFA for indicators development
The concept and practical application of the MFA contributed much to the process of preparation
of the draft set of indicators. Data compilation through international joint study with Austria, Germany, the
Netherlands and the United States was timely, which enabled us to present actual figures of historical
trends of some indicators. However, there still remain many issues to be solved.
The first essential problem is data availability, in particular, poor data availability for estimating
imported hidden flows. Japanese part of the international joint study mostly relied on the data provided by
the Wuppertal Institute, but we may have to collect our own country-specific data reflecting origin of
natural resources imported to Japan.
Many other detailed technical questions arise from indicators of recycling. For example,
definition of "recycle ratio of specific materials" is not simple. Denominator could be domestic production,
supply including import, consumption, or consumption plus abandoned stocks. Numerator could be
recovered, input of secondary material to production, production of refined material from secondary input,
including/excluding imported secondary materials and/or materials refined from recycled materials.
Disaggregation of the MFA by industrial sectors and by type of materials is another important
subject, in order to link indicators and policy measures to improve environmental efficiency of industrial
54
activities. If this disaggregation of MFA is made consistent with the Input-Output table, it will provide us
with further opportunity of integrated analysis of the environment and the economy.
5. Conclusion and future perspective
Recently, studies on Material Flow Accounting in Japan have been activated through
international collaborations in this field, and MFA has already been used as a part of the state of the
environment reporting. The use of MFA in the development of environmental indicators for the Basic
Environmental Plan is just a first step of potential policy applications of MFA. Planning and target setting
processes based on the new basic law for promoting recycle-based society may give another great
opportunity for policy-relevant use of MFA.
REFERENCES
Adriaanse, A., Bringezu, S., Hammond, A., Moriguchi, Y., Rodenburg, E., Rogich, D., and Schuetz,
H.(1997), ‘Resource Flows: The Material Basis of Industrial Economies’, World Resources Institute,
Washington D.C., 66pp.
Clean Japan Center ed. (1997), ‘Keywords of Recycling –3rd edition-’, 241pp. (In Japanese)
Environment Agency, Quality of the Environment in Japan, Annual publication.
Hondo, H., Tonooka, Y., Uchiyama, Y. (1998), ‘Environmental burdens associated with production
activities in Japan using an Input-Output Table’, CRIEPI Socio-economic research center report,
Y97017. (in Japanese with English abstract)
Koike, K. (1999),’Environment and Resource Account of Forest/Timber in Southeast Asian Countries’,
Proc. 3rd ConAccount workshop: Ecologizing Societal Metabolism, CML report 148, 69-73.
Kondo, Y., Moriguchi, Y., and Shimizu, H. (1996), ‘Creating an Inventory of Carbon Dioxide Emissions
for Japan－Comparison of Two Methods－’, AMBIO, 25(4), 304-308
Matthews E., Amann, C., Bringezu, S., Fischer-Kowalski, M., Huettler, W., Kleijn, R., Moriguchi, Y.,
Ottke, C., Rodenburg, E., Rogich, D., Shandl, H., Schuetz, H., van der Voet, E., and Weisz, H.
(2000), ‘The weight of nations –Material outflows from industrial Economies-‘, World Resources
Institute, Washington D.C., 125pp.
Moriguchi, Y. (1997), ‘Material flow indicators for the Japanese Basic Environment Plan’, Proceedings of
the ConAccount Conference 11-12 September 1997, (Wuppertal Special 6, Analysis for Action:
Support for Policy towards Sustainability by Material Flow Accounting), 108-114.
Moriguchi, Y. (1999), ‘Recycling and waste management from the viewpoint of material flow accounting’,
Journal of Material Cycles and Waste Management, 1(1), 2-9
Nakamura, S. (2000), ‘Input-Output Analysis of Waste Treatment and Recycling’, J. Japan. Soc.Waste
Man. Experts, 11(2), 84-93. (in Japanese with English abstract)
OECD(1994), ‘Natural Resource Accounts: Taking Stock in OECD Countries’, OECD Environment
Monographs No. 84, Paris, 30pp.
Stahmer C, Kuhn M, Braun N (1998) ‘Physical Input-Output Tables for Germany, 1990’, Eurostat
Working Papers 2/1998/B/1, 119pp.
Uno, K., and Bartelmus, P. (1998), ‘Environmental Accounting in Theory and Practice’, Kluwer,
Dordrecht. 459pp.
55
Table 1 Summary of "Set of comprehensive environmental indicators"

for the Basic Environment Plan of Japan (tentative translation)
*:core indicators
A. Sound Material Cycle +:supplementary indicators
(1) Indicators set for material and energy flows
Indicators for "Material" Flows Indicators for "Energy" Flows
*Direct Material Input (total) *Total Primary Energy Requirement
+DMI disaggregated by sector *Total Final Energy Consumption
+DMI per capita +TFEC disaggregated by sector
+DMI per GDP in constant price *TPER per GDP
+Domestic Hidden Flows(Ecological Rucksacks) *TFEC per GDP
+Imported Hidden Flows(Ecological Rucksacks) *Ratio of renewable energy to TPER
*Total waste generation +Energy efficiency by sector
+Waste generation disaggregated by sector
+Waste generation per capita
+Waste generation per GDP in constant price
*Recycle ratio of waste generated
*Recycle ratio of specific materials
(steel, aluminum, paper, glass, plastics)
+ Nitrogen and phosphorus cycle
(2) Indicators set for sound water cycle

*Natural water cycle indicator
=function (runoff ratio by type of ground surface, measures to enhance infiltration)
*Man-made water cycle indicator
=function (reuse of industrial water, public water and rainfall)
(3) DSR Indicators for environmental issues associated with material and energy flows
DSR indicators by 15 environmental themes (See separate table 2)
B. Harmonious Co-existence with nature

(1) State of the nature
Driving force indicators State indicators Response Indicators
*Land use conversion *Spatial continuity of forest *Protected areas
*Degree of human disturbance
to vegetation
*Indicative animal species
*Total number of spieces
- endangered spieces
+Function for land conservation
(2) Access to the nature

Actual Potential
*Indicators for recreational use of the nature * Indicators of space for recreational use
* Indicators of demand for recreational use
* Indicators of support for access to the nature
C. Participation
Indicators for actions by national government
Indicators for actions by local government
Indicators for actions by industry
* Indicators for actions by citizens
D. International activities
(1) Indicators for international partnership for global environmental policy
(2) Indicators for support to developing countries (fund & personnel)
Remark:This table is a tentative, not-authorized translation of Japanese original one.
56
Table 2. List of D-S-R indicators by 15 environmental themes

+:supplementary indicators
Environmental Theme Driving Force Indicators State Indicators
Global warming GHG emissions aggregated by GWP + Atmospheric Concentration of GHGs
Emissions of each GHG (CO2,CH4,
N2O,HFCs,PFCs,SF6,CFCs,HCFCs)
CO2 emission per capita
Ozone layer depletion ODS emissions aggregated by ODP + Total ozone amount above Japan
Acid Deposition SOx emissions Total deposition of SO4(2-) and NO3(-)

NOx emissions Deposition of NH4(+)
+ Deposition of H(+) or pH of precipitation
Photochemical oxidant NOx emissions Achievement rate of EQS for Ox
NMHC emissions Percentage of stations above
alarm level of Ox
Urban air pollution NOx emissions in specific polluted areas Achievement rate of EQS for NO2
Achievement rate of EQS for SPM
Hazardous air pollution T.B.D Achievement rate of EQS for
hazardous air pollutants
Nuisance Traffic volume of trunk roads Population exposed to noise above
(Noise, Vibration and Odors) Total number of the Shinkansen Railway trains
specified levels
or Achievement rate of EQS for Noise
+ Grievances about vibration
Areas exposed to aircraft noise + Grievances about odors
Organic pollution of water (Rivers) (Rivers)
Organic pollutants emissions Achievement rate of EQS for BOD
(Coastal areas & Lakes) (Coastal areas & Lakes)
COD emissions Achievement rate of EQS for COD,N,P
Nitrogen emissions
Phosphorous emissions
Safety of water T.B.D Achievement rate of EQS for
toxic substances
Marine environment T.B.D T.B.D
Soil pollution T.B.D Agricultural areas requiring measures

Urbanized areas requiring measures
Ground subsidence Drawn groundwater Areas where subsidence is recognized
Solid waste Amount of municipal wastes + Cumulative total of areas reclaimed by

Amount of industrial wastes disposal of wastes
Toxic waste T.B.D T.B.D
Environmental risk of T.B.D T.B.D

chemicals
Remarks:
Response indicators for all themes will be selected later.
This table is a tentative, not-authorized translation of Japanese original one.
GHG:Green House Gas

GWP:Global Warming Potential
ODS:Ozone Depleting Substances
ODP:Ozone Depleting Potential
NMHC:Non-Methane Hydro Carbons
EQS:Environmental Quality Standard
T.B.D:To be determined
57
MATERIAL FLOW ACCOUNTING: MATERIAL USE FOR NATIONAL CONSUMPTION AND

FOR EXPORT:
The experience of Sweden26
This report was prepared for the Third International Conference of the European Society for Ecological Economics
Theme 9. Ecological Economics Modelling and Material & Energy Flows that was held in the spring 2000 in Vienna,
Austria.
Abstract
Statistics Sweden has developed economy-wide material flow accounts for Sweden, as a part of
an assignment for the government. The main purpose with the analysis is to be able to link the flows of
materials in the society to the existing data on economic activity and environmental impacts. The approach
aims to improve our knowledge of how materials are used in society and to enhance the efficient use of
resources. The study shows possibilities of enhancing international comparability of aggregated data on
material use, by making a division between consumption and export purposes. Direct Material Input
(DMI), including production and imports, per capita amounts to 24-27 tonnes per capita for the period
1987-1998, with highest values 1989 and 1995. The fossil fuel input varies only slightly over the period,
from 3,2 tonnes/capita in 1991 to 3,6 in 1996, a level which is unsustainable according to the Swedish
EPA. Renewable raw materials vary between 8 and 9 tonnes per capita. The non-renewable categories ores
and minerals vary between 11 and 15 tonnes/capita. Around 5 tonnes per capita are exported, while the
rest, around 20 tonnes per capita, is national consumption. This calculation puts Sweden above the
estimates made for Germany, USA and Japan, and in the same size as the Netherlands. There are several
explanations for the differences, mainly the relative size of exports and the data sources. Total Material
Requirements (TMR) mounts to at least 45 tonnes per capita. Comparisons of the input with waste
statistics indicate that waste (excluding mining waste) amounts to about 10 % of the input. The recycled
materials are about 5% of the amount of virgin resources brought into society each year. The materials that
can be classified as hazardous for health (metals and imported goods excluded) amounts to around 20% of
the direct flows.
1. Introduction
Many environmental problems are connected to production and use of materials and energy. It
would therefore be desirable to have an information system that gives consistent and complete information
on material and energy flows. Such a system would even be more useful if it could be connected to
economic data. The environmental accounts provides such a framework, but comprehensive statistics on
material flows have not been available until now. Statistics Sweden is now in the process of developing
material flow statistics for Sweden. One part of this work focuses on an aggregate description of the total
material throughput for the society, with a methodology similar to what has been suggested internationally
(WRI, 1997). This knowledge can be used to work towards eco-efficiency by improving the resource
productivity. The results contribute to the work with environmental accounts and provide a link between
society’s use of materials and natural resource accounting.
We are interested in describing the environmental pressures from a nations economic activities
through its producers and consumers. The producers purchase energy, materials, labour and capital to
produce goods and services. This production process also produces waste, air emissions and water
emissions (Figure 1).
26 Paper prepared by Ms. Viveka Palm and Ms. Kristina Jonsson, Statistics Sweden (SCB).
59
Figure 8. Figure 1. Input to and output from the economy
energy goods
material services
Economy
work force waste
capital emissions
The environmental accounts aim at quantifying these resource flows and the relationships
between them. From a sustainable development point of view, the economic system should produce the
desired goods and services with a minimum of environmental impacts. This can only partly be achieved by
reducing emissions from specific processes by end-of-pipe solutions, and reducing waste by recycling.
The size and composition of the inputs in the production process is of great importance for the
resulting environmental pressure. This is particularly clear for carbon dioxide emissions from fossil fuel
input, as well as for the use and spreading of chemicals. As the economy grows, so do the inputs of
materials into the system. However, a counter-balancing factor is the energy and material efficiency gains
that occur in the development of new products and new equipment.
If, for example, these effects balance out each other, we are facing a development with constant
environmental pressures. This is not a satisfactory situation, as the recently defined Swedish environmental
quality objectives will only be met if the environmental pressures are reduced.
Until now, there has been no coherent statistics on material flows to follow the development of
material inputs. By establishing material flow statistics, policies on materials and eco-efficiency will have
a firmer point of reference.
The forming of material flow statistics has been guided by two principles: to start with existing
statistics, and to start with a wide range of materials. These principles lead to a suggestion to divide the
materials description into three different areas. Natural resources, Chemical indicators and Substance flows
for a limited number of chosen substances. The results couple to the newly suggested environmental
quality objectives of primarily ‘a non-toxic environment’, ‘a good urban environment’, and ‘limited
influence on climate change’. Here we will present some results from the natural resource area, expressed
as an aggregate description of the total material throughput for the society.
The model used is organised similarly as the System of National Accounts. Main primary
categories of direct material inputs (DMI) in the society are included (total inputs from foreign trade,
agriculture, forestry, mining and fuels). Time series from 1987-1997 are presented. The time series have its
main focus on the input side. Attempts are made to link the input data on natural resource use to emissions
and waste by means of input/output analysis. However, the results point to that a further disaggregation of
the monetay input/output tables are needed in future studies.
In the future, results from similar studies may show the increase of stocks in society. However,
due to incomplete solid waste data, no such estimates are included now. The time series have their main
focus on the input side, as the output statistics have not been produced with the same frequency. To make it
possible to compare the use of material between different sectors of society, the input and output is divided
into aggregated industries (by NACE).
60
2. Method
This study is an attempt to calculate the total national yearly input of raw materials with an
economic value. In order to avoid double counting, those materials that are taken from biosphere to
technosphere during the year are singled out. The method concentrates on the primary input, and avoids
calculating the materials, substances and products that have already been accounted for as primary input.
This means, for example, that figures from plastic industry are not of primary interest in this
work, since their raw material is already counted for as crude oil, industrial minerals or other primary
goods. It also means that meat from the food industry is mainly recorded as the feed to produce poultry,
etc. However, as the nation’s material use is also covered by imports of raw materials as well as products,
this method has its limitations. In accordance with similar studies, the import of goods is therefore
recorded separately. Ideally, these goods should of course have been recalculated to their primary input.
Since methods and data are lacking for that kind of recalculation, they are merely presented by their
weight.
The method gives an aggregated figure of yearly raw material input. It shows the total and the
share of renewable and non-renewable materials. In this form, it can be compared with figures from similar
studies, and possibilities for resource efficiency can be highlighted.
As a further development, the figures have also been complemented with export figures. This is
made in order to single out the input for national consumption from the input used in production for export
purposes. Sweden is dependent on its export of raw materials such as ores and pulp. If consumers in
Sweden would become more resource efficient, while the exports of raw materials to other countries
increased, it is important that our measures on resource use can distinguish between these phenomena.
Final demand is presented as national consumption and exports. National consumption is defined as
domestic production plus imports minus exports.
The food referred to in the report consists of beef, pork, lamb, reindeer meat, game, poultry, fish,
milk and eggs. Also cereals, fruit and vegetables. With regard to beef cattle, pigs and poultry, we have
chosen to show the passage from nature to society by means of the feed they consume. Where sheep are
concerned, we have assumed that the bulk of their fodder comes directly from nature, even if some
auxiliary feeding generally occurs.
3. Results
The total material flows consist of one third renewable materials and two third non- renewable
materials. Note that 20% of the non- renewable materials are fossil fuels. Construction minerals cause the
largest material flow, followed by forestry and ores. Domestic production in Sweden consists mainly of
construction minerals, ores and forestry. Construction minerals are consumed in Sweden, while ores and
forest products are exported to a large extent. Seventy per cent of the Swedish exports consist of ores and
forestry.
The Swedish imports consist mainly of fossil fuels. The greater part of raw material for food
production come from domestic production, only a small amount is imported. The use of raw material for
food production is mainly domestic material consumption, only a small part is exported.
One third of the domestic material consumption is renewable materials, mostly forestry.
Construction minerals also stand for one third of the domestic material consumption. Twenty percent of the
non- renewable materials consumed in Sweden are fossil fuels not stored in society, but consumed and
dispersed as carbon dioxide.
61
Figure 9. Figure 2. Material flows in the Swedish economy, 1997
DMI per capita

DMI per capita amounts to 24-27 tonnes per capita for the period 1987-1998, with highest values
1989 and 1995. The fossil fuel input varies only slightly over the period, from 3,2 tonnes/capita in 1991 to
3,6 in 1996. Renewable raw materials vary between 89 and 9 tonnes per capita. The non-renewable
categories ores and minerals vary between 11 and 15 tonnes/capita. Around 5 tonnes per capita are
exported, while the rest is national consumption (Figure 3).
62
Figure 10. Figure 3. DMI per capita in Sweden, 1987- 1997
International comparisons
This calculation puts Sweden above the estimates made for Germany, USA and Japan, and in the
same size as the Netherlands. There are several explanations for the differences. Differences can be
explained by system boundaries, by data sources and by how the exports are treated. In the table (Table 1)
below the comparisons are shown. (WRI, 1997 and own calculations). For the Netherlands, we have
chosen not to compare the disaggregated figures, as they were partly erroneous. For the total DMI, Sweden
and the Netherlands appear to be of the same size however.
For the fossil fuels, Sweden and Japan have similar per capita figures of about 3 tonnes/capita.
The industrial minerals are also on the same level as the other countries.
For the construction minerals, Sweden has a high figure. This can be explained by what statistics
have been available, as the economic statistics may have different system boundaries than what is wished
for. In this study, raw material statistics was often taken from other sources than the industrial statistics.
Earlier studies (Bergstedt et al, 1999) have shown that construction materials are underestimated in the
Swedish industrial statistics, as the cut-off on numbers of employees have excluded many small
companies.
For the renewable materials, Sweden’s per capita input appears to be about 5 tonnes higher than
the other countries, which is probably explained by the dominating forest industry.
The system boundary chosen for the ores has a major impact on the results. The international
comparison made it clear that earlier studies have had a boundary different to the one chosen in the English
study (Eurostat, 2000). In the international study from the World Resource Institute the concentrates of
metals have been considered direct flows, while in the Eurostat report we considered the ores (but
excluded the rock) as direct flows. In our Swedish study (SCB, 2000) the ores are recalculated as
concentrates, and those figures are presented here. The calculation based on ores rather than concentrates
gave much higher figures, between 2,0 to 2,8 tonnes/capita over the time period (not shown). A
preliminary calculation of the total material requirement (TMR), where so-called hidden flows such as
mining waste are included, shows figures of about 45 tonnes per capita.
63
Table 4. Table1. DMI in tonnes per capita for USA, Japan, Germany and Sweden in 1991
DMI/ capita DMC/ capita

USA Japan Germany Sweden Sweden
Non- renewables
Fossil fuels 8 3 6 3 2
Ores 1 1 1 2 0
Industrial minerals 0 2 1 1 1
Construction minerals 7 9 10 11 11
Total non- renewables 16 15 18 17 14

Renewables
Raw material for food production 2 2
Forestry 6 4
Total renewables 4 2 3 8 6
Total non- renewables and 20 17 21 25 20

renewables
Domestic material consumption

The DMI can be divided into domestic consumption and exports (see for example Hüttler et al,
1998). In order to distinguish the Swedish domestic material consumption, DMC, DMI have also been
complemented with export figures. This is made in order to single out the input for national consumption,
from the input used in production for export purposes. Sweden is dependent on its export of raw materials
such as ores and pulp. If consumers in Sweden would become more resource efficient, while the exports of
raw materials to other countries increased, it is important that our measures on resource use can distinguish
between these phenomena.
Final demand is presented as national consumption and exports. National consumption is defined
as domestic production plus imports minus exports. This estimate has its limitations and ought to be
refined in future work.
Especially for small countries such as Sweden or the Netherlands the effect of export goods may
dominate international comparisons between countries. To be able to assess the resource input for
consumption, export will have to be subtracted from the DMI.
Apart from construction minerals, the time series of material use are fairly constant during the
twelve year period. What natural resource use could then be regarded as sustainable? By comparing with
scenarios that have been produced by the Swedish EPA in a project on sustainable Sweden, the current use
of fossil fuels is clearly unsustainable, as a reduction of carbon dioxide emissions with more than 80% is
needed to reach the environmental goal on climate gases by 2050. A phase out of the fossil fuels put
greater stress on the use of forest resources both for transportation and heating purposes. A sustainable
level of wood use was also established, with criteria for meeting the environmental goals on biodiversity,
which is higher than today’s use. Also for the food sector some changes were recommended, in order to
reach the environmental goals on eutrophication. To compare these scenarios with the consumption of
material resources in tonnes is a future task. In future studies, we also intend to account for the energy
consumption needed to handle these materials today, in order to estimate the options for energy efficiency
measures such as recycling and substitution of materials.
4. Conclusions
DMI per capita amounts to 24-27 tonnes per capita for the period 1987-1998, with highest values
1989 and 1995. The fossil fuel input varies only slightly over the period, from 3,2 tonnes/capita in 1991 to
3,6 in 1996. Renewable raw materials vary between 8 and 9 tonnes per capita. The non-renewable
categories ores and minerals vary between 11 and 15 tonnes/capita. Around 5 tonnes per capita are
exported, while the rest is national consumption.
64
This calculation puts Sweden above the estimates made for Germany, USA and Japan, and in the
same size as the Netherlands. There are several explanations for the differences, such as the relative size of
exports and the data sources. For the fossil fuels, Sweden and Japan have similar per capita figures of
about 3 tonnes/capita.
For the construction minerals Sweden has a high figure. This may be explained by what statistics
have been available for the different studies. The so-called industrial statistics, which are primarily used for
economic assessments, have system boundaries that cover most of the economic flows, but not necessarily
the physical flows. In this study, raw material statistics was often taken from other sources than the
industrial statistics.
For the renewable materials, Sweden’s per capita input appears to be 5 tonnes higher than the
other countries, which may be explained by the dominating forest industry. Also for food production the
figures are higher than the USA, Germany and Japan. This may be due to the choice of statistic sources.
The fact that Sweden is a small country with a relatively large export of raw materials such as
forestry and ores is evident in the results. If we separate the export from the domestic material
consumption the per capita figures decrease with about 5 tonnes per capita. This difference is important to
be able to distinguish, in order to separate life style changes from exports of raw materials.
Acknowledgements
The results are based on two reports. A Swedish report, financed by the ministry of environment,
and an English report prepared on commission from EUROSTAT, who supports and co-ordinates
development of environmental accounts in the EU member states. The European commission (DG XI) has
contributed financially to the latter project. Several people have been engaged in different areas and in
preparing these reports. Thanks to Annica Isacsson (emissions and waste), Irene Linder (renewable
materials), and Anders Wadeskog (input-output analysis).
REFERENCES
Bergstedt Elisabeth, Linder Irene, DG Regional Policy and Eurostat, 1999. A Material Flow Account for Sand and
Gravel in Sweden.
Eurostat, 2000. Material flow accounts, DMI and DMC for Sweden 1987-1997. Report 2/2000/B/2
Hüttler Walter, Schandl Heinz, Weisz Helga, The department of Social Ecology of the Institute for Interdisciplinary
Studies of Austrian Universities (iff), 1998. Are industrial economies on the path of dematerialization?
Material flow accounts for Austria 1960- 1996: Indicators and international comparison.
Konijn, 1994. The make and use of commodities by industries. On the compilation of input-output data from the
national accounts.
SCB, 2000. En framtida nationell materialflödesstatistik. Miljöräkenskapsrapport 2000:4.
Schütz Helmut, Bringezu Stefan, Wuppertal institute, 1998. Economy-wide Material Flow Accounting (MFA).
SWEEA- Swedish Environmental and Economic Accounts. Statistics Sweden, Environment Statistics.
World Resources Institute et al, 1997. Resource Flows: The material basis of industrial economies.
65
PART II. METHODOLOGICAL ISSUES
67
MATERIAL FLOW ACCOUNTING – FRAMEWORKS AND METHODS:

Presentation by Eurostat27
Material flow accounts (MFA) have a role to play in informing policy formation. For example,
resource use and resource efficiency is listed among the future key policy issues in the Review of the EU’s
Fifth Environmental Action Programme (Decision N° 2179/98/EC) and the Presidency conclusions of the
informal meeting of the EU Environment Ministers in Helsinki, 23-25 July 1999. Key long-term themes of
the new (6th) EU Environmental Action Programme 2000-2010 will be related to material flows (incl.
climate change and use of resources). The UNSD Sustainable Development Indicator set and the EU
Environmental Headline Indicator set include Resource Use indicators.
This paper looks at the various kinds of MFA approaches (with a focus on economy-wide MFA),
their uses in policy, the frameworks and methods adopted for compiling MFA, the derivation of indicators
from the accounts and the interpretation of these indicators. Finally, the paper identifies key issues for
future advancement in MFA. As the basic framework for MFA is fairly simple, key issues in practice relate
to availability of the primary data needed for the accounts, their comparability across countries and over
time and the interpretation and uses of MFA results. As MFA will rarely be the only information source for
policy formation, links to, and complementarily with, other data sets and information systems need to be
fully understood.
1. Introduction
MFA are accounts in tonnes describing the extraction, transformation, consumption and final
disposal of chemical elements, raw materials or products. Many applications focus on the flows of
materials through the economy but some also include flows of natural origin and within nature. The first
law of thermodynamics (conservation of matter) provides the conceptual basis for MFAs and ensures that
inputs and outputs balance. The technique has been widely applied to materials and substances of specific
environmental importance including fuels, strategic metals, timber, pesticides, solvents, zinc, cadmium,
etc. The principles of statistical approaches towards material flow accounts and material balances have
been formulated in the 1970s.28 In Europe, MFAs are part of official statistics in several EU Member States
and EFTA countries.29
With environmental policy focus shifting towards waste, health/chemicals, climate change and
sustainable development, interest in materials management policies and accounts increased and recent
years saw a strong increase in MFA applications and research. The European Commission funded
ConAccount concerted action (1996-1998) was instrumental in advancing research and establishing
networks.
A key development in the 90’s was the wide application of economy-wide MFA and material
balances starting from early beginnings in Austria,30 Germany,31 Japan32 and the USA.33 Unlike single-
material or substance accounts, economy-wide MFA and the indicators derived from these accounts
provide an aggregate picture of the entire material and energy throughput of a society/economy. Terms
such as ‘industrial metabolism’34 or ‘societal metabolism’35 metaphorically suggest to consider modern
27 Papers prepared by Mr. Anton Steurer, European Commission - Eurostat.

28 United Nations (1976): Draft guidelines for statistics on materials/energy balances - UN document E/CN.3/493.
29 Eurostat (1997), Materials Flow Accounting: Experience of Statistical Offices in Europe. Luxembourg.
30 Steurer, A. (1992), Stoffstrombilanz Österreich 1988. Schriftenreihe Soziale Ökologie, Band 26. Wien.
31 Schütz, H. and S. Bringezu (1993), Major Material Flows in Germany. Fresenius Env. Bull. 2: 443-448.
32 Japanese Environmental Agency (1992), Quality of the Environment in Japan 1992. Tokyo.
33 Rogich, D.G., et al (1992), Trends in Material Use: Implications for Sustainable Development.
34 Ayres, R.U. (1989): Industrial Metabolism. In: Ausubel, J.H. and H. Sladovich (eds.): Environment and technical change, Washington DC
69
economies as living organisms whose dominance in, and impact on, a given eco-system can be indicated
by the size and structure of its ‘metabolic profile’.36 The material throughput of mature economies is a
central sustainability issues not just because of using up natural resources but also and foremost from the
point of view of reducing and managing the output of residuals and other environmental impacts associated
with the extraction, production or use of materials, including that socio-economic systems accrue more and
more of total biomass production and of land.
2. Mapping MFA activities with policy issues
MFA activities in research and statistics cover a wide field along several dimensions and do not
easily lend themselves to a simple categorisation. The materials covered range from bulk materials such as
water, wood or total material throughput to selected priority substances such as heavy metals or chlorinated
chemicals. MFA are made for a diverse range of economic, administrative or natural entities including
whole economies, industries or establishments; nations, regions, cities; and catchment areas or eco-zones.
The SEEA37 offers the following categories: flow accounts of non-produced raw materials, product flow
accounts, residual flow accounts and asset (stock) accounts. A distinction often found in research is
between material and substance flows, where substances tend to mean ‘pure’ chemical elements or
compounds and materials the actually observed flows of raw materials, products and residuals which are
often (but not always) a mixture of various substances.
In view of the many dimensions it may be unnecessary to develop a rigorous classification of
types of MFA. There are strong links, overlaps and synergies among different types and in practice various
aspects are treated simultaneously. For example, accounts for substances are often derived - by application
of technical coefficients - from the flows of observed ‘carrier’ materials (e.g. carbon content of biomass or
fuels, zinc or cadmium content of steel or fertiliser, etc.). The issue is rather to develop a set of tables
tailored towards the specific uses the accounts are made for.
Figure 1 below gives a very stylised presentation of the main area of interest for MFA accounts
in relation to categories of policy use. The materials in the ellipse are assumed to have environmental
impacts (specific impact per tonne multiplied by the volume of the flow) of broadly the same order of
magnitude. Flows outside this range are of lesser interest from an environmental point of view or simply
do not exist.
Three overlapping clusters or groups of materials may be identified:
− Low volume flows with a high specific toxicity (hazardous substances). The key issue is the
characteristics of the substances (eco-toxicity, etc.). Policy focuses on control of the uses or
releases.
− Basic industrial materials ranging from packaging materials to cement, steel and timber. Here
the policy focus is on materials and resource management, resource efficiency, waste
minimisation, etc. These materials are of interest for various reasons including life-cycle
considerations (e.g. highly polluting or energy intensive production processes, transport, etc.)
and economic importance.
− High volume flows ranging from e.g. timber or sand and gravel to the total material
throughput of an economy. In this area often more general sustainability considerations
apply, related to efficiency of use, long-term sustainability of supply, economic restructuring,
etc.
35 Fischer-Kowalski, M, and H. Haberl (1993), Metabolism and Colonisation. Modes of Production and the Physical Exchange Between
Societies and Nature. Wien.
36 Schandl, H. and Schulz, N. (2000): Using Material Flow Accounting to operationalise the concept of Society’s Metabolism. A preliminary
MFA for the UK 1937-1997. ISER Working Papers, Paper 2000-3, University of Essex.
37 United Nations (1993): System of Integrated Environmental and Economic Accounting, New York. The SEEA is currently under revision by
the London Group on Environmental Accounting. Draft chapters of the new SEEA are posted for public review at
http://ww2.statcan.ca/citygrp/london/publicrev/pubrev.htm.
70
The three clusters may also be related to types of policy instruments with ‘hard’ policies (bans
and phase-out protocols) dominating the policy mix for hazardous substances and ‘soft’ instruments for the
high-volume end (e.g. taxes, voluntary agreements or information). This in turn may be related to the
policy planning horizon which tends to be longer-term at the sustainability end (this does not exclude the
need for regular monitoring of e.g. hazardous substances).
Figure 11. Figure 1. A stylised map of the materials of particular interest for accounting
38
Source: Steurer (1996), as developed with W. Radermacher (StBA) in 1995
For some groups of MFAs there is an immediate policy and management use (e.g. linked to
chemicals policy or recycling). Others are intended to serve longer-term environmental policy objectives
and sustainability considerations (e.g. economy-wide MFA describing the material metabolism of a
nation). For the latter, the policy uses (and instruments) are only just developing. Future applications and
analyses have to prove that MFAs are a useful and efficient tool.
3. Frameworks for MFA

At a conceptual level MFA can be considered as the grand parallel to monetary economic
accounting and the importance of MFA (and physical accounting more generally) is widely accepted in the
context of environmental accounting.39 At a practical level the parallel does not hold as easily. While
money is a common numeraire, the goods consist of a vast range of materials with greatly varying
economic and environmental importance and impact. Moreover, money circulates and serves a variety of
transactions, whereas materials in circulation change their form and composition and losses (waste and
emissions) occur at each stage in the production and consumption chain.
38 Steurer A. (1996): Material Flow Accounting and Analysis: Where to go at a European Level. In: Statistics Sweden (ed.): Third meeting of the
London Group on Natural Resource and Environmental Accounting – Proceedings Volume, Stockholm, pp. 217-221.
39 Chapter 3 of SEEA 2000, draft for public review of 1 May 2000, available at: http://ww2.statcan.ca/citygrp/london/publicrev/pubrev.htm.
71
The basic principle used for MFA is the first law of thermodynamics which states that matter
(mass/energy) is neither created nor destroyed by any process but only converted from one form into
another so that for any transformation the total inputs and outputs are identical (taking account of residuals
and of changes in stocks). This allows to apply accounting techniques when making MFA, to interpret
discrepancies in a meaningful way and to fill gaps in basic data with informed estimates.
The basic framework for MFA is very similar to the national accounts framework. In its most
comprehensive form, the framework used is directly based on the input-output tables or supply and use
tables.40 The input-output framework provides for two basic equations:
1. along the rows of an input-output table (commodity flow) the basic equation for a given
commodity is: domestic production plus imports equals domestic use plus exports
(plus/minus changes in stocks). Domestic use can be intermediate consumption or final
consumption or fixed capital formation (i.e. investment goods).
2. down the columns of an input-output table (production function) the basic equation for a
given production process is: total input of materials equals total output of goods (including
by-products) and residuals (waste and emissions to air and water).
In both cases, the equations can be calculated in terms of a (composite) material or a particular
substance or element carried; or both, since often the accounting for substances is based on information on
the use of materials. For the whole economy the total amount of materials taken from nature (and
imported) equals the total amount of residuals generated plus the goods exported plus the net accumulation
of material in the economy.
MFAs often provide detail adapted to the characteristics of the material studied with separate
tables for:
− extraction/harvesting (e.g. agriculture, forestry, mining and quarrying)
− transformation/conversion (e.g. refineries, basic metal industries)
− intermediate uses (e.g. in manufacturing)
− final uses (e.g. households)
− accumulation (capital formation, durable consumer goods)
− imports/exports
− waste management and recycling
− residuals (waste and emissions)
− Flows and accumulation in nature
This general framework is in practice very much adapted to the characteristics of the materials
analysed and to their profile of use. For example, if uses are direct or dissipative (e.g. fuels, nutrients or
solvents) so that the materials enter the environment quickly, stocks do not play a major role. In other cases
the accumulation in the economy (e.g. for construction materials, metals or PVC) or in the environment
(e.g. heavy metals, persistent chemicals) is important. For some materials recycling is important (e.g. for
paper, glass, metals or CFC) but not for others, etc.
Actual applications can range from very elaborated and detailed accounts to very simple
indicators of apparent consumption only derived from production and foreign trade statistics. An example
for the first approach is the ‘Zinc Balance for the Netherlands, 1990’ by Statistics Netherlands, an example
40 See for example the Eurostat Working paper Nr. 2/1998/B/1: ‘Physical Input-Output Tables for Germany, 1990’, by C. Stahmer/M. Kuhn/N.
Braun of the German Federal Statistical Office, or ‘Material Flows and Input-Output Analysis’ by P. Konijn/S. de Boer/J. van Dalen (1995),
Statistics Netherlands.
72
for the latter approach is ‘Hazardous Chemicals’ by the Swedish Chemical Inspectorate and Statistics
Sweden41.
The level of detail chosen (and hence the resources needed) for an MFA will depend on the
purpose of the account and on the characteristics of the material. For the analysis of policy options and for
policy formulation more detailed accounts with links to economic information will be needed whereas
monitoring of the broad trends in material use will require less detail. Dangerous materials which are used
for a variety of purposes and for which substitutes are not readily available (such as zinc) will require more
effort. Data availability is of course important. For example the good representation of a material or
substance in the standard product classifications or the existence of a chemicals register can be very
helpful.
4. A framework for economy-wide MFA

Eurostat is developing a framework and recommendations for economy-wide MFA with the aim
of harmonising terminology, coverage, categories of materials to be distinguished and practical
compilation and estimation procedures. A project is ongoing to establish economy-wide MFA EU-wide
following this methodology. The basic outline of the framework is summarised in a composite material
balance (Fig. 2).
Figure 12. Figure 2. Economy-wide material balance with derived resource use indicators
INPUTS (origin) OUTPUTS (destination)
Domestic extraction Emissions and wastes
Fossil fuels (coal, oil…) Emissions to air
Minerals (ores, sand…) Waste landfilled
Biomass (timber, cereals…) Emissions to water
Dissipative use of products and losses
Imports (Fertiliser, manure, seeds, corrosion...)
DMI - direct material inputs DPO - domestic processed output to nature
Unused domestic extraction Disposal of unused domestic extraction

From mining/quarrying From mining/quarrying
From biomass harvest From biomass harvest
Soil excavation Soil excavation
TMI – total material input TDO - total domestic output to nature
Exports
TMO – total material output
Indirect flows associated to imports
TMR - total material requirements Net Additions to Stock (NAS)
Infrastructures and buildings
Other (machinery, durable goods, etc.)
Indirect flows associated to exports
Note: excludes water and air flows (unless contained in other materials).
Source: Eurostat (2000): Economy-wide Material Flow Accounts and Balances with derived Resource Use Indicators. Draft for Public
Review 10 October 2000.
41 See Eurostat (1997): Material Flow Accounting - Experience of Statistical Institutes in Europe, Luxembourg.
73
5. Indicators derivable from a material balance

A set of indicators can be derived from the composite balance including:
Input indicators
− Direct Material Input (DMI)
− Total Material Input (TMI)
− Total Material Requirement (TMR)
− Domestic Total Material Requirement (domestic TMR)
Output indicators
− Domestic Processed Output (DPO)
− Total Domestic Output (TDO)
− Direct Material Output (DMO)
− Total material output (TMO)
Consumption indicators
− Domestic material consumption (DMC)
− Total material consumption (TMC)
− Net Additions to Stock (NAS)
− Physical Trade Balance (PTB)
These indicators are linked by accounting identities. For example, DMI = DPO + NAS + Exports
= DMO + Exports, TMI = TMO + NAS, NAS = DMC – DPO, etc. DMC and TMC are calculated as
follows:
− DMC (domestic material consumption) = Domestic extraction (used) + Imports - Exports
− TMC (total material consumption) = Domestic extraction (used and unused) + Imports +
indirect (‘hidden’) flows associated to Imports - exports - indirect flows associated to Exports
More indicators could be derived from the accounts, e.g. by setting the boundaries of the
accounts differently or by compiling indicators per broad material category (biomass, fossil fuels,
construction minerals, etc.). Most of the indicators listed above, taken individually, can be derived from
individual accounts (see the sequence of accounts below) without the need to compile a complete material
balance. For presentational purposes it is useful to relate these indicators to socio-economic indicators such
as GDP or population.
Which resource use indicator will be considered the most relevant and useful in the longer term is
difficult to predict at this stage - only future use in analysis can provide a sound basis for such
recommendations. The choice of the most relevant indicators will depend on the policy focus and on
proven usefulness and applicability of indicators in policy analysis. At this stage, only a set of criteria for
the selection of indicators can be offered:
− Ease of understanding the meaning of an indicator
− Ease of compilation
− Compatibility with the national accounts
− Potential for policy use
− Data availability
− Completeness of the indicator
At present it appears that good candidates for core indicators would be the input indicators DMI
and TMR as well as the consumption indicators DMC and, maybe, TMC (the latter being difficult to
74
estimate because of the need to estimate also indirect flows associated to exports). NAS and PTB may be
interesting supplementary indicators.
‘The Weight of Nations’42 documents the feasibility and usefulness of output indicators.
However, the availability of primary data for regularly compiling output indicators is currently more
limited than for input indicators (e.g., data on waste disposal and emissions to water are often incomplete).
Compilation of fully-fledged economy-wide material balances (or aggregated input-output tables)
describing both inputs and outputs of materials by broad material groups is instrumental for ensuring that
the accounts balance but is also resource-intensive. A (sub)set of economy-wide material flow accounts
should be compiled annually, starting with accounts for direct material flows. Full-fledged economy-wide
material balances should be compiled at regular intervals, for example linked to availability of e.g. waste
and waste water statistics.
6. A sequence of accounts
For practical compilation work a system of accounts has been set up. The sequence of these
accounts is such that the material flows for which data are more likely to be available are presented first.
Progressing through the sequence of accounts, more primary data and compilation work will be required.
The individual accounts are:
1. Direct Material Input Account,
2. Domestic Material Consumption Account,
3. Physical Trade Balance Account,
4. Net Additions to Stock Account,
5. Physical Stock Account,
6. Direct Material Flows Balance,
7. Unused Domestic Extraction Account,
8. Indirect Flows Account,
9. TMR account,
10. TMC account.
It is important to have the indicators in a long time series in order to identify longer-term trends
and isolate changes that are due to economic cycles.
7. Basic data
Data sources include the following:
− Industry or production statistics: extraction of non-agricultural raw materials, recycling
(partly),
− Agricultural and forestry statistics: extraction/harvest of renewable resources,
− Trade statistics: imports and exports,
− Energy statistics, i.e. energy balances: extraction of energy carriers,
− Environmental statistics: several output flows (air emissions, solid waste, waste water) and
recycling,
These data sources will be complemented by estimates as needed. Information from other
institutions (trade associations, government agencies) and especially direct contacts with the producers of
the materials under consideration are also often used.
42 Matthews, E. et al (2000): The Weight of Nations: Material outflows from industrial economies, World Resources Institute, Washington D. C.
75
8. Selection and interpretation of indicators

Resource use and resource efficiency has emerged as a major issue for long-term sustainability
and environmental policy at EU and Member States level. Objectives include to substantially increase the
resource efficiency of the economic system, thereby reducing the use of natural resources and the related
negative impacts on the environment. Two main themes have been identified as policy relevant: the ‘total
quantity used’ and the ‘efficiency in use’. This implies that the resource issue should be analysed in terms
of ‘resource efficiency or productivity’ - a main current policy focus - and in terms of absolute use levels
and the ‘scarcity’ of resources.
Resource use indicators must be seen in context with other indicators. For example, in the EU set
of Headline Indicators the resource use indicator is complemented by other resource-related indicators,
such as ‘water quantity’ (water use), ‘waste’, ‘climate change’ or ‘land use’. Thus, a resource use indicator
should focus on the total resource use by the economic system, thereby complementing other indicators
which focus on specific environmental themes. For the first publication of EU Headline Indicators,
Resource Use will be described by Gross Inland Energy Consumption, in addition to a material use
indicator.
As resource indicators are typically related to GDP at constant prices to show trends in resource
productivity (efficiency), it is useful to compare the way resource use indicators are constructed with the
way GDP is defined. GDP can be defined in several ways. The closest parallel to material flow accounting
can be established for the calculation of GDP as the sum of final uses. The textbook formula for this is
GDP = C+I+X-M, with C for final consumption of households and government, I for investment (gross
capital formation), X for exports and M for imports. Exports are added and Imports are deducted whereas
for material input indicators Imports are added, and for DMC and TMC Exports are deducted. This is
because physical and money flows have opposite directions.
Consumption indicators are the closest equivalents to national accounts aggregates (GDP, NDP
or gross capital formation, as the case may be). DMC or TMC are thus best integrated with macroeconomic
performance indicators such as GDP and better suited for calculating overall efficiency or productivity
measures in which the numerator and denominator are consistent. Note also that key energy indicators such
as Gross Inland Energy Consumption are consumption indicators.
9. The trade bias in material input indicators

The difference between input and consumption indicators can be of practical importance,
especially for small economies. Recent results for Austria and Finland43 illustrate that efficiency increases
are higher when measured as GDP (at constant prices) per tonne of material consumption (TMC and
DMC) and compared with GDP per tonne of material input (TMR and DMI). and that this gap widens over
time. One reason for this is that both imports and exports of materials have grown fast over the past
decades, due to trade liberalisation and EU accession (internal market). Special factors may magnify trade-
induced biases, e.g., the ‘Rotterdam’ or ‘Antwerp’ effect.44 Table 1 shows preliminary results of ongoing
analysis related to differences between DMI and DMC. As could be expected, for the EU as a whole the
difference is small and does not change much over time, whereas for small economies the differences tend
to be more important (and increasing).
43 Muukonnen (2000): Material flows accounts - TMR, DMI and material balances, Finland 1980-1997, Eurostat Working Paper 2/2000/B/1,
Luxembourg. Mäenpää, I. (1999): Towards a sustainable Finnish economy, Helsinki. Gerhold, S. and B. Petrovic (2000): Material flow
accounts - material balance and indicators, Austria 1960-1997, Eurostat Working Paper 2/2000/B/6, Luxembourg.
44 Rotterdam (NL) and Antwerp (B) are important EU entry points for materials. For an analysis of the Dutch case see Adriaanse, A., S.
Bringezu, A. Hammond, Y. Moriguchi, E. Rodenburg, D. Rogich, H. Schütz (1997): Resource Flows - The Material Basis of Industrial
Economies. Washington: World Resources Institute.
76
Table 5. Table 1. Comparing resource use indicators – DMI and DMC

DMI in % of DMC, 1997 Increase in distance DMI – DMC between 1980
and 1997, in %
EU-15 105.5 EU-15 1.3
IRL 107.7 EL 3.1
E 109.7 F 3.1
EL 112.0 E 3.2
P 112.0 IRL 3.5
I 113.8 D 4.5
D 114.1 I 5.9
F 116.9 P 6.3
FIN 119.2 UK 7.2
A 120.1 A 8.7
UK 120.1 DK 10.4
DK 123.9 FIN 11.9
S 129.3 NL 15.3
B/L 180.2 S 16.6
NL 187.8 B/L 22.6
Source: Eurostat and Wuppertal Institute (2000), preliminary results
There may also be other reasons for possible biases in input indicators: For example, the quickly
increasing domestic extraction of North Sea oil in the 70s/ 80s and the subsequent increase in physical
exports (partly compensated by a decrease in imports) in the UK or Norway could affect time series of
input indicators. However, comparable data and analyses are still limited. More data, research and analyses
are needed before firm conclusions on indicator selection and interpretation and on the practical
importance of country-specific factors can be drawn.
Apart from ensuring comparability of data sets across countries, more research will be essential
to guide interpretation of indicators in future. This includes research into the precise meaning of the
various indicators and their relation to key socio-economic indicators, more experience with analytical
applications and analysis and improved understanding of the reasons for changes in the indicators.
Analysis of the correlation between different indicators is highly relevant in this context.
10. Conclusions and outlook
The basic framework of MFA is simple and universally applicable. The main data sources are
often standard statistics of output, inputs and trade. The details of the accounting procedures are adapted to
the characteristics of the materials chosen. For accounts of e.g. hazardous materials technical expertise as
well as direct information from individual firms is often as important as the standard statistical sources.
Approaches vary from simple indicators of apparent consumption derived from standard statistics
to detailed analyses of all aspects of the materials under consideration, even including flows of natural
origin or within environmental media.
Material flow accounts are often made at non-regular intervals, or as an ad hoc reaction to
specific user demands or policy debates.
The resource requirements for MFA vary but are often considerable. Regular accounts can only
be made for a limited number of materials/indicators.
For economy-wide MFA, data availability is a crucial factor for defining the scope (and level of
detail) of the accounts and the derived indicators. As a first heuristic, data on the input side are often
available in long time series, whereas data on the output side (e.g. waste) still need some development.
Data on indirect (‘hidden’) flows are generally not directly available and must be estimated.
A material balance approach (or even fully-fledged physical Input-Output tables) will facilitate
derivation of indicators tailored to satisfy future policy needs. The underlying data-structure will support
more detailed analyses. However, primary data needed for full material balances are not always readily
77
available. Simple input or consumption indicators should thus be produced annually, whereas full material
balances and physical input-output tables could be compiled at intervals.
MFAs are rarely the sole tool for decision support. Rather, they complement other sets of
information. The links to other information systems need to be taken into account to optimise the MFA
work and to ensure compatibility with other approaches (e.g., the NAMEA system45).
The role of imports and exports is worth considering to assess a nation’s contribution to world-
wide trends on resource exploitation (including e.g. land use or emissions abroad due to imports).
Therefore, indirect or ‘hidden’ flows should be included in the framework and development of data on
indirect flows advanced.
Economy-wide MFA have seen enormous progress in recent years. It is essential now to:
− ensure comparability of the accounts across countries so as to support policy use and
analysis,
− advance the understanding of the meaning of the various indicators that can be derived from
the accounts,
− advance the analysis of results, including by analysing the correlation among indicators and
by identifying the impact of country-specific factors on the indicators,
− identify ways to reduce the costs of compilation.
45 For an overview see e.g. European Commission (1999): Pilot Studies on NAMEAs for air emissions with a comparison at European level,
Office for Official Publications, Luxembourg.
78
MATERIAL FLOW ACCOUNTING – METHODOLOGY AND FRAMEWORKS:

Presentation by the United States46
1. Introduction
All accounting for material flows (MFA) has, at its basis, a mass balance approach to
understanding economic activity. Whether at the macro-level of a national economy, or on the scale of a
region, economic sector, firm, substance, or even product life cycle, MFA, in principle, attempts to account
for all material inputs to that scale, the transformations, if any, of those inputs, and all the final outputs
from that scale – products, emissions, leakages, and wastes – by balancing the materials that go into the
economy with those that leave the economy (less materials added to stocks-in-use). The purpose of
understanding an economy in material terms is to identify inefficiencies, understand the
environment/economic nexus, and provide policy makers with the information necessary to set priorities,
identify points of policy leverage, and gain insight into the potential unforeseen consequences of material,
technological, and policy intervention.
2. The national economy

The goal of MFA at the national level is to provide an overall picture of material flows and so
identify broad trends, the magnitude of various flows, and the interaction of those flows with the monetary
economy. MFA at this scale allows the creation and tracking of indicators of the whole economy – a
necessary overview of how the economy works and where important flows exist. Because of its broad
nature, MFA at the national scale cannot be a materials management tool, but it can highlight areas where
such tools are needed.
The data required for painting a material picture of the national economy is much less detailed
than that required to understand smaller economic scales. While detailed inventories of materials at the
smaller levels would be valuable in constructing the national picture, in practice investigators are forced to
deal with only partial inventories, information on gross flows, a generalized understanding of processes
(extraction, transformations, emissions, leakages), and limited information on final use and the ultimate
fate of materials.
3. Frameworks
There are two frameworks that are used to construct and understand material flows at the national
level. One is the national mass balance approach that looks at the physical inputs to the economy and the
physical outputs from the economy (e.g., Figure 2). This is the approach used in the recent publication of
The Weight of Nations: Material Outflows from Industrial Economies. (Matthews, et al., 2000) The second
is the creation of physical input-output tables that, in their final form, parallel the monetary input-output
tables of the system of national accounts. These track materials from industrial sector to industrial sector
with the additional dimension of flows to the environment. This is the approach used in the recent
publication Physical Input-Output Tables for Denmark (Pedersen, 1999). This approach also requires the
creation of mass-balances, but at a more detailed level than does the first. The choice of frameworks
depends on the purpose of the MFA activity. In the first framework the purpose was international
comparison and a comparison of trends in material use efficiency. In the second, the purpose was to create
46 . Paper prepared by Mr. Eric Rodenburg, Minerals and Materials Analysis Section, U.S. Geological Survey, U.S. Department of the Interior.
79
a component of a system of “green” national accounts that does not assign monetary value to resource
depletion, and pollution.
Figure 13. Figure 1. A materials flow view of an industrial economy (Matthews, et al., 2000, p.7)
Foreign Add. Air Water

Hidden and Water Vapor
Flows
Imports Exports
ECONOMIC
DMI PROCESSING Domestic
TMR Processed
Domestic
Outputs DPO
Extraction
(to Air, Land,
and Water)
TDO
STOCKS
Domestic Domestic
DOMESTIC ENVIRONMENT
DMI (Direct Material Input) = Domestic Extraction + Imports

TMR (Total Material Requirement) = DMI + Domestic Hidden Flows + Fo reign Hidden Flows
DPO (Domestic Processed Output) = DMI – Net Additions to Stock – Exports
TDO (Total Domestic Output) = DPO + Domestic Hidden Flows
NAS (Net Adddition to Stock) = DMI - DPO - Exports
Figure 14. Figure 2. The material flow balance of Germany (Matthews, et al., 2000, p.37)
1996 - Million tons
Material Economy Material

Input Output
Imports 475 Additional

stock
938
Abiotic Exports 228
raw materials
3443 Waste disposal
(excl.
used: incineration)
- minerals 898 2329
- energy - controlled
carriers 253 waste disposal
unused: 119
- non saleable - landfill and
extraction mine dumping
1996 2210
- excavation *
296 Emissions **
to air 1005
* - CO2 990
** - NO2,SO2,CO,
and others 15
Air 1080 Emissions of
water from
materials 639 ***
* Biotic raw * Erosion 126
materials ** Dissipative use
(fresh weight) 225 of products and
** Erosion 126 dissipative losses 47
*** Emissions
to water 37
TOTAL INPUT 5348 TOTAL OUTPUT 4411
Source: Wuppertal Institute, 12/99
80
4. Boundary Conditions
MFA requires an explicit definition of the conceptual boundaries of the system under study.
National borders make up one such arbitrary, yet sensible set of boundaries. But this is equally necessary
when developing MFA at the level of the firm, or a region, or a household.
Figure 15. Table 1. Categories of boundary decisions

Economic– national, firm, household
Geographic – national, region, international
Temporal – instantaneous, time series
Material – exclusions
In national MFA using input-output tables, the boundary is set narrowly to within the monetary
economy and within national borders. In Pedersen (2000), only materials that are priced are included. In
order to maintain parallelism with the monetary input-output tables, physical input-output tables represent
the transactions in materials that are priced that occur between industrial sectors.
Figure 16. Figure 3: Conceptual model of physical input-output tables with boundaries
(Wuppertal Institute for Climate, Environment, Energy, 1998)
In the national level material balance MFA (Matthews, 2000), the boundary is also set at the
national border, but includes flows that are created ancillary to the production of commodities or the
creation of infrastructure (hidden flows or environmental rucksacks) and sets as its economic boundary
those materials that are priced and enter the economy and those that exit the economy. The distinction
between hidden flows and commodity flows is crucial. Hidden flows are not priced, but are large in
81
magnitude. They include such materials as overburden, eroded soil, and tailings. Commodities are counted
as materials at their first monetary transaction. This definition means some overlap occurs. Tailing piles at
copper mines are derived from a non-priced ore and so are defined as hidden flows. The functional
equivalent of tailings (red mud) from the production of alumina from bauxite ore can occur either before or
after the sale of bauxite and so could be considered a hidden flow or a process waste.
Industrial economies are characterized by international trade and so raw materials, semi-
manufactures, and finished products are imported. Hidden flows associated with these imports, while
occurring in other countries, can also be included in a material balance MFA if the total burden – foreign
and domestic – of material use on the planet is an important purpose of the study. Resource Flows: The
Material Basis of Industrial Economies (Adriaanse, et al., 1997) explicitly includes these foreign hidden
flows to allocate responsibility for offshore environmental effects of material use.
Figure 17. Figure 4: Foreign and domestic hidden flows in four industrialized countries, 1994.
(Adriaanse, et al., 1997)
100
Domestic
80
Foreign
M etric tons per capita
60
40
20
0
Germany Japan Netherlands USA
Not all material flows need be included in MFA. Water is often excluded or is counted only as
water embodied in another material flow. The flow and use of water, for cooling, power generation, and
waste disposal is often well documented, but is so large, that it would overwhelm the flows of other
materials. Air moved for ventilation, is similarly large in quantity and relatively benign and is usually
excluded from MFA. The products of combustion (carbon dioxide, water vapor, carbon monoxide, sulfur
dioxide, and nitrogen oxides), on the other hand, include oxygen and sometimes nitrogen derived from the
air. These oxides are sometimes counted with the embodied oxygen (Matthews, et al., 2000) and
sometimes without (Adriaanse, et al., 1997). The choice depends on the purpose of the MFA and is
somewhat arbitrary.
MFA accounting can focus on a single year or be given a time dimension. A set of input-output
tables, by definition, looks only to a single base year. In contrast, material flows are shown in a twenty-two
year time series in the five countries in Matthews, et al. (2000). In the U.S. data in that study, outflows to
the environment are instantaneous, that is, the outflows are reported in the year of production. The other
four countries report actual outflows.
5. Approaches
There are several basic approaches to constructing national MFA systems. The Physical Input-
Output Tables for Denmark (Pedersen, 1999) is based upon the construction of material balances for 2940
82
different commodities associated with the goods and services found within the Danish economy. These
were distributed across industries according to the relative share of transactions found in the monetary
input-output tables and aggregated to a 27 X 27 industry input-output matrix. Residuals, that is, flows to
the environment, are characterized by mass only and not by material.
With the exception of the data for the United States, the data shown in Weight of Nations:
Material Outflows from Industrial Economies (Matthews, et al., 2000) was based, in general, on estimates
of inputs and separate estimates of output (emissions, wastes, and other flows to the environment) using
relatively good governmental statistics and extrapolations from special studies. Of special importance were
studies that defined emission factors associated with the processing of materials.
The United States data shown in Weight of Nations: Material Outflows from Industrial
Economies (Matthews, et al., 2000) were based on calculations of 460 individual material flows – in part
due to limited data on waste flows compared to the European countries in the same report. Production,
imports, and exports combined to give inputs to the economy. Estimates of extraction and process losses,
emissions, and ancillary flows were calculated. Information on the final use of products from these flows
was used to allocated them to the environment. Each flow to the environment of any kind was
characterized by one of seven modes of first release, (e.g., M1. Flows contained or controlled on land as
solids), five quality categories (e.g., Q3. Flows that have not been chemically processed but are chemically
active [salt], or biologically hazardous [asbestos]), and three categories of velocity through the economy
(e.g., V3. Flows that stay in the economy for more than 30 years). This allows a first order
characterization of potential effects of these flows on the environment.
6. Estimation
In the best of all worlds, governments would collect all the data required to characterize the
physical economy. But, this is Earth and here investigators must deal with data gaps, including
inappropriately aggregated data, missing data, aged data, and time series data that abruptly end.
Investigators must use a variety of means to estimate flows where such constraints exist. Good work on
MFA requires that such means be explicitly documented for analysts and policy makers.
Where available, existing governmental datasets provide the starting point for MFA. In general,
good data are available for commodities entering the economy through production or importation. But, this
is not always the case. Especially for imports, data are usually available for the value of products entering a
country in aggregated categories, but disaggregation of those flows and conversion of value to mass are
sometimes problematical.
Special studies can be used to supplement production data. For example, studies of coal
chemistry can be used to characterize emissions to bottom ash, fly ash, and the atmosphere from coal
combustion. In The Weight of Nations: Material Outflows from Industrial Economies (Matthews, et al.,
2000) participating institutions sometimes used emissions coefficients developed by the others to estimate
flows by analogy. Not all imports are consumed in country, some are transshipped (especially in smaller
trading countries, e.g. the Netherlands) some are used as materials that are then exported. These special
cases require special handling. Academic research such as Accounting for Resources, 1: Economy-wide
Applications of Mass-Balance Principles to Materials and Waste (Ayers and Ayers, 1998) provide
essential information on process losses and the universe of materials that flow through processes. Expert
opinion is another source of information on otherwise intractable information problems, especially related
to the efficiency of processes, the life of materials, and the ultimate fate of products. The filling of gaps
through extrapolation is a universal strategy and one that should be documented.
Detailed review of the data, assumptions, and sources is required to ensure that MFA provides
the best data possible. At the national level, MFA statistics should provide a view of the economy that is
correct in the magnitude and nature of flows.
83
7. Indicators
The purpose of material flow accounting is to inform, not simply to amass statistics. Data are not
information. The statistics generated by MFA must be aggregated in a way that tells the story of the
material economy while maintaining accuracy and transparency. The monetary accounts of a nation use
indicators in this way. Gross domestic product (GDP) is a measure of the monetary economy and how it
changes over time that has meaning to the public. Analogous indicators must be constructed for the
environment. Both The Weight of Nations: Material Outflows from Industrial Economies (Matthews, et al.,
2000) and Resource Flows: The Material Basis of Industrial Economies (Adriaanse, et al., 1997) provide
such indicators – Total Material Requirement (TMO) and Total Domestic Output (TDO) – that measure,
respectively, the dependence of economies on material flows and the burden on the environment from
those flows.
But, just as GDP is made up of an underlying database that is also used to construct a host of
economic indicators, the database created under MFA can be used to generate a number of other relevant
indicators that describe the economy/environment interaction. Material that flows out to the environment
can, for example, be aggregated by material specific carcinogenicity factors to suggest the potential cancer
burden. Greenhouse gas emissions, ozone depletion, GDP, aquatic toxicity, are other characterizations can
be applied to the data to create policy relevant indicators.
8. Conclusion
MFA provides a tool to understand the economy in a new and policy-relevant manner. It is
applicable at a variety of scales, but at the national level it provides measures of the burden of national
economic activity on the planet. Based on the mass-balance approach, MFA accounts for everything that is
mobilized from or delivered to the environment. Until now, governments have not tended to collect
comprehensive data that would allow investigators to compile MFA in detail and analyze the data to
manage the impact of material use on the environment. But, MFA has been developed to where countries
can look at the changes in their physical economies over time and compare their experience with those of
other countries. Indicators of these changes and comparisons provide previously unavailable information to
the public and policy makers. MFA provides a powerful and necessary complement to the monetary
system of national accounts.
REFERENCES
Adriaanse, Albert, Stefan Bringezu, Allen Hammond, Yuichi Moriguchi, Eric Rodenburg, Donald Rogich,
Helmut Schutz, 1997, Resource flows – the material basis of industrial economies: World Resources
Institute, Washington, D.C. 66 p.
Ayres, Robert U., Leslie W. Ayres, 1998, Accounting for resources, 1: economy-wide applications of
mass-balance principles to materials and waste: Edward Elgar, Cheltenham, U.K, 245 p.
Matthews, Emily, Christof Amann, Stefan Bringezu, Marina Fischer-Kowalski, Walter Huttler, Rene
Kleijn, Yuichi Moriguchi, Christian Ottke, Eric Rodenburg, Don Rogich, Heinz Schandl, Helmut
Schutz, Ester Van Der Voet, Helga Weisz, 2000, The weight of nations – material outflows from
industrial economies: World Resources Institute, Washington, D.C.,125 p.
Pedersen, Ole Gravgard, 1999, Physical input-output tables for Denmark: Statistics Denmark, Copenhagen,
189 p.
Wuppertal Institute for Climate, Environment, Energy, 1998, Modeling a socially and environmentally
sustainable Europe: Wuppertal Institute for Climate, Environment, Energy, Wuppertal, Germany,
http://www2.wupperinst.org/Projekte/SuE/.
84
PART III. LINKS WITH INDICATORS AND OTHER ACCOUNTING TOOLS
85
LINKS BETWEEN THE MICRO AND THE MACRO FLOWS: SUBSTANCE FLOW ANALYSIS:
Presentation by the Netherlands, Leiden University47
1. Introduction
Many of the environmental problems we are facing today are a result of society’s processing of
materials. One of the challenges of economics is to relate the generation of waste and emissions to societal
developments. This can be approached from many sides. One possibility is to look at the societal system in
environmental terms. This approach is taken by Ayres (e.g. Ayres 1989; Ayres et al.1989; Ayres and
Simonis 1994) in his introduction of the concept of industrial metabolism. This concept argues the analogy
between the economy and environment on a material level: the economy's "metabolism" in terms of
materials mobilisation, use and excretion to create "technomass" is compared to the use of materials in the
biosphere to create biomass. Whereas the biosphere has had billions of years to evolve and attune its
processes to such a state that waste generated in one process is converted into a resource for another, the
economy is still in its early stages of wastefulness. In order to speed up 'economic evolution', society must
look to the biosphere for guiding principles. The description of the economy thus is limited to a description
of the physical economy. The research area of industrial ecology (e.g. Jelinski et al., 1992; Socolow et al.,
1994; Graedel and Allenby, 1995; Ayres and Ayres, 1996; Allenby, 1999) occupies itself among others
with elaborating and operationalizing this concept, and takes the physical economy as its primary subject.
An important research principle in this field is the materials balance, (e.g. Kneese et al., 1970;
Ayres and Ayres, 1996) a tool for describing the materials regime of the economy based on the Law of
Mass Preservation, analogous to the long-standing practice of investigating ecological materials cycles.
Analytical tools based on the materials balance are Materials Flow Analysis (MFA) and Substance Flow
Analysis (SFA). These tools are useful for supporting environmental policy (Bringezu et al., 1997): they
enable policy makers to trace the origins of resource depletion and pollution problems and to evaluate the
appropriateness of the societal management of materials and substances.
2. Material Flow Accounting (MFA)
MFA is a widely used technique within environmental science. Studying the material basis of the
world’s economies can generate valuable insights in the process of formation of environmental hazardous
emissions and waste flows and of depletion of natural resources. MFA studies vary in complexity, ranging
from a “simple” materials bookkeeping for a certain year to complex dynamic analyses calculating future
flows and stocks. Next to that, the aim of the MFA can be very diverse: studies focussing on the
metabolism of industrial systems like the metals industry the chlorine industry, pioneered by Robert Ayres,
studies focussing on the total or bulk material flows within modern economies (brought into the political
arena mainly by the Wuppertal Institute), and studies focussing on the flows of one specific (group of)
substance(s) within economy and environment (mainly in Scandinavia, Austria, Switzerland and the
Netherlands). However diverse these studies and their aims may be, the techniques which are used within
the analyses are similar. The fact that, in the end, inflow equals outflow is the simple central argument in
all studies.
47 . Paper prepared by Mr. Rene Klein and Ms. Ester van der Voet, Centre of Environmental Science (CML), Leiden University, Netherlands.
87
3. Bulk-MFA
The type of MFA which is also called bulk-MFA ,is characterised by the fact that all material
flows are accounted for on a mass basis regardless of their nature. The result of such studies normally
consists of a list of inflows and outflows for the studied country, region or city. The material flows are then
aggregated on a mass basis into single indicators like DMI and DPO (excluding hidden flows) and TMR
and TDO (including hidden flows). These indicators are often used as general indicators for societal
metabolism and compared with economic and social indicators like GDP and population.
Material flows generated by human societies are at the heart of many of the most important
environmental problems. The material metabolism of societies should thus be an important object of study.
In order to get insight in the functioning of the metabolism of societies a complete quantitative overview of
flows of materials entering, leaving and accumulating within the system is needed. A bulk-MFA provides
such information on the metabolism of societies. The accounts are set up in a systematic manner which
make the results comparable between different countries and different moments in time.
The desegregated information which is being collected in bulk-MFA studies provide useful
insights in the material basis of human societies. In industrial societies the dominance of water and fossil
fuels is vividly illustrated for example. Flows of drinking water dwarf all other material flows which
emphasises the excessive need for clean fresh water in those societies. The next largest type of flow are
those connected to the use of fossil fuels which indicates our dependence on unsustainable sources of
energy.
An important strength of bulk-MFA is the fact that the results can be presented in the form of one
aggregated indicator. Because of this the methodology provides a macro-indicator which can be compared
with the main macro economic indicator: GDP. Decoupling of economic growth and material flows can
thus easily be analysed. However, like GDP, completely aggregated indicators like DMI and DPO clearly
have their limitations. A substitution of material intensive processes by processes with emissions of small
quantities of toxic compounds for example will result in a decrease in bulk-MFA indicators while actual
environmental impacts will most likely increase. Thus bulk-MFA indicators are not always good indicators
for environmental pressure since the differences in environmental impacts of the materials are not taken
into account. If the focus is on specific environmental problems linked to specific (groups of) substances
another type of MFA is needed: Substance Flow Analysis (SFA).
4. Substance Flow Analysis (SFA)
SFA aims to provide relevant information for an overall management strategy with regard to one
specific substance or a limited group of substances (van der Voet, 1996). In order to do this, a quantified
relationship between the economy and the environment of a geographically demarcated system is
established by quantifying the pathways of a substance or group of substances in, out and through that
system. In SFA the economy thus is viewed only in terms of flows of a specific (group of) substance(s).
The methodology is similar to bulk-MFA but the applications may be different. Mass and bulk flow studies
provide macro economic indicators (von Weizsäcker et al. 1997; Adriaanse et al. 1997), while studies on
flows of substances can be related to specific environmental problems and thus provide input for a
pollutants policy. An example of the aggregated results of an SFA is given in Figure 1.
In general terms, substance flow studies comprise the following three step procedure (van der
Voet et al. 1995b): (1) definition of the system, (2) quantification of the overview of stocks and flows, and
(3) interpretation of the results. All three steps involve a variety of choices and specifications, each of
which depends on the specific goal of the study to be conducted, as will be argued below. The first step in
88
any materials flow study is to define the system. The system must be determined with regard to space,
function, time and materials. If necessary, the system can be divided into subsystems. The various
categories of processes, stocks and flows belonging to the system must be specified. Finally, this results in
a flow chart: the specification of the network of nodes. To define the SFA system, a number of choices
must be made with regard to the following aspects: (1) spatial demarcation, (2) functional demarcation, (3)
time horizon, and (4) materials to be studied.
4.1 SFA modelling
The quantification of the network is the next step. This involves identifying and collecting the
relevant data on the one hand, and modelling on the other. Three possible ways of modelling the system
are briefly discussed here, all three types having their own data requirements, as well as their own potential
for policy support: (1) accounting or bookkeeping, (2) static modelling, (3) dynamic modelling
Accounting
The first way to ‘model’ the system is to treat it as an accounting system. The input for such a
system consists of data regarding the size of the system's flows and stocks of goods and materials, that can
be obtained from trade and production statistics, and if necessary also data regarding the content of specific
substances in those goods and materials. Emissions and environmental flux or concentration monitoring
can be used for the environmental flows. A combination of those data together with application of the mass
balancing principle then must lead to the desired overview of flows and stocks.
The accounting overview may also serve as an identification system for missing or inaccurate
data. Missing amounts can be estimated by applying the mass balance principle. In this way, inflows and
outflows are balanced for every node as well as for the system as a whole, unless accumulation within the
system can be proven. This technique is most commonly used in materials flow studies, and can be viewed
as a form of descriptive statistics (e.g. Ayres et al. 1988; Olsthoorn 1993; Fleckseder 1992; Palm and
Östlund 1996; Tukker et al. 1996, 1997; Kleijn et al. 1997; Hansen and Lassen 2000). There are, however,
some examples of case studies that specifically address societal stocks (Bergbäck and Lohm 1997;
Bergbäck, Johansson and Molander 2000) and use these as an indicator for possible environmental
problems in the future.
Static modelling
In the case of static modelling, the process network is translated into a set of linear equations
describing the flows and accumulations as dependent on one another. Emission factors and distribution
factors over the various outputs for the economic processes and partition coefficients for the environmental
compartments can be used as such variables. A limited amount of accounting data is required as well for a
solution of the set of equations, but the modelling outcome is determined largely by the distribution
pattern. The description of the system as such a matrix equation opens possibilities for various types of
analysis: the existence of solutions, the solution space, and the robustness of the solution can be studied by
means of standard algebraic techniques, as is shown by Bauer et al. (1997) and by Heijungs (1994, 1997)
for the related product Life Cycle Assessment.
Static modelling can be extended by including a so-called origin-analysis in which the origins of
one specific problematical flow can be traced at several levels (van der Voet et al. 1995c; Gleiss et al.
1998). Three levels may be distinguished:
89
− direct causes, derived directly from the nodes balance (for example, one of the direct causes
of the cadmium soil load is atmospheric deposition);
− the economic sectors, or environmental policy target groups, directly responsible for the
problem, identified by following the path back from node to node to the point of emission
(for example, waste incineration is one of the economic sectors responsible for the cadmium
soil load);
− ultimate origins, found by following the path back to the system boundaries (for example, the
import of zinc ore is one of the ultimate origins of the cadmium soil load).
Furthermore the effectiveness of abatement measures can be assessed with static modelling.
The result of steady state modelling may be regarded as a caricature of the present management
regime. It is therefore most suitable for comparisons between management regimes. Static and steady state
models have been proposed by Anderberg et al. (1993) and applied for example by Schrøder (1994),
Baccini and Bader (1996), Boelens and Olsthoorn (1998), van der Voet et al. (2000) and also, for purely
environmental flows, by Jager and Visser (1994).
Dynamic modelling:
The main difference between static and dynamic SFA models lies in the inclusion of stocks in
society (Ford 1999): substances accumulated in stocks of materials and products in households or in the
built environment. Until recently, MFA has concentrated mostly on flows. During the past few years, MFA
researchers have realised that stocks may be equally or sometimes even more important. One of the
environmental issues where stocks play an important role in SFA is in the prediction of future emissions
and waste flows of products with a long life span. In some way or another information on the societal
stocks of PVC is needed to supply policy makers with information about future outflows: today’s stocks
are tomorrow’s emissions and waste flows. In some studies the magnitude of the anthropospheric stocks
has even been the primary focus of the study (e.g. Bergbäck & Lohm 1997 and Patel, 1997). These
exercises have established the importance of considering stocks. Large stocks have accumulated in the
societal system which must be dealt with in some way or other (Brunner & Baccini, 1992; Obernosterer et
al., 1998). Future CFC emissions from present (1998) stocks for example are estimated, even assuming a
world-wide successful implementation of the Montreal protocol, to roughly equal 75% the total added past
emissions (Kleijn & Van der Voet, 1998). Other studies that are dedicated to the analysis of accumulated
stocks of metals and other persistent toxics in the societal system are: Gilbert and Feenstra 1992; Bergbäck
and Lohm 1997; Baccini and Bader 1996; Fraanje and Verkuijlen 1996; also Lohm et al. 1997; Kleijn,
Huele and van der Voet 2000. Such build-ups can serve as an ‘early warning’ signal for future emissions:
one day, the stocks may become obsolete or recognisably dangerous (as has happened with asbestos,
CFCs, PCBs and mercury in chlor-alkali cells). Then the stocks may be discarded and end up as waste and
emissions. In some cases, this delay between inflow and outflow can be very long indeed. Bergbäck and
Lohm (1998) also draw attention to stocks of products no longer in use, but not discarded yet: old radios or
computers in basements or attics, out-of-use pipes still in the soils, old stocks of chemicals no longer
produced such as lead paint or pesticides. They conclude that such “hibernating stocks” could be very
large. In order to estimate future emissions, which is a crucial issue if environmental policy makers are to
anticipate problems and take timely action, it appears that such stocks cannot be ignored.
When using MFA or SFA models for forecasting, stocks therefore should be a vital part. Flows
and stocks interact with each other: stocks grow when the inflows exceed the outflows of a (sub)system
and certain outflows of a (sub)system are proportional to the stocks.
90
For this dynamic model, additional information is needed with regard to the time dimension of
the variables: the life span of applications in the economy, the half life of compounds, the retention time in
environmental compartments and so forth.
Calculations can be made not only on the ‘intrinsic’ effectiveness of packages of measures, but
also on their anticipated effects in a specific year in the future, and on the time it takes for such measures to
become effective. A dynamic model is therefore most suitable for scenario analysis, provided that the
required data are available or can be estimated with adequate accuracy
4.2 SFA Indicators
The use of indicators can be great help in the interpretation of the results. SFA indicators can be
selected from the overview of flows and stocks, by singling out a specific flow or stock as the relevant one
to follow, or they can be calculated directly from the overview. Indicators may be defined for
environmental flows and/or stocks, as an addition to the numerous environmental quality indicators already
existing. Other possibilities are indicators for economic substance flows, or indicators for integrated chain
management, which bear on (possible, future) losses from the economy to the environment; i.e. 'leaks' out
of the economic cycle. Examples include materials intensity, economic throughput, the technical or energy
efficiency of groups of processes, secondary vs. primary materials use and so on (Ayres 1997a). Another
possibility is to compare economic mobilisation of a certain substance with natural mobilisation; as a
measure of potential risk (Huele, Kleijn and van der Voet 1993). This goes in the direction of the study of
biogeochemical cycles and their transformation by man's activity into anthropo-biogeochemical cycles.
Indicators should be designed to provide information of relevance for an integrated substance

chain management policy, for example regarding: (1) the existence and causes of environmental problems
related to the substance; (2) the management of the substance chain or cycle in society; (3) early
recognition of future problems and (4) the influence of policy measures, including both their effectiveness
and various types of problem shifting. In addition, requirements can be defined for the indicators as a
group, which must be suitable for evaluating an SFA overview for a specific year but also for evaluating
changes in flows and stocks over time as well as alterations thereof, as induced by environmental policy.
Therefore, a comparison between different regimes must also be possible. See also Guinée et al.(1999) and
Moolenaar et al. (1997) for agricultural soils and systems.
4.3 Studying substance flows in relation to each other
One of the most important aspects of SFA is that the flows of substances are studied in relation
with each other. An example of this comes from the study of the flows of chlorine and its compounds
through society. Chlorine flows within industrial systems are very much interlinked. A large number of
processes within the chlorine industry, including the incineration of chlorinated waste flows, produce
hydrochloric acid as a by-product. This hydrochloric acid is often used within the production of PVC,
which by some will be regarded as a great example of industrial ecology while others will think of it as just
another argument to ban PVC. Because of this link between the production of PVC and the rest of the
chlorine chain a ban on PVC will have an enormous impact on the whole chlorine industry: one of it’s
most important by-products will run into a ”dead-end” and become a major waste flow which would have
to lead to a major restructuring of the chlorine industry.
Another example can be found within the realm of metal flows. The use of cadmium in diffuse
applications can lead to critical levels in the environment. Emission prevention, and maybe even a ban on
certain cadmium containing products therefore seem rational actions for policy makers. However, MFAs
have led to the insight that emission prevention and even product bans will not solve the problem. Since
91
inflow always equals outflow the only way to reduce the outflow to the environment is to reduce the inflow
to society. Cadmium however, is produced not on demand but as a by-product of the zinc industry. Like
when one tries to close the exit of a colony of ants, immediately after closing one exit, another one will be
created. The only real solution to this problem is thus to either reduce the amount of zinc being produced
or to immobilise the cadmium directly after production. Some have already argued that the production of
statues would be a good option.
5. Links between bulk-MFA en SFA
As discussed above there are important similarities between bulk-MFA and SFA. The type of
accounting and the mathematical techniques which are used are almost identical. The object of study
however is different and therefor the applications are different. Where the outcomes of bulk-MFA can be
used single indicators of the material basis of societies, SFA can be used to trace back and analyse the
substance flows which are related to specific environmental problems. In our view however bulk-MFA and
SFA strengthen each other when the are used in combination. The following Links between the two can be
thought of:
− with the aid of e.g. mass fractions of substances in products or materials the data generated by
the one can be used as an input for the other;
− bulk-MFA is usually focussed on the inflows and outflows of the system while in SFA the
focus is on the internal flows in society and the relations between those flows. Physical
Input/Output tables may function here as a traité d'union;
− one of the most common criticisms on bulk-MFA is that the link between the on mass base
aggregated flows of materials and environmental problems is very indirect. This problem is at
least partly solved when bulk-MFA is combined with a number of SFAs quantified in kg
relevant substance or in environmentally weighted kg;
− in SFA different types of modelling are used to evaluate policy measures and to predict future
flows and stocks. The modelling that is used in SFA could be introduced in bulk-MFA;
− SFA and bulk-MFA get very close when flows of bulk materials are studied such as water,
steel or paper;
Since a number of important links exist between bulk-MFA and SFA, both tools can benefit from
the combining the two. Also in a policy context the use of both tools together will broaden the scope of
their possible applications.
92
Figure 18. Figure 19. Example of an overview of the results of an SFA-study for cadmium
234
CADMIUM
1990
665907
mining and
extrac tion 15090
12032
5581
633204
4331
FOREIGN COUNTRIES
209837 3810
ENVIRONMENT
production
7560
5000
72881
5503
156
144648
15649
usage
4911
101383
64853
3240
482
waste
treatment 13070
45229
600
93
REFERENCES
Adriaanse, Albert, Stefan Bringezu, Allen Hammond, Yuichi Moriguchi, Eric Rodenburg, Donald Rogich
and Helmut Schütz (1997), Resource Flows: The Material Basis of Industrial Economies,
Washington DC: World Resources Institute.
Allenby, B.R., 1999. Industrial Ecology, Policy Framework and Implementation. Prentice Hall, New
Jersey.
Anderberg, Stefan, Gerd Bauer, Yuri Ermoliev and William M. Stigliani (1993), Mathematical Tools for
Studies of Industrial Metabolism, Laxenburg Austria: International Institute of
Applied Systems Analysis (IIASA) Working Paper WP-93-9, 28 pp.
Ayres, R.U., Ayres, L.W., 1996. Industrial Ecology: Towards Closing the Materials Cycle. Edward Elgar
Publishers, 379 pp.
Ayres, R.U., Norberg-Bohm, V., Prince, J., Stigliani, W.M. and Yanowitz, J., 1989. Industrial
Metabolism, the Environment, and Application of Materials-Balance Principles for Selected
Chemicals. IIASA report RR-89-11, Laxenburg, Austria, vi + 118 pp.
Ayres, R.U., 1989. Industrial Metabolism. In: Ausubel, J., Sladovich, H., Technology and Environment,
National Academy Press, Washington DC.
Ayres, R.U., Simonis, U.E. (eds.), 1994. Industrial Metabolism, Restructuring for Sustainable
Development. United Nations University Press, Tokyo, 376 pp.
Ayres, Robert U. (1997a),‘ Metals recycling: Economic and environmental implications’, Resource
Conservation and Recycling, 21, 145-173.
Ayres, Robert U., Leslie W. Ayres, J. McCurley, M. Small, Joel A. Tarr and Rolande C. Widgery (1988),
An Historical Reconstruction of Major Pollutant Levels in the Hudson-Raritan Basin 1800–1980,
Technical Memorandum (NOS OMA 43), Rockville, MD: National Oceanic and Atmospheric
Administration.
Baccini, P. and Bader, H.-P., 1996. Regionaler Stoffhaushalt. Spektrum Akademischer Verlag, Heidelberg,
Germany, 420 pp.
Bauer, Gerd, Manfred Deistler, Andreas Gleiß, Emmanuel Glenck and Thomas Matyus (1997),
‘Identification of Material Flow Systems’, Environmental Science and Pollution Research, 4(2),
105-112
Bergbäck, B. and Lohm, U., 1997. Metals in Society. In: Brune, Chapman, Gwynne and Pacyna (Editors),
The Global Environment. Scandinavian Science Publisher, Wiley - VCH.
Bergbäck, Bo and Ulrik Lohm (1997), ‘Metals in society’, in D. Brune and V. Chapman (eds), the Global
Environment - Science, Technology and Management, Oslo, Norway: Scandinavian Scientific Press,
pp. 276–289.
Bergbäck, Bo and Ulrik Lohm (1997), ‘Metals in society’, in D. Brune and V. Chapman (eds), the Global
Environment - Science, Technology and Management, Oslo, Norway: Scandinavian Scientific Press,
pp. 276–289.
Bergbäck, Bo, K. Johansson and Ulf Molander (2000 forthcoming), ‘Urban metal flows – a case study of
Stockholm; review and conclusions, water, air and soil pollution’, special issue Metals in the Urban
and Forest Environment, forthcoming.
Boelens, J. and Olsthoorn, A.A., 1998. Software for material flow analysis. In: Vellinga, P., Berkhout,
F.and Gupta J. (Editors), Sustainable Sustainability. Kluwer Publishers, Dordrecht, The Netherlands.
94
Bringezu S., Kleijn R., 1997. Short Review of the MFA work presented. In: Bringezu, S., Moll, S., Fisher
Kowalski, M., Kleijn, R., Palm, V. (Editors), Regional and National Material Flow Accounting,
“From Paradigm to Practice of Sustainability”. Proceedings of the 1st
ConAccount Workshop 21 - 23 January, at Leiden. Wuppertal Institute, Wuppertal, Germany: 306-308.
Brunner, P.H. and Baccini, P., 1992. Regional Materials Management and Environmental Protection.
Waste Management & Research 10: 203-212.
Fleckseder, Hellmut (1992), ‘A Nitrogen Balance for Austria’, Water Science and Technology,
26(7-8), 1789-1795
Ford, Andrew (1999), Modeling the Environment; An Introduction to System Dynamics, Washington DC:
Island Press.
Fraanje, Peter J. and Evert Verkuijlen (1996), Balansen van non-ferro metalen in de Nederlandse
woningbouw in 1990, Amsterdam: IVAM-onderzoeksreeks no. 76.
Gilbert, Allison J. and Johan F. Feenstra (1992), An Indicator of Sustainable Development – Diffusion of
Cadmium, Amsterdam: IvM-VU publ. R-92/06, pp vi + 66
Gleiß, Andreas, Thomas Matyus, Gerd Bauer, Manfred Deistler, Emmanuel Glenck and Christoph Lampert
(1998), ‘Identification of material flow systems - extensions and case study’, Environmental Science
and Pollution Research, 5(4), 245-253.
Graedel T.E., Allenby B.R., 1995. Industrial Ecology. Prentice Hall, New Jersey, 412 pp.
Guinée, J.B., van den Bergh, J.C.J.M., Boelens, J., Fraanje, P.J., Huppes, G., Kandelaars,
Hansen, Erik and Carsten Lassen (2000 in press), Paradigm for Substance Flow Analyses - Guide for SFAs
, Report Prepared for the Danish EPA, Copenhagen: the Danish Environmental Protection Agency.
Heijungs, Reinout (1994), ‘A generic method for the identification of options for cleaner products’,
Ecological Economics, 10(2), 69-81
Heijungs, Reinout (1997), Economic Drama and the Environmental Stage, Formal Derivation of
Algorithmic Tools for Environmental Analysis and Decision-support from a Unified
Epistemological Principle, Leiden, The Netherlands: Centre of Environmental Science, Leiden
University, PhD thesis.
Huele, Ruben, René Kleijn and Ester van der Voet (1993), Natural Resource Accounting, the Search for a
Method, the Hague: Ministry of Environment Series Environmental Strategy 1993/3.
Jager, D.Tjalling and C.J.M. Visser (eds) (1994), Uniform System for the Evaluation of Substances
(USES), version 1.0, National Institute of Public Health and Environmental Protection (RIVM),
Ministry of Welfare, Health and Cultural Affairs (WVC), Ministry of Housing, Spatial Planning and
the Environment (VROM), Bilthoven, The Netherlands: RIVM Distribution No. 11144/150.
Jelinski, L.W., Graedel, T.E., Laudise, R.A., McCall, D.W., and Patel, C.K.N., 1992. Industrial Ecology:
Concepts and approaches. Proc. Natl. Acad. Sci. USA, vol. 89, pp 793-797. Proceedings of a
colloquium "Industrial Ecology" organised by C. Kumar and N. Patel, May 20-21 1991, at National
Academy of Sciences, Washington DC.
Kleijn, René and Ester van der Voet (1998), ‘Chlorine in Western Europe, a MacTempo case study, Annex
3’, in Paul H. Brunner et al., Materials Accounting as a Tool for Decision Making in Environmental
Policy (Mac TEmPo), Final Report, Technical University of Vienna, Linköping University, ETH
Zürich and Leiden University.
Kleijn, René, Arnold Tukker and Ester van der Voet (1997), ‘Chlorine in the Netherlands Part I, An
Overview’, Journal of Industrial Ecology, 1(1), 95-116.
Kleijn, René, Ruben Huele and Ester van der Voet (2000), ‘Dynamic substance flow analysis: The
delaying mechanism of stocks, with the case of PVC in Sweden’, Ecological Economics, 32(2), 241-
254.
95
Kneese, A.V., Ayres, R.U., de Arge, R.C., 1970. Economics and the Environment. Resources for the
Future, Washington DC, 120pp.
Lohm, Ulrik, Bo Bergbäck, Johan Hedbrant, Arne Jonsson, John Svidén, Louise Sörme and Catarina
Östlund (1997), Databasen Stockhome, Flöden och ackumulation av metaller i Stockholms
teknosfär, Linköping, Sweden: Tema V Rapport 25, Linköping University.
Moolenaar, Simon W., Sjoerd E.A.T.M. van der Zee and Theo M. Lexmond (1997), ‘Indicators of the
sustainability of heavy metal management in agro-ecosystems’, The Science of the Total
Environment, 201(2), 155-169.
Obernosterer, R., Daxbeck, H., Gagan, T., Glenck, E., Hendriks, C., Morf, L., Paumann, R., Reiner, I. and
Brunner P.H., 1998. Urban metabolism - The City of Vienna; Materials Accounting as a Tool for
Decision Making in Environmental Policy. 4th European Commission Programme for Environment
and Climate. Institute for Water Quality and Waste Management, Vienna University of Technology,
Vienna, Austria, 129 pp.
Olsthoorn, Kees S.M. (1993), Stikstof en Fosfor in Nederland, 1990 (Nitrogen and Phosphorus in The
Netherlands, 1990), Voorburg: Kwartaalberichten Milieu 93/1.
P.P.A.A.H., Lexmond, Th.M., Moolenaar, S.W., Olsthoorn, A.A., Udo de Haes, H.A.,
Verkuijlen, E. and E. van der Voet (in press). Evaluation of metal flows and accumulation in economy and
environment. Ecological Economics, in press.
Palm, Viveka and Catarina Östlund (1996), ‘Lead and zinc flows from technosphere to biosphere in a city
region’, The Science of Total Environment, 192(1), 95-109.
Patel, M.K., 1997. Plastics in Germany - an analysis of production consumption and waste generation.
Draft paper Fraunhofer-Institute ISI, Karlsruhe, Germany.
Schrøder, Hans (1995), ‘Input management of nitrogen in agriculture’, Ecological Economics. 13(2), 125-
140.
Socolow, R., Andrews, C., Berkhout, F., Thomas, V. (eds.), 1994. Industrial Ecology and Global Change.
Cambridge University Press, Cambridge.
Tukker, A., Kleijn, R., van Oers, L. and Smeets, E.R.W., 1996. A PVC substance flow analysis for
Sweden. TNO-report STB/96/48 part I to IV, TNO, Apeldoorn, The Netherlands, 51, 81, 75 78 pp..
Voet, E. van der (1996). Substances from cradle to grave - Development of a methodology for the analysis
of substance flows through the economy and the environment of a region. Thesis, Leiden university,
Leiden, The Netherlands, 348 pp.
Voet, Ester van der, Jeroen B. Guinée and Helias A. Udo de Haes (eds) (2000), Heavy Metals: a Problem
Solved?, Dordrecht, The Netherlands, Boston, MA, USA and London, UK: Kluwer Academic
Publishers.
Voet, Ester van der, Reinout Heijungs, Paul Mulder, Ruben Huele, René Kleijn and Lauran van Oers
(1995c), ‘Studying substance flows through the economy and environment of a region - Part II:
Modeling’, Environmental Science and Pollution Research, 2(3), 137-144.
Voet, Ester van der, René Kleijn, Lauran van Oers, Reinout Heijungs, Ruben Huele and Paul Mulder
(1995b), ‘Studying substance flows through the economy and environment of a region - Part I:
systems definition’, Environmental Science and Pollution Research, 2(2), 89-96.
Weizsäcker, Ernst-Ulrich von, Amory B. Lovins and L. Hunter Lovins (1997), Factor Four: Doubling
Wealth – Halving Resource Use: the New Report to the Club of Rome, London: Earthscan.
96
LINKS BETWEEN MACRO AND MICRO FLOWS: THE SECTORAL APPROACH:

Presentation by Austria, IFF48
1. Introduction
Looking at developments regarding the theoretical and methodological approach to MFA and
empirical works undertaken within the last decade several preliminary conclusions can be drawn.
Firstly, the research community heavily concentrated on methodological agreements to foster
comparability. This is especially true for economy wide analysis. Nevertheless, this process resulted in two
different solutions. One being the international agreement reached within the attempt to harmonise
different approaches for economy wide MFA under the umbrella of the World Resources Institute (WRI).
There, principal solutions both for inputs into and outputs from a macroeconomic system were found
(Adriansee et al. 1997, Matthews et al. 2000).49, 50
A parallel effort has been undertaken at the German Statistical Office. Their work resulted in a
physical input output table (PIOT) for the German economy that differentiates between 71 activities
(Stahmer et al. 1998).51 The table, which has been established for the year 1990 is very much similar to the
methodological logic of economic input output tables (Leontief 1970).52 Also recently, a PIOT for
Denmark has been released (Pedersen 1999).53
Considering this two strands of research, we have to acknowledge that the process of
methodological stringency actually led to two answers.
What are the main differences of the two approaches? WRI, on the one hand, provides accurate
data for inputs and outputs in a time series approach for several countries. This data, which is rather easily
to produce, gives a first overview of the physical dimension of a national economy. Clearly, in this
approach, inputs and outputs stand rather loosely aside. Since the economic system is portrayed as a black
box, there is no information available for intra economic processes. Hence, this approach focuses on the
environmental performance at the boarder between society and the environment.
Opening up the black box economy is exactly the strength of the PIOT approach. The internal
economic activities are the focus of the research. On the basis of this tables, we can exactly know, how
materials are transformed within an economic system an hence discuss environmental problems related to
these activities. At the same time, the research effort to establish such a table seems to be a massive one. It
does not come as a surprise, that the most recent available data set is only for 1990. Therefore, we might
have to accept, that the huge amount of information stored in this tables will not be very relevant for
political decision making processes.
48. Paper prepared by Heinz Schandl and Helga Weisz, Institute for Interdisciplinary Studies of Austrian Universities (IFF), Vienna, Austria.
49. Adriaanse, A., S. Bringezu, A. hammond, Y. Moriguchi, E. Rodenburg, D. Rogich, and H. Schütz 1997. Resource Flows. The Material Basis
of Industrial Economies. Washington, DC: World Resources Institute.
50. Matthews, E., C. Amann, S. Bringezu, M. Fischer-Kowalski, W. Hüttler, R. Kleijn, Y. Moriguchi, C. Ottke, E. Rodenburg, D. Rogich, H.
Schandl, H. Schütz, E. Van der Voet, and H. Weisz. The Weight of Nations. Material Outflows from Industrial Economies. Washington, DC.
World Resources Institute.
51. Stahmer, C., M. Kuhn, and N. Braun 1998. Physical Input-Output Tables for Germany, 1990. Report prepared for DG XI and EUROSTAT.
Working paper No. 2/1998/B/1. Luxemburg: EUROSTAT.
52. Leontief, W. 1970. Environmental Repercussions and the Economic Structure. An Input-Output Approach, Review of Economics and
Statistics, Vol. 52, pp. 262-271.
53. Pedersen, O. G. 1999. Physical Input-Output Tables for Denmark. Products and Materials 1990. Air Emissions 1990-92. Kopenhagen:
Statistics Denmark.
97
We might think that these two approaches should only be linked to gain the advantages of both
approaches at the same time. This would be a promising strategy but is undermined by methodological
incompatibilities.
There are some differences in the two approaches on the very basic level of system definition.
The PIOT first and most important, applies the economic input/output framework to MFA by referring to
the law of conservation of mass as the physical equivalent to the basic equation (Leontief equation) of
monetary input output analysis, which considers of the economic system as a closed system with input
and output values necessarily to be equal at all levels of disaggregation. While this is quite compelling on a
conceptual level, it quickly becomes difficult at a more operational level. The application of the law of
conservation of mass to an input output framework, clearly demands additional conceptual decisions. In
fact, recognising the physical openness of the economic system being the most important. Hence, defining
the boundary between the economic system and the natural environment is probably the most difficult
problem to solve. The two approaches brought forward by PIOT and WRI diverge in their conclusions with
regard to two important aspects. PIOT’s solution to the economy/nature boundary is to treat the production
of plant biomass as a process taking place within the economic system. This clearly does not reflect an
SNA logic, but the way this is done, indeed does reflect a strict and consistent application of the input
output logic. WRI on the other hand considers agricultural crops as well as timber from managed forests as
elements of nature, which enter the economy as harvested plant biomass. This is certainly closer to the
SNA classification.
To put forward a second example, we might consider respiration of humans and livestock
animals treated both in PIOT and WRI. Whereas PIOT treats this feature in a systemic logic, WRI fails to
apply a consistent systems theory framework, and hence fails to provide comparable input and output data.
We suggest to take both criteria, SNA compatibility and a consequent systems logic, as essential and to
combine both within the conceptual framework of MFA. Still there will be some contradictions when
operationalizing both criteria, (and there will be even more when operationalizing other criteria, such as
policy relevance). However, there should be a methodological solution that represents an optimum
combination.
Beside methodological efforts, the question of policy relevance of a bulk flow approach was the
other major challenge within the discussion. Pretty soon indeed, this question was on the agenda. One
answer given did come quiet naturally. The MFA concept was a political concept as such from the very
beginning on, because it strongly questions the dominant approach which portrayed the environmental
problem as a problem of toxic outputs. Without doubt, MFA could successfully show that environmental
pressures stem from every step in the extraction - production – distribution - consumption chain. Being an
overall approach based on a consistent concept MFA should serve as a guideline to anticipate future
environmental problems that are connected to the huge amount of, at first glance, harmless materials. To
put it in a nutshell, the CO2 problem is a prominent example in this context.
The MFA approach could also claim to be the only integrated approach, both with respect to the
theoretical foundations, the methodological solutions and the arrangement of large sets of data. Differently
form traditional environmental indicators, MFA provides a framework which allows for answers to a
variety of questions, but still being able to integrate the answers into a common and consistent framework.
The main question still not sufficiently answered is the one concerning the political strategies. It
often has been asked, at which level of disaggregation policy intervention towards environmentally sound
economic behaviour could take place. Questions, such as if the picture provided by an economy wide MFA
is more than a retrospective indicator of political success, are very justified.
One prerequisite for political intervention, independently at which level it should take place, is
disaggregated information. Whether it is the local or regional level, the level of the companies, the
economic sectors, or even groups of materials where political efforts would be worth to be undertaken is
not excessively clear. But consequently, each level of interest would be reflected in a level of indicator
98
availability. Subsequently, this results in the discussion about the relation of the different levels. This has
often been addressed as a need to link the macro, meso, and micro level.
Though we perfectly know that a sectoral economic and environmental policy is not part of
current economic or environmental policy practice within OECD countries, we notwithstanding think that
the economic sector, representing an important level of integration and system differentiation should be an
important factor in this discussion.
2. Methodological preconditions for a sectoral approach

Above, we have argued that the process of development of MFA methods has actually resulted in
two different products, the WRI approach and the PIOT approach. Here, we will suggest that these two
strands could successfully be linked within an aggregated input/output framework. We will argue that
getting things ordered at this level would enable us to more successfully establish a sectoral account. This
might also be true for all other levels of disaggregation.
A guiding principal for structuring our discussion will be the principal of policy relevance. If we
accept that MFA first of all needs to be a political relevant concept several decisions are made at the same
time.
Firstly, data sets must be provided at a regular basis in order to provide indicators of sufficient
actuality. This can only be achieved, if we strictly follow a top-down approach. Doing indicators “quick
and dirty” might not be desirable for a scientist but certainly is for the political system.
Secondly, indicators must successfully reduce complexity. This can be fostered, if our decisions
of the system under observation is theoretical sound and plausible for all levels of disaggregation. What is
part of the economy/society and what is part of nature has to be decided on grounds of theoretical decision
and has to be true for the overall system as well as for its parts and components (for a detailed discussion
see Fischer-Kowalski 1997, Schandl and Schulz 2000).54, 55
Thirdly, it is definitely not enough to only understand the interrelationship between a system and
its natural and social environment.56 Policy making also needs a picture of internal systems dynamics. On
the other hand, it is not convincing to argue for an approach that delivers data with a delay of a decade.
Consequently, we have to open up the system by linking inputs to outputs within an overall understanding
of internal interlocking. To be able to do this, one should start with a rather superficial picture of internal
complexity, as provided by a three sector model of an economy. This model necessarily should be
established within an input/output framework (see Weisz et al. 1999).57
Having understood the systematic of this, we might think of a more complex sector model at least
including eight to ten sectors but not more than that. In such a model, we still can analyse data on a top-
down basis ensuring readiness and actuality. In a next step, we can think of the relation of a certain
economic sector to the overall structure. On the one hand, data for a sector can be organised in the same
way as data for the whole economy. On the other hand we can zoom out one sector from the overall table
and also link it back again.
Let us just briefly explain the logic of a rather simple input output model and mention where the
data to fill such a model might stem from.
54. Fischer-Kowalski, M. 1997. Society’s Metabolism. On the Childhood and Adolescence of a Rising Conceptual Star. The International
Handbook of Environmental Sociology. Redclift, M. and G. Woodgate. Cheltenham: Edward Elgar. Pp 119-137.
55. Schandl, H. and N. Schulz 2000. Using Material Flow Accounting to Operationalize the Concept of Society’s Metabolism. A Preliminary
MFA for the United Kingdom. ISER Working Papers 2000 (3). Colchester: University of Essex.
56. A macro-economic system interacts with its natural environment represented by the domestic nature and with other social units represented by
other macro-economies.
57. Weisz, H., H. Schandl, and M. Fischer-Kowalski 1999. OMEN – An Operating Matrix for Material Interactions Between the Economy and
Nature. How to make Material Balances Consistent. Kleijn, R., S. Bringezu, M. Fischer-Kowalski, and V. Palm, eds. Ecologizing Societal
Metabolism: Designing Scenarios for Sustainable Materials Management. CML report 148. Leiden: CML.
99
The aggregated physical Input Output Table (aggregated PIOT) as well as any sub-table consists
of three quadrants, the input quadrant (left below), the processing quadrant (left above) and the output
quadrant (right above). Additional rows or even whole quadrants, which allow to estimate additional items
of interest, such as unused extraction and indirect flows associated to imports (for terminology see
EUROSTAT 2000)58 might be provided. All input flows within the table are shown vertically along the
columns from the bottom up whereas all output flows are shown horizontally along the rows from left to
right.
The input quadrant contains all inputs into the system. We call these inputs primary inputs to
emphasise that they cross the border of the system to be balanced. For an aggregated PIOT calculation on
the economy-wide level we differentiate between four input categories. Domestic extraction of resources
(raw materials, water and air) and imports (inputs from other economies). The sum total of the primary
inputs is the direct material input (DMI).
The processing quadrant contains all material flows within the economic system. Rows and
columns are equally differentiated into three aggregated economic sectors, the primary production sector
(agriculture and mining), the industry sector (including construction) and the service sector (including,
besides other activities the public and private households). For these sectors the sectoral input has to equal
the sectoral output. Stock changes, independently where they occur, are treated as a separate sector. Here
inputs do not have to equal outputs.
Table 6. Table 7. The general structure of an aggregated PIOT table
Primary Services, stock Deliberate

Industry Export Emissions
production households changes disposals
domestic
Primary output
x11 x12 x13 x14 goods o11 o12 o13
production (Σo11 to o13)
(Σ x12+x13+x14)
domestic
output
Industry x21 x22 x23 x24 goods o21 o22 o23
(Σo21 to o23)
(Σ x21+x23+x24)
domestic
Services, output
x31 x32 x33 x34 goods o31 o32 o33
households (Σo31 to o33)
(Σ x31+x32+x34)
stock stock outputs output
x41 x42 x43 x44 o41 o42 o43
changes (Σ x11+x42+x43) (Σo41 to o43)
total proces-
secondary secondary secondary stock deliberate
sing matrix exports emissions direct
input input input inputs disposals
(Σx11+ x12+ .. (Σo11 to o41) (Σo12 to o42) output
(Σ21+x31+x41) (Σx12+x32+x42) (Σx13+x23+x43) (Σx14+x24+x34) (Σo13 to o43)
+x44)
domestic
Domestic
i11 i12 i13 i14 extraction
Extraction
(Σ i11 to i14)
water
Water i21 i22 i23 i24
(Σ i21 to i24)
Air i31 i32 i33 i34 air (Σ i31to i34)
imports
Imports i41 i42 i43 i44
(Σ i41to i44)
primary primary primary primary
input input input input direct input
(Σ i11 to i41) (Σ i12 to i42) (Σ i13 to i43) ( i14 to i44)
Source: Weisz et al. (1999)
58. EUROSTAT 2000. Economy-wide Material Flow Accounts and Balances with derived Resource Use Indicators. A Methodological Guide.
Draft for Public Review. Luxembourg: EUROSTAT.
100
The cross line in the processing quadrant contains intra-sectoral flows, referring to flows that are
transferred between actors within the same sector. Sectoral inputs which are transferred within the
economic system by other sectors of the same economy are defined as secondary inputs. These secondary
inputs make up the total secondary input of a sector (here, intra-sectoral flows are not included). Primary
inputs and secondary inputs make up the total input of a sector.
Analogous, output flows that go from one sector within the economy to other sectors of the same
economy can be aggregated to domestic goods produced in one sector.
The output quadrant contains all flows that leave the economic system to be balanced. Preferably,
here it should be distinguished between exports, emissions (to land, air and water) and dissipative use and
loss of materials (such as fertilisers, seeds, pesticides, but also abrasion and leakages, etc.). All flows in the
output quadrant can be summed up to make the direct output.59
For the total consistency check the calculation of the aggregated PIOT table has to fulfil the
equation:
direct input = direct output + stock inputs – stock outputs
For a sectoral consistency check, total sectoral input (i.e. primary input plus secondary input) has
to equal total sectoral output (i.e. domestic goods plus output into nature and into other economies). This
sectoral equation rule is true for the economic sectors within the OMEN table but is not be valid for stock
changes.
This kind of table directly represent a flow chart and vice versa. An aggregated economy-wide
material balance as presented here is still too complex to be calculated in one step. For this reason material
flows might be divided into five groups and the balances would be calculated separately for each group,
using sub-tables. Sub-tables clearly have to be structured in a way that allows for aggregation to an overall
table by summing up. The difficulty with this approach is, that both double counting’s must be avoided and
completeness must be achieved.60
It has to be stated, that the suggestion to consider material categories as a starting point for sub-
tables is not the only opportunity to take. Especially for a sectoral approach, a functional differentiation
along raw materials, semi-manufactured materials and final goods might be more promising. Nevertheless,
for raw materials the difference between biomass, mineral materials and energy carriers might make sense
again. Actually, this solution has been favoured by the input/output approach (Stahmer et al. 1998).61
The input/output approach as represented in the aggregated PIOT might also help to structure
available data sources. Data to fill the input quadrant stems from Agricultural, Forestry, Fishing and
Mining Statistics, and from Foreign Trade Statistics. Data to fill the intermediary quadrant can be obtained
from Industry and Trade Statistics and Monetary Input Output Tables. Finally, data to fill the output
quadrant is available in Foreign Trade Statistics and Waste Statistics, Emission Monitoring Statistics,
respectively.
59. Matthews et al. (2000) successfully tries to set a standard for indicator classification both for inputs and outputs. One major indicator for
outputs there, namely DPO (Domestic Processed Output) does not subsume exports. The argument established there is, that exports will
contribute to waste and emissions in other countries.
60. Of course double counting is a general problem of any MFA. The decision for sub-tables, however, adds additional traps for double counting.
61. A definition of sub-tables according to five major groups of input materials, namely water , air, biomass, fossil fuels and mineral materials as a
consequence introduces a substance logic into the structure of the aggregated PIOT, which generates some consistency problems. This can
easily be illustrated by considering the sub-table for fossil fuels.
The input quadrant contains the gateways from which fossil fuels enter the economy. These gateways are domestic extraction and imports.
Difficulties emerge if the materials are not raw materials but semi-manufactured materials or final goods. These groups of materials usually
represent a mix of different raw material categories and cannot easily be related to either one of the main categories (fossils, minerals or
biomass). On the input side this is the case for imports. One decision could be to integrate imported products according to their main
component into the different sub-tables. This would indeed help to avoid overlaps between different sub-tables, which would have to be
considered when summing up.
101
The aggregated PIOT approach mainly offers two prerequisites for a sectoral approach. It offers a
framework to define a specific economic system and to organise data accordingly. This enormously
supports consistency. On the basis of this systems approach every level of interest (from macro to micro
level) can be structured alike. Secondly, the aggregated PIOT is a first step to reduce the system
complexity on which the sectoral balance can build upon. Hence, we regard the aggregated PIOT as a
necessary intermediate step of investigation.
3. Problems witch mainly occur at the sectoral level

Economy wide MFA on principle relies on data periodically available at official statistical bodies
or international data compendia (such as the OECD or FAO data base). Similarly, sectoral MFA can be
established using the same data sources but additionally taking industry and trade statistics into account.
These additional statistics traditionally cover material inputs into a sector (input lists) and the production of
a sector (production lists). Preliminary estimates for wastes may be derived from environmental and waste
statistics, but unfortunately not always disaggregated for economic sectors. Once again, other data
compilations and frameworks, such as the NAMEA framework could provide useful data.
Data concerning sectoral origin and destination of commodities are provided in the conventional
Input/Output Statistics. These are usually provided periodically, but record monetary values only. Clearly,
the statistical sources mentioned here differ from one another with respect to data collection procedures,
terminological usage and levels of aggregation; this, in turn, causes methodological problems when it
comes to attributing data sets to the specific categories.
As said before, also a sectoral material balance accumulates data at an aggregated level, with at
least the following aggregates listed separately: On the input side, one should distinct quantities of water,
air, raw materials as well as semi-manufactured and auxiliary materials used in production.62 On the output
side, the balance should record goods produced (exports as well as domestically used goods), wastes and
emissions (differentiated by gateways).
Looking at the individual aggregates one cannot fail to notice that they normally differ with
respect to the extent to which they are documented in the official statistics.
The sum total of water used often is estimated by means of reference to the industry and trade
statistics. In order to map outputs as a function of inputs one would have to differentiate between water
used in production and water used as a cooling medium or for transport. The statistical sources available,
however, simply do not differentiate among these different forms of using water. No statistical data are
available for the quantity of air used.
All other materials are recorded in the input lists monitoring the use of raw and auxiliary
materials, as well as in foreign trade statistics. Final goods produced within a sector are registered in the
production lists of the industry and trade statistics.
Waste statistics or environmental statistics list the amount of waste and emissions produced for
several years. However, the quality of waste data is often weak, since it documents actual amounts of
waste but also waste potentials. It also leaves important quantities out. These different approaches cannot
easily be distinguished.
Several methodological problems seem to be characteristically for a sectoral approach, despite
some of these are also relevant at the economy-wide level. These problems are (a) conversion of data
reported in different units to metric units, (b) handling of secret or protected data, which more frequently
appear on a sectoral level, (c) handling of flows, which are not reported within statistical data bodies, (d)
intra-sectoral double counts, and (e) a high number of data of which big parts might be of minor relevance.
62. The definition of these aggregates largely follows the systematics in which data are presented in the statistical sources.
102
Not underestimating the seriousness of all of these problems, we will only discuss intra-sectoral
double counts as a specific problem of sectoral balances here. Industrial and trade statistics are computed
on the basis of company surveys. Empirical monetary and physical data gained in these surveys are
ultimately summed up to estimate total aggregates. In this case, each substance used by a firm is counted as
an input. For a sectoral material balance, however, only those materials are considered as inputs which are
bought from other sectors viz. other economies, or extracted from nature. From the point of view of a
consistent sectoral material balance, materials which are supplied from another firm within the same sector
– i.e. materials traded intra-sectoral – must not be considered inputs of the sector in question.
This implies that simple summation of the positions included in the lists of raw and auxiliary
materials leads to an overestimation of sectoral inputs. Hence, adjusting these lists for intra-sectoral double
counts is, in our view, one of the crucial methodological challenges when establishing a material balance
for an economic sector. That this challenge can successfully be met we have shown in a case study for the
chemical sector in Austria (Schandl and Zangerl Weisz 1997).63 We will not go into further detail here. It
should just be mentioned, that the methodological solution is a mathematical one, which combines all
available data (including monetary IOT) in a certain way.
4. Discussion
What we have suggested here is to structure the material dimension of economic processes within
a consistent aggregated Input Output framework. This framework is based on criteria such as theoretical
soundness, methodological clarity, policy relevance and feasibility.
This framework of an aggregated PIOT brings together the positive achievements of the currently
two most prominent approaches, represented in the two WRI reports and the German (and Danish) PIOT.
From the first it acknowledges the system definitions and the top-down approach as well as its feasibility.
From the latter it acknowledges the SNA compatibility and the attempt to open up the black box of the
economic system.
These advantages should be brought together to not only add a new feature to the economy-wide
approach but to also serve as a structural device for disaggregation at any level. The system definitions, the
difference between stocks and flows, the recognition of physical components of the economy/society, to
mention some of the relevant principles, guide the arrangement of MFA datasets, on the economy-wide,
the sectoral, the regional and also the company level.
Once empirical work at different levels keeps going in the same direction, results can be related
to one another.
It has often been argued, that consumption is a driving force within the economic system, maybe
even the dominant one. This understanding implicitly guides large parts of the efficiency and
dematerialization debate. A prominent example of such an understanding is the MIPS concept (Schmidt-
Bleek 1994).64 There, symbolic responsibility is attributed to the agents (the consumers), who, by buying a
good also decide to “buy” the material rucksack. The same approach can be found within life cycle
analysis (LCA) approaches, despite being useful political concepts, they overestimate the potential of the
consumer.
The aggregated PIOT approach and the related suggestion for sectoral investigations can be seen
as a descriptive method. The method mainly concentrates on the question, which amounts of material are
processed by an agent (be it an individual consumer or a company or an economic sector, etc.) and how
efficiently these materials are used.
63. Schandl, H. and H. Zangerl-Weisz 1997. Materialbilanz Chemie. Methodik sektoraler Materialbilanzen. Schriftenreihe Soziale Ökologie. Vol.
47. Wien: IFF.
64. Schmidt-Bleek, F. 1994. Revolution in Resource Productivity for a Sustainable Economy – a new Research Agenda. Fresenius Environmental
Bulletin 2. pp. 245-490.
103
Clearly, this focuses on certain steps in the production-consumption-chain, where different actors
occur and contribute to the environmental soundness of an economic system or expand environmental
pressures. Within this economic process, the monetary logic and the material logic is quite contradictory.
While materials gain value within this process (added value) they, at the same time, lose weight (“lost
weight” or “added waste”). The material analogue to value added of a company or economic sector is the
difference between the mass of material input and the mass of produced outputs, i.e. the amount of waste
and emissions.
Sectoral material balances can document this relation of input to output (as an material to
material efficiency indicator) but can also relate the material performance of a sector to its economic
performance. These results can be linked back to the economy-wide picture but also to the performance of
certain companies that are part of a sector. Benchmarking could only be one consequence of these
information systems. We certainly regard this information as highly relevant for policy decision making.
The systems approach discussed here does not only offer a comprehensive understanding how to
link the macro – meso and micro level. It furthermore fosters congruence with the economic tradition since
Leontief and with the principals of SNA. In the context of sustainability, where economic and ecological
understanding should be linked, this congruency is of major importance.
104
LINKS WITH NATURAL RESOURCE ACCOUNTS AND RELATED INCIDATORS:

THE CASE OF WATER RESOURCE & USES ACCOUNTS
Presentation by France65
1. Objectives of water Resource & Uses accounts
Water resource and uses accounts are part of environmental accounts in the domain of water.
Two approaches of water accounts can be considered. The first one is media oriented. It aims at describing
the origin and quantity of the resource and the quality of the water. The second approach is social-
economic oriented and deals with the uses of the water resource (by sectors), the polluting discharges (by
sectors), the protection and operation expenditures and costs and the value of the assets.
The expected output of water resource and use accounts is indicators related to :
− Resource Availability: e.g. Groundwater replenishment rate, Natural inflows, Annual
available resource (net…)..)
− Resource Use: e.g. Intensity of abstraction by sectors, Total consumption of water, Returns of
waste water)
− Resource Depletion: e.g. Evapotranspiration inducted by irrigation, Depletion of groundwater
(net accumulation…)…)
2. History
Water resource and uses accounts have been currently produced by hydrologists, agronomists,
water or energy companies or urban managers since centuries… However, they have been developed in a
structured format a few years ago in France in the context of the so-called Natural Patrimony Accounts
project. In 1990, OECD proposed to experiment work on Natural resource accounting in two domains :
forests and inland water. It was decided that the "Pilot study on inland water accounts" would consist in a
test of the French methodology. Several countries started to collect statistics in this framework
(Netherlands, Finland…). The most complete set of accounts (quantity, quality and expenditure) was
produced in Spain.
3. Selected recent developments in the continuation of the OECD exercise

Although the subsequent history may seem hectic (no decision has been made so far considering
its implementation), it is remarkable to note that the basic model as examined at OECD has continued
being tested in pilot projects, but also as the foundation of the development of data base on water.
In France, a second generation of accounts was tested by IFEN, and results presented to the new
"Committee of environmental accounts and economics" chaired by the Minister of Environment. First
simplified accounts of the quality of rivers (1999) have been computed; a prototype of a model for
calculation of the emissions to rivers has been successfully tested on a large drainage basin (the Loire
river) and will be generalised next year to all of France, accounts of protection expenditure and of costs of
production of the resource are produced on a yearly basis. As compared to the early times, three points can
65 . Paper prepared by Mr. Jean-Louis Weber, IFEN, France.
105
characterise this new generation of accounts : breakdown by catchments, breakdown by quarters, use of
physical modelling for integrating the data delivered by the monitoring stations.
Pilot studies were worked out during the period in Indonesia (in the context of the
implementation of a water agency in the catchment of River Brantas) and in Columbia (in parallel with the
implementation of the law on water).
One of the most recent achievements as taken place in the Republic of Moldova with the
establishment in 2000 of preliminary accounts for the years 1994 and 1998. Consideration is now given to
the possibility of using the water accounts for streamlining the information system on water and as a
reference in the “national water strategy”.
Another pilot study has been recently published for a catchment in Chile (River Aconcagua). It is
based on the Spanish exercise and the time series cover Resource & Use (1992-97), Quality of rivers
(1986-1997), Production & Expenditure (1997).
A similar work has been done in Australia (with publication in 2000). It covers Resource & Uses
accounts by catchments, details for sectors. The report introduces the concept of divertible resource (the
part of the overall available resource which can be use in acceptable environmental conditions).
Last but not least, the SEEA under revision will incorporate some elements of the water accounts.
4. The accounting framework of water R & U

As all accounting framework Water R&U Accounts are based on a system analysis. It can be
illuminated by the following scheme which describes all the flows. Some key point have to be underlined:
accounts are established for a territory (a country, a region, a river basin,…); this territory exchanges water
Outside territory
Territory of reference
Precipitations Evapotranspiration
Atmosphere
Inland water INLAND WATER SYSTEM

Soil & Vegetation Inland water
Internal
Inflows Underground Water

transfers
Outflows
Snow & Ice
Lakes & Dams Sea
Rivers
Primary Returns + Evapotranspiration
withdrawals Irrigation
Users Exports
USERS SYSTEM
Users
Imports
with other territories and the sea. The overall system is split in two subsystems, the hydrological system
and the users systems; they are connected by withdrawal of water (on the one hand) and returns of water,
including wastewater (on the other hand). The water cycle starts with the precipitation and finished with
the evapo-transpiration and the flowing out to the sea. Within each subsystem, breakdowns are produced to
106
show in which way the water resource is available (humidity of soil, groundwater, snow, lake, river,…)
and who are the users. Accordingly, matrixes describe how the initial input (precipitation, withdrawal,…)
is distributed to final resource (through internal transfers) and to final users (through supply and purchase).
From this scheme is derived the water resource and uses accounting framework. The presentation
will be illustrated by the water accounts of Moldova. (Please, note that the results, being provisional,
cannot be quoted).
Basically, the R & U water accounts are made of 3 sets of accounts :
− The tables of the users system : uses by sectors and transactions between sectors
− The account of the hydrological system : origin, transfers between natural subsystems and
making of the water resource
− The synthesis balance sheet (where the flows are given a + or a -, according to the way they
are going)
The users are sectors or activities of the National Accounts; for analytical purposes, key sectors
in the domain of water are highlighted (water supply of irrigation water and of drink water, sewerage,
fisheries, agriculture and energy). Industry is aggregated here but some breakdowns could be relevant at
the regional level where a particular industry is a key user. Specific operations of the users deal with water
extraction and imports, distribution, irrigation, returns of wastewater, exports. Leaks and losses are
considered as a use. In situ uses (bathing in rivers, navigation) are not taken into account here. Import and
export means that water enters or flows out of the territory in an artificial feature (pipe or channel).
The accounts of the hydrological system calculate the total input (precipitation, outside natural
influents, returns of water (incl. wastewater and irrigation, spontaneous internal transfers) and the total
output (spontaneous internal transfers, primary withdrawals, evapo-transpiration, natural outflows to other
territories and the sea). The accounting balance between input and output is the Net Accumulation of
Water (change in stocks). The spontaneous internal transfers between subsystems are described in a matrix.
The Synthesis Balance Sheet presents side by side the flows accounts of the hydrological and of
the users systems. These flows are given a + or a - according to the fact they are an inflow or an outflow. In
addition, the Balance Sheet integrates the stock account of the various components.
5. Examples of indicators of material flows derived from the water resource & uses accounts
Indicators can be easily extracted from the accounts and give useful information. Below,
comparisons are made between Moldova in 1994 and 1998, France in 1981 and Spain (average year for
resource, 1992 for uses).
A first information is given on the respective hydrological profiles (e.g. m3 per hectares).
Considering the rain, Moldova is closer to Spain (with an average level of precipitation even lower) than
France. Due to hot summers and evapo-transpiration, there is a recurrent deficit of water in Moldova, a
country where agriculture is a key activity. At the same time, Moldova can rely on an external resource, the
River Prut and the River Nistru or Dniestr, which is not the case for Spain. But they are international rivers
which operation is limited by international conventions.
If we consider the uses (m3 per ha. or per capita) we can see that in spite of similar climate
conditions and existing irrigation schemes, the intensity of use water by agriculture is much lower in
Moldova than in Spain. Even though we consider the very high level of evaporation generated by irrigation
in Spain, the gap between Moldova (339 m3/ha in 94) and Spain (1196 m3/ha) suggests that there are
margins for developing the irrigation in the latter country. We can also observe the impact on water of the
economic crisis in Moldova in the sectors of energy and agriculture (although, in the latter case, the high
level of precipitation in 1998 has to be taken into account).
107
6. Next steps (very incomplete panorama)

In general terms, one can say that the original model of water accounting for R&U as well as for
the quality of the rivers is robust. Solutions have been identified and implemented for producing such
accounts starting from the existing databases and improving the quality of the numbers step by step. These
solutions encompass physical modelling, sampling techniques and geo-statistics.
As they are now, the water accounts prove to be very useful for securing the best possible
comparability over time and in space (between catchments or countries). This is generally not guaranteed
with the current statistics based on the aggregation of results from monitoring stations when the monitoring
networks have not been designed with sampling preoccupation. Accounts also facilitate the cross checking
of the statistics and help in identifying gaps or inconsistencies.
The main improvements which are required and foreseen are the following :
− a more operational definition of the "annual available resource" to take into account the
physical limitations (the minimum to keep in the ecosystems and the purely physical
constraints) as well as the quality of the resource; interesting proposals have been made in
this way by Eurostat (Water accounts task force) and by the ABS in its recent publication.
− A full development of the rivers quality accounts for the causes of changes of quality, which
constitute a gateway to economic activities.
− The development of a methodology to account for the quality of groundwater.
− The development of accounts of emissions to water. In this domain, a reasonable balance has
to be found between the detail of accounting by activities (NAMEA type, similar detail in the
ABS accounts) required for macro-economic analysis and the breakdown by catchment areas
required for the physical modelling of the leakage from non point sources (now) and the
relation between emissions of pollutants and the quality of the water (in the future).
− The development of costs calculations of the resource (asset and product) and of the
maintenance of its quality.
108
(provisional results, not to be quoted)
109
INLAND WATER ACCOUNTS / RESOURCE ACCOUNTS IN RAW QUANTITIES
T2 - HYDROLOGICAL SYSTEM FLOWS ACCOUNT (INPUT-OUTPUT TABLE)

YEAR : 1998 - COUNTRY : REPUBLIC OF MOLDOVA - UNIT : Mm³
T2A - TOTAL INPUT AND OUTPUT TO THE (FROM THE) HYDROLOGICAL SYSTEM
w1 w2 w3 w4 w5
TOTAL
Lakes &
Soil & vegetation Groundwater Snow & ice Rivers
reservoirs
f11 Precipitation
23516,3 " 362,5 290,0 24168,9
f12 Outside natural influents
1100,0 17750,0 18850,0
f311 Returns of lost water (incl. leaks)
113,7 113,7
f312 Returns of waste water
73,5 973,2 1046,6
f321 Irrigation
f23 A SPONTANEOUS INTERNAL INPUT

310,0 2123,0 2453,0
D TOTAL INPUT TO THE HYDROLOGICAL SYSTEM

23516,3 1597,1 362,5 21136,2 46632,2
f23B SPONTANEOUS INTERNAL OUTPUT
2299,0 134,0 20,0 2453,0
f34 Primary withdrawals (extraction…)
196,0 196,0
f16 Evapo-transpiration 21277,6 415,9 332,8 22026,3
f131 Natural outflows towards territories (regions, basins…)

1267,1 19420,0 20687,1
f132 Natural outflows towards the sea
E WITHDRAWALS AND FINAL OUTPUT 21277,6 1463,2 415,9 19752,8 42909,5

F CHANGES IN STOCKS (NET ACCUMULATION OF WATER) = (H - E)
-60,3 -53,4 1363,4 1269,7
TOTAL OUTPUT FROM THE WATER SYSTEM AND NET
G
ACCUMULATION = (f23B + E + F) 23516,3 1597,1 362,5 21136,2 46632,2
H GLOBAL AVAILABLE ANNUAL RESOURCE = (D - f23B) 21217,3 1463,2 362,5 21116,2 44179,2
T2B - INTERNAL TRANFERS BETWEEN HYDROLOGICAL SUB-SYSTEMS
f23B
w1 w2 w3 w4 w5
SPONTANEOUS
Lakes &
to recipient Soil & vegetation Groundwater Snow & ice
reservoirs
Rivers INTERNAL
from origin OUTPUT
w1 Soil & vegetation
310,0 1989,0 2299,0
w2 Groundwater
" 134,0 134,0
w3 Snow & ice
w4 Lakes & reservoirs

" "
w5 Rivers
" 20,0 20,0
f23 A SPONTANEOUS INTERNAL INPUT 330,0 2123,0 2453,0
110
COMPARISON PER HECTARE OF WATER RESOURCE AND USES (UNIT : m³/ha)

MOLDOVA MOLDOVA FRANCE SPAIN
1994 1998 1981 1992
STOCK/UNDERGROUND WATER 44313 44313 40000 28162
STOCK/RESERVOIRS 630 630 164 435
PRECIPITATIONS 4140 7140 10455 6719
EXTERNAL INFLOW (TOTAL) 2984 5569 691 0
EVAPOTRANSPIRATION GENERATED BY IRRIGATION 288 54 237 1057
EFFECTIVE REAL EVAPOTRANSPIRATION 3990 6514 5552 4866
NET INTERNAL AVAILABILITY (= EFFICIENT RAIN) 348 668 4985 2277
NET ANNUAL AVAILABLE RESOURCE 3331 6237 5676 2277
USES/AGRICULTURE (INCL. TOTAL RESOURCE FOR IRRIGATION) 339 64 239 1196
USES/ENERGY (HYDROPOWER & COOLING) 438 219 355 447
USES/INDUSTRY 19 14 39 37
USES/HOUSEHOLDS 71 66 45 61
COMPARISON PER CAPITA OF WATER RESOURCE AND USES (UNIT : m³/capita)
USES/AGRICULTURE (INCL. TOTAL RESOURCE FOR IRRIGATION) 174 33 84 623

USES/ENERGY (HYDROPOWER & COOLING) 345 172 362 579
USES/INDUSTRY 15 11 40 48
USES/HOUSEHOLDS 56 52 46 79
111
BIBLIOGRAPHY
Tafi, J.(DASS, Moldova) & Weber, J.-L.(Ifen, France) : "Inland Water Accounts of the Republic of
Moldova – Preliminary Results of Resource Accounts in Raw Quantities, 1994 & 1998", draft
technical report for Eurostat, DASS & Ifen, July 2000.
Crouzet, Ph. (Ifen), Germain Ch. & Le Gall, G. (BETURE-CEREC) : "Les Comptes de la Qualité des
Cours d’Eau", Etudes et travaux n°25, Ifen, 1999. Synthesis available in English.
Meza, F. (PUC), Jiliberto, R. (TAU), Maldini, F.(ECONAT) et alii : Cuentas Ambientales del Recurso
Agua en Chile, Documento de Trabajo Nº 11, Serie Economía Ambiental, Pontificia Universidad
Católica de Chile, Facultad de Agronomía y Ciencias Forestales, Departamento de Recursos
Naturales, 1999.
McLennan, W., "Water Accounts for Australia, 1993-94 to 1996-97", ABS, Catalogue N° 4610.0,
Australian Bureau of Statistics, 2000.
112
WORKING GROUP ON ENVIRONMENTAL INFORMATION AND OUTLOOKS
LIST OF PARTICIPANTS OF THE SPECIAL SESSION ON MFA
Chair: Mr. Yuichi MORIGUCHI, Japan
AUTRICHE/AUSTRIA
Ms. Caroline VOGL
Federal Ministry of Agriculture, Forestry,
Environment and Water Management
CANADA
Mr. Martin LEMIRE Mr. Dennis O'FARRELL
Statistics Canada Environment Canada
Environment Accounts and Statistics Division Indicators and Assessment Office
COREE/KOREA
Ms. Young-Sun KIM Mr. Yeon-Soo PARK
Ministry of Environment Permanent Delegation of Korea to the OECD
DANEMARK/DENMARK
Mr. Preben ETWIL Ms. Astrid LADEFOGED
Danmarks Statistik Danish Environmental Protection Agency
ETATS-UNIS/UNITED STATES
Mr. Frederick (Derry) ALLEN Ms. Elizabeth COTSWORTH
US-EPA US-EPA
Office of Policy, Economics and Innovation Director, Office of Solid Waste
Mr. James LEE Ms. Angela LEITH
US-EPA US-EPA / Office of Solid Waste
Mr. James C. MORANT Mr. Eric RODENBURG
US-EPA Chief, Minerals and Materials Analysis Section
Office of the Administrator U.S. Geological Survey, U.S. Department of the Interior
Mr. N. Phil ROSS Ms. Breck MILROY
US-EPA / Office of Environmental Information Permanent Delegation of the United States to the OECD
Center for Environmental Information Statistics
FINLANDE/FINLAND
Mr. Leo KOLTTOLA
Statistics Finland
FRANCE
Mr. Jean-Louis WEBER
Institut Francais de l'Environnement (IFEN)1
HONGRIE/HUNGARY
Mr. Elemer SZABO
Ministry of Environment
ITALIE/ITALY
Mr. Aldo FEMIA
ISTAT
JAPON/JAPAN
Mr. Kazuhiro MIYAUCHI Dr. Yuichi MORIGUCHI (Chair of Special Session on MFA)
Environmental Planning Division Resources Management Section
Planning & Coordination Bureau Social and Environmental Systems Division
Environment Agency of Japan National Institute for Environmental Studies
Mr. Akio TAKEMOTO
Permanent Delegation of Japan to the OECD
113
MEXIQUE/MEXICO
Mr. Yosu RODRIGUEZ ALDABE Mr. Armando YANEZ
Dirección de Estadísticas Ambientales Dirección de Estadísticas Ambientales
Secretaria de Medio Ambiente, Recursos Naturales y Pesca Secretaria de Medio Ambiente, Recursos Naturales y Pesca
(SEMARNAP) (SEMARNAP)
NORVEGE/NORWAY
Mr. Øystein NESJE
Ministry of Environment
Section for Environmental Data
PAYS BAS/NETHERLANDS
Mr. René KLEIJN Mr. Cornelis VIJVERBERG
Leiden University, Centre of Environmental Science (CML) Ministry of Housing, Spatial Planning and the Environment
(VROM),
REPUBLIQUE TCHEQUE/CZECH REPUBLIC
Mr. Erich LIPPERT
Ministry of the Environment
Department of Environmental Statistics
ROYAUME-UNI/UNITED KINGDOM
Ms. Pat KILBEY
Environment Protection Statistics and Information Management
Department of the Environment, Transport and the Regions
SUEDE/SWEDEN
Mrs. Eiwor HÖGLUND-DÁVILA Mrs. Manuela NOTTER
Statistics Sweden Swedish Environmental Protection Agency
Environment Statistics Environmental Monitoring
SUISSE/SWITZERLAND
Mr. Peter GLAUSER Mr. Thomas KLINGL
Office fédéral de la statistique Office fédéral de l’environnement, des forêts et du paysage
Section Environnement Département fédéral de l’environnement, des transports
de l’énergie et de la communication
CCE/CEC
Mr. Anton STEURER
Statistical Office of the European Communities - Eurostat
Luxembourg
BIAC IFF
Mr. André BOURASSA Mr. Heinz SCHANDL
Business and Industry Advisory Committee (BIAC) Institute for Interdisciplinary Studies of Austrian
International Council on Metals and the Environment Universities (IFF) – Department for Social Ecology
(ICME), Ottawa, Canada Vienna, Austria
TELLUS INSTITUTE WUPPERTAL INSTITUTE

Dr. John Stutz Mr. Stefan BRINGEZU
Tellus Institute Wuppertal Institute for Climate, Environment and Energy
Boston, USA Department Material Flows and Structural Change
Wuppertal, Germany
WORLD RESOURCES INSTITUTE
Ms. Fran IRWIN Ms. Emily MATTHEWS
World Resources Institute Senior Associate, Information Program
Washington D.C. USA World Resources Institute
Washington D.C., USA
YALE UNIVERSITY
Mr. Reid LIFSET
Associate Director, Industrial Environmental Management
Programme, Yale University
Editor-in-Chief, Journal of Industrial Ecology (Yale/MIT)
New Haven, USA
114
OECD SECRETARIAT / SECRÉTARIAT DE L’OCDE
Environment Directorate
Ms. Joke WALLER-HUNTER Director

Mr. Kenneth RUFFING Deputy Director
Environmental Performance and Information Division
Mr. Christian AVÉROUS Head of Division
Ms. Myriam LINSTER
Mr. Heino VON MEYER
Environmental Health and Safety Division
Ms. Soizick de TILLY
Mr. Henrik HARJULA
Pollution Prevention and Control Division
Mr. Peter AERTKER
Mr. Fabio VANCINI
Statistics Directorate
Mrs. Anne HARRISON
Mr. Carl OBST
115

Special Session On Material Flow Accounting: OECD Working Group On Environmental Information and Outlooks (WGEIO)

Uploaded by

Copyright:

Available Formats

Special Session On Material Flow Accounting: OECD Working Group On Environmental Information and Outlooks (WGEIO)

Uploaded by

Document Information

Original Description:

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Special Session On Material Flow Accounting: OECD Working Group On Environmental Information and Outlooks (WGEIO)

Uploaded by

Copyright:

Available Formats

OECD Working Group on Environmental Information

and Outlooks (WGEIO)

PAPERS AND PRESENTATIONS

Questions addressed included the following:

Working Group on Environmental Information

PART II. METHODOLOGICAL ISSUES................................................................................................... 67

Introductory speech by Ms. Joke Waller-Hunter

Mr. Chairman, Dear participants,

LINK TO SUSTAINABLE DEVELOPMENT

LINK TO INFORMATION AND DECISION MAKING

USE OF ENVIRONMENTAL ACCOUNTING

OBJECTIVES OF THIS SPECIAL SESSION

OECD Working Group on Environmental Information

SPECIAL SESSION ON MATERIAL FLOW ACCOUNTING

LINKS WITH ENVIRONMENTAL AND SUSTAINABILITY INDICATORS

PART I. DEVELOPMENT AND MOST PROMISING USES

HISTORY AND OVERVIEW OF MATERIAL FLOW ANALYSIS2,3

1. Material flow analysis and accounting

2 Paper prepared by Mr. Stefan Bringezu, Wuppertal Institute.

2. Concepts for sustaining the metabolism of economies

3. Impacts of material flows and types of indicators

Volume based indicators Impact based indicators

5. Use of material flow related analyses

− the derivation of indicators for progress towards sustainability.

Add. Air Water

7. MFA based indicators

Adriaanse, Albert, S. Bringezu, A. Hammond, Y. Moriguchi, E. Rodenburg, D. Rogich, H. Schütz (1997),

Matthews, Emily, C. Amann, S. Bringezu, M. Fischer-Kowalski, W. Hüttler, R. Kleijn, Y. Moriguchi, C.

INTERNATIONAL COMPARISON OF INFLOWS AND OUTFLOWS12

MATERIAL FLOWS TRACKED FOR THE STUDY

International Comparison of Material Flows

 Conventional wastes (emissions, discharges, solid wastes, system

comparable across countries.

indicators of flows are physical descriptors of an economy, just as GDP

 The actual quantities of material going to stock appear to be higher in

however not easy; they may reflect differences in building practices

MATERIAL FLOWS AND DECOUPLING

Detailed Composition of Flows: the U.S. example

A disaggregation of outputs by environmental media (air, land, water) shows that.

Selected individual flows in the U.S.

Many of these substances occur not at the processing or

MATERIAL FLOW ACCOUNTING:

2. Material and Energy Flow Accounts in Canada

2.1 Why account for materials and energy?

3. The MEFA accounting framework

4. Uses of the Material and Energy Flow Accounts

5. Material and Energy Flow Indicators

7. Future directions for Material and Energy Flow Accounts

ANALYSIS OF MATERIAL FLOWS FOR SUSTAINABILITY POLICY:

2. Data and method of analysis

3.1 Absolute material use by vertically integrated branch

Materials of domestic origin Imported materials

3.2 Intensity of use of materials by vertically integrated branch

Figure 1. Average Intensity of use by vertically integrated sector

1,5 Agriculture, breeding, forest, fishery

PA and non market services

Non-energy minerals- imported

Undetermined materials- imported

Conventional wastes (emissions, discharges, solid wastes, system

The actual quantities of material going to stock appear to be higher in