Framework for Chemical Risk Management under REACH: Regulatory Decision-Making Steffen Erler
iSmithers – A Smithers Gr...
60 downloads
1025 Views
1MB Size
Report
This content was uploaded by our users and we assume good faith they have the permission to share this book. If you own the copyright to this book and it is wrongfully on our website, we offer a simple DMCA procedure to remove your content from our site. Start by pressing the button below!
Report copyright / DMCA form
Framework for Chemical Risk Management under REACH: Regulatory Decision-Making Steffen Erler
iSmithers – A Smithers Group Company Shawbury, Shrewsbury, Shropshire, SY4 4NR, United Kingdom Telephone: +44 (0)1939 250383 Fax: +44 (0)1939 251118 http://www.rapra.net
First Published in 2009 by
iSmithers Shawbury, Shrewsbury, Shropshire, SY4 4NR, UK
1st Edition ©2007, Steffen Erler 1st Revised Edition ©2009, Smithers Rapra
All rights reserved. Except as permitted under current legislation no part of this publication may be photocopied, reproduced or distributed in any form or by any means or stored in a database or retrieval system, without the prior permission from the copyright holder.
A catalogue record for this book is available from the British Library.
Every effort has been made to contact copyright holders of any material reproduced within the text and the authors and publishers apologise if any have been overlooked.
ISBN: 978-1-84735-400-6 (Hardback) 978-1-84735-401-3 (Softback) 978-1-84735-402-0 (ebook) Typeset by iSmithers Printed and bound by Lightning Source Inc.
C 1.
2.
ontents
Introduction ......................................................................... 1 1.1
The New EU Chemicals Policy - REACH ................... 1
1.2
Why Regulate Chemicals? .......................................... 2
1.3
Risk Definitions .......................................................... 6
1.4
What are Chemicals? - Defining the scope of the research project ........................................................ 10
1.5
Policy, Risk Management and Risk Reduction Strategies .................................................................. 13
1.6
Rationale and Choice of Countries ........................... 16
1.7
Conclusion & Research Questions .......................... 18
Literature Review ............................................................... 21 2.1
2.2
Chemical Risks ......................................................... 25
2.1.1
Hazard and Exposure ................................25
2.1.2
Toxicology.................................................30
2.1.3
Exposure Assessment.................................35
2.1.4 Risk Characterisation ................................38 Obstacles to Risk Management ................................ 41 2.2.1
2.3
Structure of Risk Management ..................41
2.2.2 Existing Chemicals Regulation ..................44 The Regulatory Process ............................................ 51 2.3.1
Regulatory Options ...................................51
2.3.2
Recent Developments in Regulation ..........55
i
Framework for Chemical Risk Management under REACH
2.3.3 2.4
2.5 3.
4.
ii
Regulatory Decision-Making .....................57
2.3.4 Regulatory Approaches .............................59 REACH .................................................................... 67 2.4.1
A New Era for Chemical Control .............67
2.4.2
Uncertainties in the New System ...............73
2.4.3
Decision-making under REACH................75
2.4.4
Potential Improvement to Health and the Environment..............................................76
2.4.5 Business Effects of REACH: Cry Wolf? .....77 Conclusions .............................................................. 80
Methodology ...................................................................... 83 3.1
Research Design ....................................................... 83
3.2
Analytical Framework .............................................. 88
3.3
Interview Selection .................................................... 91
3.4
Interview Technique.................................................. 92
3.5
Initial Analysis ......................................................... 94
3.6
Interviewee Response Validation – Cross-Checking .. 96
3.7
Systems Framework .................................................. 97
3.8
Strengths & Limitations of this Research .................. 99
3.9
Conclusion ............................................................. 104
National Approaches........................................................ 105 4.1
Chemical Landscapes .............................................. 106
4.2
Regulatory Approaches .......................................... 109
4.3
Regulatory Administrations .................................... 117
4.4
Policy Styles and Implications for Risk Management ................................................... 126
4.4.1
Policy Styles.............................................126
4.4.2
Strengths and Weaknesses .......................131
4.4.3
EU Decision-Making ...............................134
Contents 4.5
4.6 4.7 4.8
5.
Social and Cultural Contexts .................................. 137
4.5.1
Legal Systems and the Role of Experts ....137
4.5.2
Future Socio-Economic Concerns and Chemical Safety .......................................139
4.5.3 Public and Political Catalysts...................141 Propensity to Change .............................................. 144 Preparing for Implementing REACH ..................... 146 Conclusions ............................................................ 149 4.8.1
Discussion and Conclusions ....................149
4.8.2
Recommendations ...................................151
A Systems Framework to Implement REACH .................. 155 5.1
The Need for Regulatory Reform ........................... 155
5.1.1 5.2 5.3
5.4 5.5 6.
Risk Assessment ......................................157
5.1.2 Risk-Reduction Strategy (RRS)................158 REACH - A Critical Analysis .................................. 166 Systems Framework ................................................ 175 5.3.1
Decision-Making Rules ...........................178
5.3.2
Technical Guidance for Decision-Making 195
5.3.3 Administrative Structures ........................205 Prioritisation of Regulatory Decision-Making ........ 209 Conclusions ............................................................ 233
Evaluating the Systems Framework .................................. 237 6.1
From Risk to Safety: Review of the Systems Framework ............................................................ 238
6.2
Testing the Systems Framework .............................. 244
6.2.1
Regulatory Outcomes ..............................244
6.2.2 6.3
Prioritisation of Regulatory Decision-Making – Results ......................255 National Approaches under the Systems
iii
Framework for Chemical Risk Management under REACH Framework ............................................................. 262
7.
6.4
Consensus on Banning Chemical Production or Use ................................................... 266
6.5
Discussion & Conclusions ...................................... 269
Conclusion ....................................................................... 273 7.1
Harmonisation of EU Risk Management ................ 274
7.2
Implications of the Research ................................... 278
7.3 7.4 7.5 7.6
7.2.1
Sustainability in the EU Chemical Industry ...................................................278
7.2.2
National Approaches and EU Decision-Making .....................................281
7.2.3
Implementing REACH ............................282
7.2.4 Risk Analysis ...........................................284 Global Dimensions of REACH and the Research.... 286 Milestones .............................................................. 287 Research Contribution ............................................ 290 Future Research ...................................................... 292
Acknowledgements .................................................................. 293 References ................................................................................ 295 Appendix ................................................................................. 361 Abbreviations ........................................................................... 409 Index ........................................................................................ 415
iv
1
Introduction
- Politics A strife of interests masquerading as a contest of principles (Ambrose Bierce, The Cynic’s Word Book, 1906)
1.1 The New EU Chemicals Policy - REACH Chemicals policy and risk management challenges European Union (EU) regulators. Decision-making is increasingly occurring at the EU level, but the process of developing, selecting and implementing riskreduction measures ultimately depends on the roles, responsibilities and resources of actors at a national level. At the time of writing (May 2007), this book describes why, after eight years of policy development, the implementation of the EU Regulation for the Registration, Evaluation and Authorisation of Chemicals (REACH) can benefit from a comparative analysis of national approaches. The book presents the regulatory structures and cultures of four Member States prominent in EU chemical risk management decisionmaking: France, Germany, Sweden and the UK. The research identifies that REACH requires structural and organisational reform beyond those offered by the legislation. In response, the book proposes a systematic process for EU regulatory risk management and communication. Taking the form of regulatory technical guidance, the ‘systems framework’ would consist of decision-making rules to guide regulators and industry through future regulatory processes. Criteria are developed that differentiate between the need for action
1
Framework for Chemical Risk Management under REACH at a national level versus an EU level, and define the inter-relationship between existing consumer, occupational and environmental legislative frameworks.
1.2 Why Regulate Chemicals? Chemicals produced by humans form the strands of a complex societal web. The network begins with the production of a single chemical that branches out into hundreds of uses. Several hundred chemicals can be used or created in the process of producing one final product. Practically every part of the built environment and industrial activity involves chemical products. While generating wealth and employment, chemical production provides countless valuable services to society: from the manufacture of construction materials to the synthesis of life-saving pharmaceuticals. Whether natural or man-made, a chemical has the potential of causing harm to a living organism. Water can present a risk of drowning to a bathing human. Iron can be fatal to humans or animals when ingested in very high quantities, yet it is an essential mineral for the functioning of most organisms. Similarly, negative effects of a given chemical depend on complex biochemical interactions. For instance, chromium acts as a potent carcinogen and skin sensitiser in a fully oxidised state, but its reduced trivalent state exhibits few signs of having a toxic mechanism in organisms [1]. The term ‘hazard’ describes the intrinsic physical and toxicological properties of a chemical. ‘Risk’ relates to the possibility and severity of damage to humans or ecosystems after exposure to a chemical (Section 2.1). Defining risk levels must also account for the perceived nature of danger, as well as the potential benefits of a risky activity (discussed in Section 1.3). Regulating chemicals responds to the need to control existing or potential risks, while retaining the socioeconomic benefits arising from the production and use of chemicals (Sections 2.2 and 2.3). 2
Introduction Regulation defines and delineates responsibility for chemical management between institutions, enterprises, non-governmental organisations (NGO), social groups and individuals. It also prohibits the production or use of some chemicals. In the EU, national policy-makers and regulators have the duty of ensuring a high level of protection to human health and the environment1. Companies must adhere to regulatory requirements and are essentially liable for controlling risks that are otherwise not specifically regulated. Distributors and retailers can also be accountable for ensuring that they supply safe products to customers, provide adequate warnings and instructions with dangerous products, or sell dangerous products only to licensed users. In turn, chemical safety inevitably depends on downstream users,2 and consumers acting responsibly when using and disposing of products. Consumers also have implicit responsibility for chemical production because customer choice3 influences market demand (Section 2.3). When legislating, political and regulatory decisions work within certain boundaries of risk ‘acceptability’, established by stakeholders and the general public. Regulating chemicals becomes a balancing act between risks and benefits to society, but what is acceptable to one person or group can be different to another person or group. To avoid the complexities associated with establishing acceptability, ‘tolerability’ has recently entered the risk policy lexicon of some EU regulators (e.g., [2]). Tolerability refers to [3]:
1
Firmly established under Article 2 of the European Community (EC) Treaty, legislators should serve to preserve, protect and improve the quality of the environment and to protect human health. Most chemicals legislation aimed to protect human health and the environment has been introduced under the EC Treaty. It is therefore referred to ‘EC’ rather than ‘EU’ legislation.
2
‘Downstream users’ refers to all organisations or individuals using industrial or professional chemicals other than actual chemical producers. Industrial uses include processing, manufacturing, and formulating, as well as using the chemical product to carry out a process or service. ‘Professional uses’ covers the use of all other chemicals except products supplied to the general public.
3
The book uses the term ‘customer’ when referring to traders, retailers and consumers influencing market demand for chemical products.
3
Framework for Chemical Risk Management under REACH ‘…a willingness by society as a whole to live with a risk so as to secure certain benefits and in the confidence that the risk is one that is worth taking and that it is being properly controlled. Tolerability does not imply that the risk will be acceptable to everyone, i.e., that everyone would agree without reservation to take the risk or have it imposed on them.’ Risk tolerability reflects politically acceptable levels of risk. Tolerability seeks to balance stakeholder and public views of acceptability as individuals and societal groups4. One interpretation of tolerability is that it refers to levels of risk that are not expected to cause public outrage. At the other extreme, an interpretation is that tolerability seeks to minimise overall risks while increasing overall benefits proportionately to those that are subjected to the risk. Risk tolerability ultimately depends on how individuals and societal groups perceive risks as encroaching on their fundamental rights and freedoms. To examine this latter phenomenon, individuals or societal groups can be described as areas of physical space or psychological space [4]. The area boundaries are defined by regulation and established by society and morality [4]. Actions or events that transgress personal or societal boundaries or intrude upon the circumscribed area without permission deny individual and societal rights and freedoms5 [4]. When released into the environment, chemicals can transcend personal and societal boundaries. Chemical contamination of shared common physical space6 or resources affects the rights and freedoms of individuals or groups to access these 4
Societal groups include stakeholder groups and other organised sub-sets of the general public.
5
In this context, justice can be seen as a system of punishment for the instigator of the action and compensation for the person whose boundary has been involuntarily crossed.
6
Animal rights will not be discussed in this book other than in the context of the impacts of chemicals on ecosystems and the use of animal testing during risk assessment processes (Chapter 2).
4
Introduction resources or to use them7. For instance, Chapter 19 of Agenda 21, the policy document that emerged from the United Nations Conference on Environment and Development in Rio, Brazil, in 1992, specifically recognises that ‘chemical risks do not respect national boundaries’. For the purposes of this PhD, which investigated chemical risk management in a European context, the basis for the rights of persons is further linked to questions of inter-generational equity. Balancing between chemical risks and benefits during decision-making must follow the principle of sustainable development enshrined in Articles 2 and 6 of the EC Treaty [5] and defined by the World Commission on Environment and the Development as development that ‘meets the needs of the present generation without compromising the ability of future generations to meet their own needs’ [5]. Instead of examining individual perceptions of risk acceptability, which is a large field of research, the book considers stakeholder group views on chemical risks and benefits8. In this way, the research project establishes tolerability according to the policy and position of the primary institutions and stakeholder associations involved in national chemical risk management. Influence of these various stakeholder associations on regulatory decision-making was determined through an analysis of their respective roles, relationships and responsibilities. The circumstances necessary for groups to take action and succeed in achieving their objectives were also investigated.
7
Environmental economists and managers continue to debate the question of individual rights to use common resources. An anthology of reading on this subject is available in Hardin and Baden’s Managing the Commons (1977) or Baden and Noonan’s collection of works published under the same title in 1998.
8
A wide range of literature is available on the subject of risk perception. For a brief overview, readers are referred to Risk and Modern Society edited by Löfstedt and Frewer (1998). For a detailed investigation with respect to social constructions of risk, a collection of writings from prominent authors is presented in Social Theories of Risk edited by Krimsky and Golding (1992).
5
Framework for Chemical Risk Management under REACH
1.3 Risk Definitions Fundamental differences in the scientific community arise with regard to the precise definition of risk. This reflects a long-standing philosophical divide between positivism and relativism. More recently, the definition of risk has been subject to debate between realists and constructionists. Positivists assert that the risks are directly observable and measurable: science is provable 9. Relativists hold the view that scientific knowledge is bounded by paradigms of our understanding of the physical world [6]. Paradigms are incommensurable, so scientific discoveries are therefore always relative [7]. Two forms of relativism are differentiated: constrained [6] and unconstrained [8]. The latter approach holds that the ‘real world’ is 100% constructed through social and cultural influences [9]. The three approaches of positivism, constrained relativism and unconstrained relativism are schematically represented in Figure 1.1. A further divide between risk analysts arises between realism and constructionism, but this debate can be seen as unifying positivist and relativist approaches. Constructionism considers how social and cultural perspectives influence risk definitions and interpretations [12]. In comparison, realists exclude social and cultural phenomena in their reference of risk, but do acknowledge their existence [12]. Constructionism resembles constrained relativism and does not represent the paradigms of the unconstrained relativists. For positivists and realists, acknowledging risk perceptions provides a potential framework to incorporate the risk perceptions into their process of risk analysis because risk perceptions can be subject to scientific analysis as social phenomena. Realists make a clear distinction between objective and perceived risks. Objective risks are defined as a function of probability and 9
6
Here, positivism is taken to include falsification, where theories should be falsifiable, where the highest number of falsifiable elements that have not yet been contested is preferable [10]. Here, theories are not necessarily ‘true’ but a choice of the least unproven one to date [11].
Introduction
(i) Observations
Real World
(ii)
‘Real World’ a
‘Observations’
‘Real World’ b
‘Observations’
Real World
(iii)
‘Real World’ 1
‘Observations’
‘Real World’ 2
‘Observations’
‘Real World’ ...
‘Observations’
The real world is directly accessible. Observations reflect the real world. This reflection can be distorted if the stimuli of the real world are weak or interfered by biasing processes.
The real world is not directly accessible. What is ‘observed’ is a ‘world’ that is a function of the real world and an input of the observer. The resulting local feed-back loops imply a historically determined ‘reality Different ‘observers’ thus may live in different ‘worlds’; none of these is more true than the other.
The view of the observer is not in any way constrained by a real world. He ‘observes’ a ‘real world’ that is 100 % constructed; which construction is at stake is a matter of historical and social influences.
Figure 1.1 Positivism (i), constrained relativism (ii), and unconstrained relativism (iii) (taken from [11]. Reproduced with permission from K.R. Popper, The Logic of Scientific Study, 4th Edition, Hutchinson and Co., London, UK, 1972. ©1972, Hutchinson and Company) consequence. For a realist, perceived risks are the results of objective risks not being understood due to a lack of scientific knowledge amongst laypersons. For instance, a common view is that the public has difficulty in differentiating between hazard and risk10. Constructionists hold that a distinction cannot be made between objective and perceived risks. As one prominent risk researcher asserts [13]: 10
Recall that hazard is the intrinsic potential to cause harm (e.g., carcinogenicity), while risk refers to the probability of causing harm by exposure to a hazard (e.g., the possibility of developing cancer following exposure to a carcinogenic chemical).
7
Framework for Chemical Risk Management under REACH ‘…the view that a separation can be maintained between objective risk and subjective or perceived risk has come under increasing attack, to the extent that it is no longer a mainstream position.’ Recent approaches to risk management seek to limit the divide between realists and constructionists [12, 14]. First, the scientific basis of cause and effect is acknowledged, regardless of the extent that science is able to examine it. Second, scientific research and discovery are seen to influence perceptions. Recent risk research therefore proposes that risk analysis and risk management recognise that the realist and constructionist approaches are interlinked [12]. From a review of risk research, the following contextual variables affect individual and societal perceptions of various degrees of risk (adapted from [12]): UÊ /
iÊiÝ«iVÌi`ÊÕLiÀÊvÊ«iÀViÛi`Êv>Ì>ÌiÃÊÀÊÃÃiÃÆ UÊ /
iÊV>Ì>ÃÌÀ«
VÊ«ÌiÌ>]ÊÃÕV
Ê>ÃÊV>ÕÃ}Ê}ÌiÀÊÀÀiÛiÀÃLiÊ `>>}iÊÌÊ
i>Ì
ÊÀÊÌ
iÊiÛÀiÌÆ UÊ +Õ>Ì>ÌÛiÊV
>À>VÌiÀÃÌVÃÊÃÕV
Ê>ÃÊÌ
iÊ`i}ÀiiÊvÊ«iÀÃ>ÊVÌÀÊ vÊiÝ«ÃÕÀiÊÌÊ>Ê
>â>À`]Êv>>ÀÌÞÊÜÌ
ÊÌ
iÊ
>â>À`Æ UÊ ÕÌÕÀiÊ>`ÊÃV>ÊLiivÃÊ>ÃÃV>Ìi`ÊÜÌ
ÊÌ
iÊV>ÕÃiÊvÊÀÃÊÀÊÌ
iÊ risk-handling actors, including trust in regulatory agencies and risk-handling institutions11. Several theories seek to explain how these factors interrelate and why risks that are sometimes minor in quantitative terms sometimes produce massive socio-political reactions while major risks are sometimes ignored. The social amplification of risk model suggests that public (or stakeholder group) response to risk can be amplified or attenuated depending on how the reporting (i.e., risk communication) of the risk interacts with psychological, social, 11
8
Research examining public trust of governments suggests that there are two dimensions: the first is a general trust dimension which accounts for perception of competence, V>Ài]Êv>ÀiÃÃÊ>`Ê«iiÃÃÆÊÌ
iÊÃiV`ÊV«iÌÊÀi«ÀiÃiÌÃÊÌ
iÊ`i}ÀiiÊvÊÃVi«ÌVÃÊ regarding how risk policies are brought about and enacted [15].
Introduction cultural and institutional processes. Social network theory builds upon the social amplification model by examining how individuals form networks and systems that influence individual perceptions and create groups or communities with similar perceptions [16]. These social units behave like organisations in terms of attitude, knowledge and behavioural structures [16]. This book considers the importance that regulatory authorities and stakeholder associations have in the social processes of risk perception. Regulatory chemical risk management is concerned not just with controlling chemicals or specific uses of chemicals, but is an overall management process that involves complex social and cultural interactions [17]. In this respect, the research investigates the relationship between various actors involved in risk management at the local, regional, national and EU level. The research project builds upon existing chemical risk management practices, identifying the strengths and weaknesses of current scientific and technical approaches. It takes the view that science ÃÊ>Ê«ÀÌ>ÌÊ«ÕÌÊÌÊ`iVÃ>}ÆÊÌ
>Ì]ÊÊ}iiÀ>]ÊÃViViÊ ÃÊÀi>LiÊ>`ÊÀLÕÃÌÆÊ>`ÊÌ
>ÌÊVÃÌÀÕVÌÃÊvÊÀÃÊV>ÊLiÊÃÕLiVÌÊ to scientific analysis. These views closely follow constructionist or constrained relativist approaches to risk because they incorporate the influence of societal risk perceptions. This book takes the scientific results of risk assessment as a basis for defining risk. Risk assessment evaluates risk in terms of hazard and exposure, but reference to risk levels must account for different perceptions of risk as well as scientific uncertainties in risk assessment. In short, this research project considers the importance of social and institutional processes in influencing risk perceptions and risk acceptability. This book therefore takes a constrained relativist approach by incorporating risk perceptions in the research framework. An unconstrained relativist perspective would imply that no scientific study is reliable or robust. By contrast, a constrained relativist approach can provide a useful basis for examining the different social and cultural factors involved in regulatory risk management. 9
Framework for Chemical Risk Management under REACH
1.4 What are chemicals? - Defining the scope of the research project Chemicals are the building blocks of life. Elemental atoms such as carbon, hydrogen and oxygen combine to form molecules which make up the physical world of gases, liquids, solids, and living organisms. From cosmic reactions to the evolution of life on earth, nature has generated millions of chemicals. Humans produce and introduce chemicals that are otherwise extremely rare (if not completely foreign) in nature. The industrial revolution was instrumental to the development of these man-made chemicals [18]. There are about 27 million substances or groups of substances that have been identified, isolated or synthetically produced [19], of which approximately 0.3% (i.e., about 100,000) are traded on the EU market [20]. Whether naturally-occurring or synthetic, the term ‘chemical’ refers to substances and preparations. The former are elements and their compounds, whereas preparations are mixtures and solutions of substances (Appendix 1.1 – Definitions)12. ‘Chemical products’ does include ‘articles’ which are manufactured products or materials with specific form (e.g., plastics). The European Commission proposes that an article should be defined as ‘an object composed of substance(s) and/or preparation(s) which during production is given a specific shape, surface or design determining its end use function to a greater degree than its chemical composition does’ [21]. Depending on the interpretation of this definition, there are between 500,000 and 5 million categories of articles13 currently sold on the EU market [22]. Given the wide availability of consumer choice, the number of different products (i.e., the same category of product produced by different manufacturers) is seemingly endless. Substances and preparations may be present within articles (e.g., ink in a pen) where the article services as a container or chemical delivery device. Substances and preparations can provide a material matrix 12
This book uses the term ‘preparation’, but the term ‘preparation’ was changed in the REACH Regulation to ‘mixture’ by the Classification, Labelling and Packaging Regulation 1272/2008 in January 2009.
13
For example, a car pertains to a different product category than a hairdryer.
10
Introduction or form an integral part of an object, where only minute quantities of a chemical may be released (such as a dyed textile). It is therefore important to understand that during manufacture, substances and preparations can be bound within the matrix of an article (e.g., pigment in a plastic) or onto the surface of a matrix (e.g., layer of paint on a plastic). Various mechanisms of binding are available, depending on the process and the chemicals/materials used. In some cases, chemicals can be readily released even when bound inside a matrix. Chemical products available on the market can be categorised according to their function and use. Because a single chemical can have hundreds of uses, the details of specific uses become a major factor when determining the appropriateness of a risk management option (Chapter 2). Every industry sector makes some use of chemicals, from electrical engineering to publishing. For instance, chemicals are used to maintain industrial equipment. General business transactions also use chemicals in forms such as computers, electrical cables, photocopiers, and printer inks.
Included (non-exhaustive list) Adhesives Organic and inorganic bases Coolants Cleaners colorants (pigments and dyes) Fragrances Inks Paints Plasticisers (e.g., in plastics) Polymers (rubbers and plastics) Sealants Solvents Stabilisers Synthetic fibres Varnishes
Excluded (Appendix 1.1) Biocides Detergents Fertilisers Fuels Foodstuff additives and flavourings to foodstuffs Medicinal products (for humans and animals) Plant protection products\ Radioactive substances
Table 1.1 Chemicals included and excluded from the scope of the research project (note that the study also considers ‘articles’) 11
Framework for Chemical Risk Management under REACH This book covers the whole range of uses of chemicals within the chemical industry and downstream manufacturing sectors, as well as all their content in consumer products. Certain substances covered by specific product regulation that requires particular risk assessment and socio-economic analysis must be excluded from the scope of the research (Table 1.1). For instance, pesticides and pharmaceuticals cannot be regulated like most products produced by the chemical industry. A detailed explanation of the excluded substances is presented in Appendix 1.2 and a description of the chemical industry business activities is provided in Appendix 1.3. RAW MATERIAL INDUSTRIES e.g. biomaterials, gas, minerals, metals, oil
CHEMICAL INDUSTRY
BASIC CHEMICALS e.g. complex salts, fibres, gases, elastomers, intermediates, resins, polymers, solvents
FINE CHEMICALS
SPECIALTY CHEMICALS
e.g. complexing agents, fragrances, intermediates
e.g. adhesives, coatings, plastics, sealants,varnishes
LIFE SCIENCE PRODUCTS e.g. agrochemicals, biotech, pharmaceuticals
OTHER INDUSTRIES Construction, Electrical & Mechanical Engineering, Paper, Textiles, etc.
CONSUMER CARE PRODUCTS e.g. cosmetics, deodorisers detergents, hygiene products
CONSUMERS
Figure 1.2 Chemical industry classification (adapted from [23, 24])
Overall, this research study concerns the management of chemicals used in industry and which are available to professional and consumer users (Figure 1.2). Articles are also investigated which widen the project’s scope to include all finished manufactured products. 12
Introduction
1.5 Policy, Risk Management and Risk Reduction Strategies As discussed in Section 1.1, regulatory chemical risk management can be viewed as a process by which chemical risks are controlled so as to reach tolerable levels of risk. The research project takes its definition for chemical risk management from the European Commission [25]: ‘The process of weighing policy alternatives in the light of the result of a risk assessment and other relevant evaluation and, if required, selecting and implementing appropriate control options (which should, where appropriate, include monitoring/surveillance).’ In general, the scope and detail of a risk assessment determines the appropriateness of a given risk management measure (Section 2.3). Additional scientific study such as lifecycle assessment (LCA) can inform managers of the appropriateness of one regulatory measure compared with another in terms of the overall effects on health and the environment (Section 2.3). A socio-economic analysis of the effects of a regulatory measure can be carried out. Ultimately, the tolerability of a risk, which depends on perceived risks and benefits of a given risky chemical or activity, also influences the appropriateness of a risk reduction measure. Policy guides decision-making (Section 2.3). Policy comprises objectives [26] and strategies – purposive courses of action – to achieve them [27]. Several policy ‘principles’ also exist to help decision-makers, regulators and stakeholders enact, interpret and implement legal measures, particularly in the face of scientific uncertainty (Sections 2.3.4). Given the many variables that enter the risk management process, regulators often adopt different national approaches to regulating chemical risks (Section 2.3.4). It is these differences that the research project seeks to examine. This research project investigated risk management processes where risk assessment is merely one input. A detailed investigation into chemical risk assessment would present a research project in its 13
Framework for Chemical Risk Management under REACH own right and extends outside the scope of this book. Instead, by examining different national approaches to risk management, the research provides insight into the future development of chemical risk assessment and, at least, the potential for the harmonisation of risk assessment. Typically, EU decision-making on chemical risk management and risk communication follows complete evaluation of a risk assessment (Section 2.2.1). In combination with a substance-bysubstance (case-by-case) EU approach to chemical risk management, an apparent lack of investigation into the commonalities and differences between previous risk management decisions arises. The research project therefore shifts the current paradigm from risk assessment to risk management by examining which information and system parameters are necessary to achieve efficient and effective EU decision-making with regards to protecting human health and the environment from hazardous chemicals. From this perspective, it examines which information from risk assessments is necessary for developing and selecting different risk management options. Ironically, the European Commission recognises that a first step to chemical risk management requires evaluation if existing regulatory measures14 are sufficient to control an existing or newly identified chemical risk in an appropriate and acceptable way [28]. Very few studies have examined the efficiency and effectiveness of existing control measures, at least from the perspective of the appropriate choice of different risk management options. There is a general lack of transparency on the interaction between various EC legislative frameworks. Similarly, the effects on EU risk management decision-making processes of different implementation of European Commission environmental directives across Member States remain largely unexplored. Currently, differences between national approaches to chemical risk management appear to play little part in EU decision-making processes, even though they may prove crucial in reaching agreement on the appropriateness of any measure. 14
For example, classifying a chemical as ‘dangerous’ triggers several regulatory controls under existing environmental, occupational and consumer protection.
14
Introduction Regulatory chemical risk control measures can be broadly separated into four categories: command and control, economic, incentivebased, and voluntary initiatives [29]. Brief explanations and examples of these various control instruments are shown in Table 1.2 and are discussed in detail in the Literature Review (Section 2.3). Control option
Description
The most commonly applied control Command instrument. It operates on the basis and of statutes by regulatory authorities Control representing the State.
Examples Environmental and product standards, emission limits.
Economic
Regulators set prices or charges, which in turn affect customer choice in markets.
Levies, insurance premiums.
Incentivebased
Customers and consumers influence markets based on information provided directly in registers or as labels.
Pollution registers, eco-labels.
Voluntary initiatives
Self-imposed controls or regulatorynegotiated agreements brought into effect by companies producing or marketing the chemicals.
Voluntary agreements or programmes by industry.
Table 1.2 Chemical-control instruments under risk management decision-making (adapted from [30]) Reproduced with permission from P. Calow, Controlling Environmental Risks from Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997. ©1997, John Wiley
Judging from previous EU chemical risk management decision-making, the choice of control instrument, scope and detail of regulatory action proposed by a Member State depends on its national regulatory approach and administrative structure [31, 32]. This affects inclusion of different actors in decision-making
15
Framework for Chemical Risk Management under REACH [17], the choice of tools used to evaluate and control risks [33], the socio-economic effects of any regulatory decision [34], as well as the practicality and tolerability of implementing a risk management option [17]. This research project analyses the approaches that the subject Member States take while considering national obligations under EU and international law.
1.6 Rationale and Choice of Countries The start of the research project coincided with the European Commission White Paper for a Future EU Chemicals Strategy that initiated the legislative process to establish a system for REACH [20]. The new policy responds to a general lack of knowledge about the properties and uses of about 100,000 existing chemical substances15 [20]. Amounting to >99% of the total volume on the EU market, the current number of existing substances marketed in volumes of >1 tonne per year is estimated at 30,000 [20]. The current risk assessment and risk management processes for existing substances is considered slow and resource-intensive, so the new system will come into force on 1 June 2007 and take 11 years to implement [35, 36]. Reform of past EU chemicals policy will see new chemicals legislation in the form of a regulation that enables more uniform regimens than a directive [37], but differences between national chemicals polices and risk management approaches will have a major role in EU decisionmaking processes (Section 2.4). To meet the aims of the strategy within the proposed timelines, agreements between Member States on chemical risk management must be reached in far shorter timeframes than in the current system. The European Commission’s First Report on the Harmonisation of Risk Assessment Procedures recognises risk management as an important area of risk-related decision-making that would benefit 15
‘Existing’ substances are defined as those substances in use within the EU before September 1981 and listed in EINECS (European Inventory of Existing Commercial Chemical Substances). Accordingly, substances not listed in EINECS are regarded as ‘new’ substances and subject to a ‘notification’ procedure (Section 2.2.2).
16
Introduction from a more harmonised approach [38]. The European Chemical Industry Council’s (CEFIC) position states that coherence between the policies of Member States is necessary to provide a predictable framework for the risk management of chemicals in the EU [39]. This PhD research project responds to these needs by developing a framework for chemical risk management under REACH that tackles issues not sufficiently addressed in the legislation. Accounting for about 25% of the total value-added EU25 chemical production, Germany’s interests resound in EU chemicals policy [40], yet it is Sweden and the UK that have played pivotal parts in the formation of the new regulation. EU Ministerial debate on chemicals policy was initiated by the UK in 1998 [41]. The UK recognised that current chemical legislation is insufficient to cope with the lack of data on the hazard and use of most chemicals on the EU market. By this time, Sweden was recognised as having developed the most stringent chemicals legislation and the most far-reaching chemicals policy in the world [42]. Identifying a need for action at the EU level for effective chemical control [43], chemicals policy became a Swedish political priority during its 2001 presidency of the European Council of Ministers [41]. In what appears to have been a co-ordinated approach between the Swedish Minister of the Environment and the Environment Commissioner (who was also Swedish), a European Commission EU chemicals policy was published that year. Before EU enlargement in 2005, France, Germany, and the UK were the key Member States in EU decision-making because together they held most votes in the qualified majority voting along with Italy (10 each out of 87 for the EU15). In this scheme, Sweden only held four votes compared with a maximum 10, but its environmental approach is similar to other northern countries (Denmark, Finland, the Netherlands) which, together, represented considerable power in yielding a possible blocking minority [44]. In short, although the Member States investigated by this research project once represented about 40% of the EU15 votes in the Council assembly, their influence on chemicals policy through the majority voting rules was greater than represented by this percentage. Although EU enlargement significantly
17
Framework for Chemical Risk Management under REACH cut these vote percentages, the contribution to chemical production from recent-accession countries represents only 5% of the EU25 industry turnover [40]. It is therefore unlikely that these countries will play a major part in future EU risk management policy. The countries considered in the research project fall into one of each of the distinguishing ‘styles’ of environmental policy proposed by Andersen and Liefferink [47] and developed by [45]. According to this categorisation, Germany is an ‘established pioneer’ in EU environmental policy, and differs from Sweden which is a ‘newcomer pioneer’ [44]. France is considered to be a ‘fence sitter’ (along with Belgium, Italy and Luxembourg) that will occasionally push for fundamental changes in environment policy on specific issues and circumstances [45]. The UK’s reluctance to legislate on environmental matters and its opposition to stringent EU-wide regulation groups it with ‘laggards’ such as Spain, Portugal, Greece, and Ireland [45].
1.7 Conclusion and Research Questions To develop a framework for EU risk management, the research project examined national chemical risk management in France, Germany, Sweden, and the UK. The study started by investigating the social and cultural contexts within which existing chemicals legislation and policies have evolved, and the differences in the regulatory infrastructures of the four Member States. A comparative study of the national chemicals policies was analysed together with data gathered from interviews with national regulators and representatives of key stakeholder associations. The framework for risk management decision-making under REACH then evolved from the findings of the comparative analysis. Judging from previous EU risk management decision-making, the scope and detail of a regulatory action proposed by a Member State depends on its national regulatory approach and administrative structure [31, 32]. Previous research has identified that France, Germany, Sweden and the UK display contrasting ‘styles’ of
18
Introduction environmental policy. It remained to be seen if the national chemicals policies and regulatory practices reflect these different styles. To date, there is no other comparative analysis between any of the chemicals policies or regulatory administrations of these four countries. The success of a framework for risk management under REACH may depend on if the framework can reconcile differences in the national approaches. Investigating chemical risk management involves more than just an analysis of national policies and regulatory administrations. Developing, selecting and implementing risk reduction measures ultimately depend on the roles, responsibilities and resources of key stakeholder associations involved in national chemical risk management. Risk tolerability not only relates to the socio-economic effects of regulatory measures, but the perceived nature of risks and benefits. On 13 December 2006, the European Parliament adopted a legal text on REACH that has been agreed by the European Commission and Council of Ministers16 [35]. The REACH Regulation 1907/2006 was finalised on 18 December 2006 and entered into force on 1 June 2007 [46]. It took three years of legislative decision-making and more than five versions of the regulation for these three institutions to agree on the final text. What is clear is that the new EU Chemicals Policy and the resulting legislation focus on hazard assessment rather than risk management (Section 2.4). Completion of the PhD project coincided with the final procedures for legislative enactment during the end of 2006. Technical Guidance Documentation for regulators and industry was also anticipated during that time, produced by the European Commission through 14 REACH Implementation Projects (RIP). As of March 2009, some of these documents must be finalised and all guidance remains open to potential revision [47]. Further technical guidance and advice 16
For updates and easy-to-use links to the REACH legislation, readers can subscribe to the free REACHReady service offered by the UK Chemical Industries Association (www.reachready.co.uk).
19
Framework for Chemical Risk Management under REACH continues to be forthcoming from the new European Chemicals Agency (such as that following several recitals of the legal text17 -[46]). Certain aspects of the regulation are also subject to review by the European Commission during the first years of implementation, including how to carry out various aspects of safety assessments and how REACH can best interact with other EC legislative frameworks [46]. Therefore, the research was well-timed to maximise its potential contribution on how to implement REACH. The three research questions were: 1. How does chemical risk management differ between the Member States and why? 2. How do the different national approaches affect EU decision-making? 3. What framework for chemical risk management emerges from the research?
17
The following recitals of the legislation state that the European Chemicals Agency should provide guidance and advice with: (24) REACH Implementaion Projects and general ÌiV
V>Ê}Õ`>ViÆÊΣ®ÊvÕw}ÊÀiµÕÀiiÌÃÊvÀÊ«Ài«>À>ÌÃÆÊÎn®]Ê{ä®Ê>`Ê{Ç®Ê >Û`}Ê>>ÊÌiÃÌ}ÆÊÇn®Ê«ÀÌÃ>ÌÊvÊÃÕLÃÌ>ViÃÊÃÕLiVÌÊÌÊÕÌ
ÀÃ>ÌÆÊÈÓ®]Ê (95), (97) risk communication. The legal text also specifies that the Chemicals Agency should establish a Manual of Decisions and Opinions of a Member State regulatory committee responsible for interpretation and implementation of certain parts of the Evaluation and Authorisation process of the REACH Regulation.
20
2
Literature Review
- Logic The art of thinking and reasoning in strict accordance with the limitations and incapacities of the human misunderstanding (Ambrose Bierce, The Cynic’s Word Book, 1906)
Introduction This Chapter provides a basis for understanding how countries adopt different regulatory approaches and why the European Union (EU) needs policy reform by examining the complexity of the risks associated with chemical production and use. As one of Europe’s most internationally competitive and successful industries, the chemical industry accounts for 18% of value added in EU manufacturing and contributes 4% to total gross domestic product (GDP) [48]. More importantly, chemical products are a primary feedstock for all manufacturing because approximately two-thirds of chemicals produced in the EU are consumed within other industry sectors [49]. Given the international trade of chemicals, regulating the chemical industry in a single country has the potential to affect industry in other EU countries and in other manufacturing sectors. The same could be said about the effect on health and the environment resulting from regulating many sources of pollution (Section 1.2). Many of the chemical risks that society faces today are the consequence of the success of the chemical industry. The number and
21
Framework for Chemical Risk Management under REACH volume of chemicals produced by human activities across the world have increased exponentially since the industrial revolution1 [50, 51]. There is no atomic difference between a ‘naturally occurring’ or ‘man-made’ chemical2 [52], but humans and ecosystems are now being exposed to many chemicals at concentrations not experienced during 3.5 billion years of evolution [53]. The term ‘synthetic’ therefore refers to manufactured products, including bio-based materials. A synthetic chemical may sometimes be distinguishable from other substances by its concentration in the environment or an À}>ÃÆÊÊÌ
iÀÊV>ÃiÃÊÌÊ>ÞÊÌÊLiÊ`iÌw>LiÊ>}>ÃÌÊL>V}ÀÕ`Ê concentrations [54]. The extent to which synthetic chemicals may be present in human bodies and environmental media is therefore unknown3. One fact is certain: the entire human population is constantly and directly exposed to synthetic chemicals in consumer products and indirectly exposed via environmental media. In the workplace, 22% of the EU workforce regularly inhales fumes, dust or vapours, and 16% handle dangerous chemicals4 [55]. Every month research published in journals such as Environmental Health Perspectives and the Journal of Exposure Analysis and Environmental Epidemiology links chemical exposures to asthma, spontaneous abortion, stillbirth, congenital anomalies, childhood cancers, reduced values of intelligent quotient (IQ) and mental illness. Recent studies even suggest that exposure to some chemicals in females can have negative effects on offspring several generations later [56, 57]. The quantity of substances being manufactured is alarming because a sufficiently high exposure to a chemical can negatively affect the 1
In terms of numbers, this trend appears to be continuing, with an increase from 10 million to 27 million new substances registered in the Chemical Abstract Services between 1991 and 2005 (see [19, 58].
2
Many substances in plants and animals can be poisonous to other species, for instance ‘green potatoes’ can be toxic to fertility in humans [59].
3
A recent study of human umbilical cord blood in the USA detected 77% of 366 tested industrial and consumer synthetic chemicals [60].
4
Exposure to hazardous chemicals is estimated to be responsible for between 18–30% of recognised occupational diseases [61].
22
Literature Review functioning of an organism or an ecosystem5 [53, 62]. Growth of the chemical industry tends to be directly linked to production volumes [63–65], so increasing exposures to synthetic chemicals can result from the strong economic performance of the industry (see [66, 67]). Exposure levels will continue to rise for a substance that enters the environment or living organisms more rapidly than it is degraded or metabolised – concepts referred to as ‘assimilative capacity’ and ‘critical load’ (see [68]). Problematically, many substances are persistent in environmental systems or biological systems and can therefore become concentrated in certain media6 or organisms. Moreover, bioaccumulative substances become even more concentrated in some biological tissues and fluids, thereby accumulating through trophic food chains. Consequently, the concentration of a very persistent and very bioaccumulative (VPVB) substance in some wildlife can correlate with levels of industrial production7 (see [69, 70]). Although the number of chemicals meeting the VPVB criterion is unknown, young people today have more synthetic chemicals in their blood than their elders [71]. Scientific knowledge was once the limiting factor for identification and control of chemical risks, but a new paradigm faces modern society. Mankind’s overall ability to evaluate chemical risks appears to be exceeded by the number of chemicals produced and the complex myriad of exposure scenarios to multiple substances. Regulators are also faced with the question of how to divide responsibility for managing chemical risks between producers, downstream manufacturers, academia, governments and stakeholder organisations. In many cases, a single substance can be independently produced by several companies across the world and can provide a raw material to several hundred downstream users. The end-user of 5
Recent research indicates that exposures to certain non-hazardous substances can interact with organisms in such a way as to inhibit detoxification mechanisms, thereby increasing the effects of known pollutants [72].
6
Global climate change may cause further increases in the growing contamination of ecosystems in the Arctic region [73].
7
The specific example to which this refers is the concentration of penta-brominated diphenyl ether (penta-BDE) in Arctic beluga whales and ringed seals [69].
23
Framework for Chemical Risk Management under REACH a substance, preparation or an article will usually be a member of the general public. Any regulatory or corporate decision on chemical risk management will have knock-on effects through the manufacturing supply chains (Section 2.2.1). From past experience, chemicals regulation has caused some losses in the industrial competitiveness of EU firms [74] and their ability to innovate new products [75]. While the industry as a whole has recovered from previous regulatory effects [75], the EU chemical industry is now facing intense international competition8 [76]. Many EU companies may not be able to overcome future obstacles caused by EU regulation [77]. Public perception appears to be focused on the risks rather than the benefits of chemicals production. A pan-European survey found that only 50% of the general public viewed the chemical industry as beneficial to society, whereas 93% consider that chemicals negatively affect human health [78]. Rarely do public debates on chemicals policy discuss the benefits that chemical products provide to society or how to best devise regulation that supports the competitiveness of EU chemical producers. The media eye sees only chemical risks and regulation responds with knee-jerk reactions. Society is faced with complex issues on the sustainability of the chemical industry, not just in terms of maintaining international competitiveness but maximising the potential for the application of innovative chemistry. In this respect, the chemical industry should not be viewed as a producer of chemical risks or as a product manufacturer. The chemical industry is creative and innovative. It applies the most upto-date science and technology to meet societal needs and demands. 8
The value of non-Organisation for Economic Co-operation and Development (OECD) chemical production is expected to almost triple by 2020, while in OECD countries it will increase by about only 60% [76]. Western Europe is predicted to experience a large reduction in growth rates when comparing 1990–2000 with predictions for 2000–2010 [76]. The shift towards service-based industries in Western countries at the expense of sectors such as agriculture, manufacturing and durable goods, further implies a move away from chemicals [76].
24
Literature Review Chemical companies hold a key position in manufacturing supply chains for generating and disseminating scientific knowledge, including how to use chemicals safely [79-82].
2.1
Chemical Risks
2.1.1
Hazard and Exposure
Sources of chemical exposure from human activities arise from the production, use and disposal of chemicals (Figure 2.1). Chemical exposure can result from direct contact with a chemical, from a chemical being present in a local surrounding, or from the chemical being transported through the wider environment. Whether in organisms or the environment, biochemical and physical processes transform the molecular structures of substances.
Figure 2.1 Pathways of exposure to chemicals [83] (adapted from [84])
25
Framework for Chemical Risk Management under REACH Exposure describes the integral of a concentration of a synthetic substance in a biological or environmental system resulting from a particular use (or combination of uses) of a chemical or group of chemicals over a specified period of time. Exposure is just one aspect vÊÀÃÆÊÜ
iÌ
iÀÊ>ÊÃÕLÃÌ>ViÊV>ÊV>ÕÃiÊ
>ÀÊÌÊ>Ê
Õ>]ÊÀ}>ÃÊÀÊ ecosystem also depends on hazard. Whereas exposure is stochastic, hazard describes intrinsic chemical properties. Consideration of exposure is critical to assessing and understanding hazard. During risk assessment, when a hazardous property is identified, exposure is used to interpret hazard assessment results in terms of what hazardous properties may exert an effect given the route and duration of exposure, as well as the particular types of biological or ecotoxicological systems involved. It should be possible to determine the hazardous properties of a substance from its molecular structure. The properties of pharmaceuticals are almost always first predicted using computational techniques such as Quantitative Structure–Activity Relationships9 (QSAR) before further product development [39]. Most industrial chemicals have been produced before these in-silico tools were available or readily accessible10. Of course, our current knowledge and understanding of science, let alone that of a risk assessor, also limits the application of such methods. The first step of a hazard assessment involves identifying an enzyme, tissue or organ with which a given substances will interact. Because many changes in biological systems and organisms can be caused by exposure to a given chemical, risk assessments target the most (potentially) significant effects. For environmental risk assessments, 9
A chemical structure can be used to predict its interaction with environmental media and biological systems. Even simple physical chemistry can accurately predict the persistent and bioaccumulative properties of certain types of chemical [86]. Equally, certain hazardous properties can be qualitatively predicted by comparing a molecular structure with similar substances with known hazard profiles [87]. Many developments have been achieved using actual computational techniques (e.g., QSAR) to quantitatively evaluate the toxic properties of substances, particularly pharmaceuticals (see [88, 89]).
10
Computer programmes have recently become available to companies to carry out such exercises at relatively low cost (e.g., [87, 90]).
26
Literature Review species that can alter the entire functioning of an ecosystem must be identified, and are referred to as the keystone species. Hazard assessments are complicated by biological respoNSE being dependent on physical environment (temperature, pressure, nutrition). Organisms also evolve while adapting to different environments. Very little is known about the structure and functioning of ‘normal ecosystems’ (if indeed there is such a phenomenon). The same could be same for many aspects of the biology and behaviour of humans. In practice, identification and evaluation of hazardous chemicals involves the piecing together of data from several sources. From industrial explosions to household spills, information reported in poison centers or monitored by health officials provides a key starting point for identifying hazards [85]. A hazard assessment usually comprises two interlinking and iterative processes: hazard identification and hazard characterisation. Hazard identification can involve specific isolated test systems (e.g., ability of a chemical to interact with deoxyribonucleic acid (DNA) in a test tube). During the characterisation stage of a hazard assessment, the ability of the chemical properties of the substance to damage humans and organisms or contaminate environmental media is evaluated. For instance, changes to a substance during interactions with cellular proteins or as a result of metabolism can alter DNA-binding potential. Without such considerations, hazard identification using limited test parameters can yield only partial conclusions on the potential hazardous properties of a substance. From a review of the literature, there does not appear to be a clear or systematic approach to differentiating between hazard assessments that are primarily based on hazard identification and those that include hazard characterisation. This appears to create obstacles in communicating the results of hazard assessment study between scientists, risk managers, policy decision-makers, as well as members of the general public. In practice, while hazard identification forms the basis of hazard assessment, hazard characterisation serves in analysing data when establishing whether a substance is ‘dangerous’ or 27
Framework for Chemical Risk Management under REACH ‘non-dangerous’11. For classification and labelling, elements of exposure guide hazard characterisation and enter the decisionmaking process by consideration of ‘normal handling and use’. Sixtyeight specific types of chemical hazard have been identified at the international level, each corresponding to a particular Risk Phrase (see Appendix 2.1). Risk Phrases can be used in particular combination, such as R14/15 which signifies that the substance reacts violently with water, creating extremely flammable gas. Several further Risk Phrases have been proposed (e.g., R320: may be harmful by inhalation after vÀiµÕiÌÞÊÀi«i>Ìi`ÊiÝ«ÃÕÀiÆÊ>`Ê,Î{ä\ÊÃiÊÀÃÊvÊV>ViÀÊV>ÌÊ be excluded after frequently repeated exposure). Risk Phrases, and the corresponding Safety Phrases that detail which protective measures should be taken (e.g., use of gloves), will soon be replaced by Hazard Statements and Precautionary Statements under a new universal system of product labelling: the Globally Harmonised System (GHS). Under GHS, eight categories distinguish physical hazards and nine categories broadly divide health and environmental hazards (Table 2.1). Physical hazards
Health and environmental hazards
UÊ Ý«ÃÛiÃ
UÊVÕÌiÊÌÝVÌÞ
UÊ>>Li
UÊ ÀÀÃÛiÉÀÀÌ>Ì}ÊÌÊÃ
UÊ"Ý`ÃiÀ
UÊ-iÀÕÃÊ`>>}iÊÌÊÌ
iÊiÞiÉÀÀÌ>ÌÊÌÊiÞi
UÊ-ivÀi>VÌÛi
UÊÕÌ>}iVÊÌÊ}iÀÊViÃ
UÊ*ÞÀ«
ÀV
UÊ >ÀV}iV
UÊ-iv
i>Ì}
UÊ,i«À`ÕVÌÛiÊÌÝVÌÞ
UÊ"À}>VÊ«iÀÝ`iÃ
UÊ/>À}iÌÊÀ}>ÊÃÞÃÌiVÊÌÝVÌÞ
UÊ ÀÀÃÛiÊÌÊiÌ>Ã
UÊ,iëÀ>ÌÀÞÊÀÊÃÊÃiÃÌÃ>Ì UÊ>â>À`ÕÃÊÌÊÌ
iÊ>µÕ>ÌVÊiÛÀiÌ
Table 2.1 Hazard categories under the Globally Harmonised System [91] 11
This book uses the term ‘dangerous’ but this term will be changed in the REACH Regulation to ‘hazardous’ by the Classification, Labelling and Packaging Regulation 1272/2008 in December 2010.
28
Literature Review Although the GHS system is unlikely to come into force in the EU until after 200812, at which point it will be progressively phased-in, the general principles for classification and labelling follow the current EU scheme. Under GHS, each hazard category shown in Table 2.1 can be subdivided in various ways, for instance: UÊ ¼VÕÌiÊ ÌÝVÌÞ½Ê V>Ê LiÊ Ãi«>À>Ìi`Ê >VVÀ`}Ê ÌÊ iÝ«ÃÕÀiÊ ÀÕÌiÊ i°}°]ÊÀ>ÊÛiÀÃÕÃÊ`iÀ>®ÆÊ UÊ ¼,i«À`ÕVÌÛiÊÌÝVÌÞ½ÊVÕ`iÃÊ>ÊÃÕLV>Ìi}ÀÞÊvʼivviVÌÊ`ÕÀ}Ê >VÌ>Ì½Æ UÊ ¼/>À}iÌÊÀ}>ÊÃÞÃÌiVÊÌÝVÌÞ½ÊÃÊÃÕL`Û`i`Ê>VVÀ`}ÊÌÊÃ}iÊ or repeat dose. A numbering scheme then distinguishes between ‘levels of hazard’ under each category (e.g., ‘1’ is very high toxicity and ‘4’ is low toxicity). For hazard communication 13, five pictographic labels represent physico-chemical data (gas under pressure, oxidising, y>>Li]Ê VÀÀÃÛi]Ê >`Ê iÝ«ÃÛi®ÆÊ >Ì
iÀÊ vÕÀÊ >ÀiÊ ÕÃi`Ê ÌÊ symbolise the various toxicological categories14 (Figure 2.2). Identifying a hazard is only a small part of the risk assessment process. Hazard must be differentiated from risk. Assessing risk involves an analysis of the likelihood that adverse effects to human health or the environment after exposure to a chemical may occur. For risk management, exposure assessments therefore play equal (if not more) important parts as evaluations of hazard. The following sections discuss how toxicology, exposure assessments, and risk characterisations contribute to the central scientific definition of risk as probability versus consequence [93-95]. 12
Some countries outside the EU have begun implementing parts of GHS.
13
Transport of dangerous products uses a separate set of pictogrammes.
14
Two pictogrammes were added in 2005 under GHS for labelling: ‘chronic toxicity’ and ‘harmful/irritant’ [91]. Chronic toxicity and other harmful properties were previously }ÀÕ«i`ÊÜÌ
ÊÌ
iʼÌÝV½Ê>`ʼÛiÀÞÊÌÝV½ÊÕ`iÀÊÃÕ>`VÀÃÃLiÃÊ>LiÆÊLivÀiÊ-]Ê a harmful or irritant substance was indicated by a St Andrew’s Cross.
29
Framework for Chemical Risk Management under REACH
Under Pressure/ Self-reactive/ Liquefied gas/ Explosive Dissolved gas
!
Harmful/ Sensitiser/ Irritant
Self-reactive/ Oxidising Pyrophoric/ Flammable/ Releases flammable gases
Acutely toxic
Aspiration hazard/ Respiration sensitiser/ Mutagen/Genotoxin/ Carcinogen/Reprotoxin/ Target organ toxicity/ Chronic toxicity
Corrosive/ Irritant
Hazardous to the aquatic environment
Figure 2.2 ‘Pictograms’ for labelling ‘hazardous’ chemicals [92] Reproduced with permission from Globally Harmonised System of Classification and Labelling of Chmeicals, UNECE. ©2005, UNECE
2.1.2 Toxicology Toxicology is the study of biological effects in organisms after chemical exposures. For the purposes of risk management, one of the most important objectives of toxicology is to determine an exposure (‘dose’) expected to have no negative effect on the functioning or health of the test system. In the EU, this dose is referred to as a No Observable (Adverse) Effect Level (NO(A)EL). Because a cell, tissue, organism or ecosystem often returns to a pre-intervention state of functioning after the removal of a test substance, the term ‘adverse’ relates to if a detectable effect appears ‘irreversible’15. Identifying the Lethal Dose for 50% of a 15
‘Irreversible’ therefore refers to damage to a cell, tissue or organ for which the natural defense or repair mechanism of the body (i.e., healing processes) cannot recover [104]. For instance, some cancers are reversible [52].
30
Literature Review test population (LD50 test) is rather archaic for testing toxicity in higher-level organisms [96, 97]. Modern technologies enable a wide variety of biological respoNSE to be reliably measured, not just death. Cell, organelle, enzyme, nucleic acid or other types of bioassays can avoid the testing on multicellular organisms or ecosystems. Such testing, known as in vitro methods, reduces the need for animal experimentation16. In vitro tests yield only a partial understanding of the hazardous properties of substances because they lack many of the complex interactions between cells and tissues present within an entire organism [98, 99]. Toxicologists therefore argue that some animal testing is required to fully assess a hazard, even though it is more costly [98]. Evidence suggests that scientists and regulators over-rely on animal studies as a result of familiarity with animal test systems rather than a proven scientific basis [100-102]. Dependence on animal testing also arises as the result of their prominence in current internationally accepted Organisation for Economic Co-operation and Development (OECD) hazard assessment methodologies [103]. Whether the test system involves a bioassay, organism, or ecosystem, the shape of a dose–response curve must be experimentally extrapolated. Acute testing often involves a ‘one-off’ dose but can be extended over periods of time (usually 14–28 days) to begin investigating the effects of a repeat dose [106]. Sub-chronic testing further extends the dosing and observation period to 28 days or 90 days [107]. Chronic testing usually refers to longer periods, from 12 months to the lifetime of an animal [97]. Compared with acute testing, sub-chronic and chronic tests enable more subtle effects at lower doses to be identified such as carcinogenic, mutagenic, 16
This is currently a hot topic in the EU, evident by the recent ban on the testing of cosmetic products on animals under the EU Cosmetics Directive [105]. In vitro cell cultures usually originate from a live animal so, ultimately, some animal sacrifice is necessary. Some synthetic systems do exist, such as semi-permeable membranes to model tissue passive transport.
31
Framework for Chemical Risk Management under REACH (b)
(a)
(c)
NOELs
NOELs NOEL
0
0
0 D os e
D os e
D os e
Figure 2.3 Dose–response curves showing NOELs (a) ‘S’ (b) ‘U’ (c) ‘U’ or reprotoxic17 properties (CMR) of substances18 [108]. Chronic testing regimens frequently use very high doses which result in pronounced effects that may not be evident at lower levels [62]. The pronounced effects observed during high-dosage testing may mask effects caused by low-dose chronic exposures (disruption to the endocrine system). Chronic NO(A)EL are usually extrapolated from sub-chronic or acute NO(A)EL due to the time and costs associated with laboratory testing [111]. Chronic effects will also assume thresholds expressed as a NO(A) EL unless a genotoxic or mutagenic ‘non-threshold’ mechanism can be demonstrated [108]. With the exception of linear dose–probability plots for non-threshold effects [112] and linear respoNSE for most non-genotoxic carcinogens, dose–response curves will usually show the ‘S’-shape in Figure 2.4a [113, 114]. Some toxins prove to be less toxic at intermediate doses than at low doses by triggering defensive or immunological respoNSE in the body19 : this is known as homeopathy. 17
As well as the potential effects on fertility and genetic alterations in chromosomes caused by exposure, reproductive toxicity can (but does not always) include tests observing the developmental characteristics of progeny [109, 110].
18
CMR substances are divided into three categories according to the level of evidence of the substance: category 1 (considerable evidence), category 2 (some evidence) and 3 (limited evidence).
19
For instance, by increasing the number of enzymes needed to metabolise the toxic substance [120].
32
Literature Review Exposure to multiple substances can also affect the dose–response curve through antagonistic, additive [115, 116] or synergistic mechanisms. Dose–response curves can therefore show a ‘U’ shape or inverted ‘U’ shape (Figures 2.4b and c) [117, 118]. Two NO(A)EL values may then need to be used during risk management. Results of in vitro and in vivo tests feed into toxicokinetic and toxicodynamic models. Toxicologists use absorption, distribution, metabolism, excretion and toxicology (ADMET) programmes to compute flows of a substance through an organism, as well as the toxicity of metabolites e.g., [119]. The amount of a substance present at a given endpoint – i.e., biological target – is calculated according to the route of exposure (dermal, inhalation or ingestion). By considering genetic differences, toxicodynamic models investigate if the effects in one cell, tissue, organism or species are relevant to others, which also serve to identify susceptible individuals within a population or subpopulation. With the rapid progress being made in understanding and mapping genomic sequences, major developments in risk assessment are expected with the development of toxicogenomics [121, 122]. It is not unforeseeable that in the near future the biological response of specific individuals to certain substances may be identified without the need for further testing. A person may be genetically screened to identify his/her probability of developing certain diseases. Many ethical questions surrounding genetic screening of human populations remain to be debated. Political and scientific attention has recently turned towards investigating substances that possess hormonal activity or alter patterns of hormone effects. Not only does detecting such changes in organisms prove to be extremely complicated, these endocrine disruptors can be active at very low doses. Considerable work for the standardisation of endocrine disruptor test methods has been undertaken at EU and international levels, but there is currently little agreement on how to test for such effects [123]. 33
Framework for Chemical Risk Management under REACH Ultimately a ‘derived no-effect level’ (DNEL) (in humans) or a ‘predicted no-effect concentration’ (PNEC)20 (in ecosystems) is calculated for a substance, group of substances or chemical mixture. Assessment factors (AF) – sometimes referred to as ‘uncertainty factors’ or ‘safety factors’ – compensate for lack of data and assumptions resulting from dose spacing and other test model parameters (adapted from [124]): DNEL or PNEC = NO(A)EL/(AFi AFii AFd AFl AFs AFu AFq) AFi accounts for variations between species AFii accounts for variations within the same specie AFd compensates for dose spacing AFl applies if a NO(A)EL cannot be established from the data set due to observation of an effect at the lowest dose, in which case a ‘lowest observable (adverse) effect level’ value is used AFs applies when a chronic or sub-chronic effect is estimated from shorter term sub-chronic or acute effect AFu accounts for incomplete test results or accounts for results from uncertainty analysis AFq considers the quality of the data or relevance of test data Regulators use DNEL and PNEC to set health or environmental standards, but usually in conjunction only with exposure levels. For non-threshold effects, when a DNEL or PNEC may not apply, probabilities of the incidence of an effect for an individual within a population are used to set regulatory limits on exposure. Recent regulatory discussions within the REACH Implementation Projects 20
For ecosystems, the collection of threshold values for NO(A)EL for several species is used to determine a ‘predicted no-effect concentration’ (PNEC) for each environmental compartment. Environmental monitoring often observes effects that different environmental concentrations have on various ecosystem sub-groups. Laboratory multi-species test systems are available, but are limited in number and rarely appear in EU chemical risk assessments (see [125]).
34
Literature Review (RIP) suggest that factors representing these probabilities be incorporated into future assessments as a ‘derived minimal-effect level’ (DMEL) for human health [126]. A combination of hazard and exposure is used to prioritise relevant regulatory activities according to risk because setting standards for health and the environment requires time and resources. For instance, setting standards to control high exposures to a substance with a low DNEL may take precedence over regulating a very low exposure to a substance with a marginally higher DNEL.
2.1.3 Exposure Assessment As with toxicology, the most accurate exposure assessments often result from physical and/or biological measurements rather than computational predictions. In general, exposure assessments are more heavily reliant on modelling than their toxicological counterparts. Most models are based on stock-flow approaches (mass balances) where a quantity of chemical (stock) flows at predicted or measured rates. For instance, human exposure to a chemical being used in a room can be estimated according to size of room, temperature, rate of airflow and the volatility of the chemical. Exposure will also depend on the time a person spends in the room and the metabolic state of the person (e.g., if the person is carrying out exercises). Concentrations will depend on if relevant protective protocols are being followed, such as operating local exhaust ventilation or wearing personal protective equipment (PPE). In general, calculations of industrial emissions rely on default values according to the industrial sector, emission rates, wind velocity and direction, anticipated substance flows through the environment, abatement technologies and wastewater treatment processes (see [114]). Site-specific assessments and local environmental exposure assessments must also account for geographic variability caused by climate, hydrology, geology, and biotic conditions [115].
35
Framework for Chemical Risk Management under REACH Environmental exposures form the basis for determining indirect exposures to the general public that will usually occur during a lifetime. Direct consumer exposure assessments prove equally challenging, and may vary from acute to chronic exposure scenarios. Ideally the data set for consumer exposure from a substance in a product should include [127]: (a) Contact data
(b) Concentration data
(c) Product use data
- frequency of product use
- weight fraction of substance in the product
- intended use of product
- duration of use per event - site of use (e.g., size of room) - air exchange rate
- concentration of substance in the products as used, after dilution or evaporation
- amount used per event - contact surface (if appropriate) - physical form of the product (e.g., aerosol, powder, liquid, gas)
Although there is large inter-individual variation in frequency, duration and amount of product used, consumers tend to follow certain routines [128]. The United States (US) National Human Activity Pattern Survey has been gathering important information for determining exposure to environmental pollutants by collecting data on time–activity patterns for various exposure scenarios [129]. In 2005, a similar project began in the EU that focusses on types of product use according to exposure scenario [130]. The residues and metabolites of a substance can be measured in an organism or an environmental medium. Alternatively, biological effects known as biomarkers that are known to be the result of exposure to a hazard can be used to determine exposure levels [131]. In some cases, monitoring biomarkers in employees (e.g., metabolites in urine) can prove cheaper than measuring airborne concentrations of a substance in the workplace [132].
36
Literature Review Many models are available for calculating exposure, but the European Union System for the Evaluation of Substances (EUSES) is the most commonly used in the EU. Variations in human populations across Member States are considered in terms of body weight, diet, and activities [133]. Consideration is also given to susceptible individuals such as children and the elderly [133]. More specific models are used in conjunction with EUSES to assess occupational dermal exposure (DERMAL), occupational inhalation (EASE) and consumer exposure (CONSEXPO) (see [134]). The models used to conduct EU risk assessments are constantly subject to validation studies and are periodically improved. Four major limitations must be addressed: UÊ *Ài`VÌÃÊvÀÊÛiÀÞÊ«iÀÃÃÌiÌÊ>`ÊÛiÀÞÊL>VVÕÕ>ÌÛiÊ6*6 ®Ê substances using physico-chemical parameters and the EUSES model often result in false-positive values [136]. UÊ /
iÊ - Ê `iÊ ÃÊ ÌÊ ÛiÀÞÊ >VVÕÀ>ÌiÊ vÀÊ >ÃÃiÃÃ}Ê iÝ«ÃÕÀiÃÊ from many open uses in the workplace [137]. UÊ ÕÀÌ
iÀÊ `>Ì>Ê Ê Ì
iÊ «iÀi>LÌÞÊ vÊ ** Ê ÃÊ ii`i`Ê vÀÊ Ì
iÊ DERMAL model [134]. UÊ /
iÊ " - 8*"Ê `iÊ vÀÊ «Ài`VÌ}Ê VÃÕiÀÊ iÝ«ÃÕÀiÊ ÃÊ limited by a lack of data on chemicals in products [138]. With regard to the last point, a recent study on the risk assessment of chemicals in toys reports that [139]: ‘The crucial parameter, for which we have too little information for a sound estimate [on human exposure], is usually the factor that describes the migration of the substance under investigation from the product. This migration parameter is not known for mouthing, skin contact with solid products and eye contact for almost all substance/ material combinations.’
37
Framework for Chemical Risk Management under REACH The standardisation of test methods for the release of chemicals from articles is particularly complex because it depends on many factors relating to the production and design of a product, which will vary across a single category of article [136, 140]. Altogether, assessing risks from articles presents companies, regulators and stakeholder groups with an extremely difficult challenge.
2.1.4 Risk Characterisation A central definition of risk as probability versus consequence pervades the literature [93, 141]. Although models such as EASE (dermal exposure) and CONSEXPO (consumer exposure) use statistical methods to express probabilities associated with risk, EU risk assessment results are usually reduced to single ‘best-estimate’ figures. Numerical or nominal results for an assessment can be deceiving. A 10–6 risk of an individual developing cancer during use of a product can represent 0.001% or 10% of a total population developing cancer. The percentage of the affected population depends on the number of persons exposed to the product. Terms often used in risk assessment such as ‘average resident’ or ‘safe’ can exclude certain groups of individuals or exposures. For instance, a domestic cleaner will be using a household furniture polish more than most members of the public, and a child is more likely to put a common household object in his/her mouth than an adult. Subsequently, different articulations of risk frequently raise confusion among risk assessors, risk managers, stakeholder representatives, and the general public [142]. In particular, there is a need to improve how uncertainties in a risk assessment are communicated between regulators and stakeholders [143-145]. Chemical risks are measured by comparing the predicted or measured exposure to the DNEL or PNEC. In risk assessment, the extent to which an exposure exceeds a DNEL or PNEC determines if a risk is deemed ‘high’ or ‘very high’. If an exposure is considerably less than the ‘no-effect’ level (e.g., a factor of q1000 when compared with a NOEL), the risk is generally considered ‘insignificant’ (see [146]).
38
Literature Review A ‘level of safety’ that represents the difference between the two scores is used by regulators when evaluating the efficacy of different risk management control measures and should be distinguished from ‘assessment factors’ used in determining a DNEL or PNEC21 (Section 2.1.3). For non-threshold effects, such as genotoxic carcinogenity, the probability of incidence of an effect is calculated according to the predicted exposure. Regulatory standards are then based according to levels of probability, e.g., a 10–6 chance of developing cancer ÛiÀÊ>ÊviÌiÊ>ÌÊ>Ê}ÛiÊiÝ«ÃÕÀiÊiÛiÆÊ>ÃÊÌi`ÊÊSection 2.1.3 REACH may incorporate this probability into a DMEL22 [126]. While analysing uncertainty and probability is a process of risk assessment, controlling exposure levels relative to a DNEL, DMEL or PNEC is a process of risk management23 [147]. As with any discipline, bias can enter the risk assessment process. Scientific results indicating a danger tend to elicit further testing whereas negative results are more often trusted [148]. Confidence in results indicating a potential health risk can also increase with increasing evidence of a danger [148]. It may be more important to avoid false-negatives otherwise exposure limits may be set too high or dangerous chemicals may not be identified [95]. False-positives can lead to unnecessary and unjustifiable burdens on industry. Bias can also arise from unethical research practices caused by financial or personal pressures (e.g., [149, 150]. Trying to identify or prove such bias has political and legal consequences. Risk assessment is primarily limited by costs rather than science. The cost of meeting the highest-level of risk assessment testing for a single 21
The term ‘margin of safety’ is sometimes used to compare exposure levels to NO(A) EL and then this margin is compared with assessment factors. DNEL already include these assessment factors in the equation, so this thesis uses the term ‘level of safety’.
22
For non-threshold effects, the term ‘margin of exposure’ rather than ‘margin of safety’ is sometimes used if assessment factors are excluded from the assessment and a unit descriptor system used to characterise the effect.
23
Risk management must also account for how to manage exposure levels when there is no DNEL, DMEL or PNEC available, how to manage exposure levels based on underlying assumptions made in the risk assessment, and how to prioritise management activities.
39
Framework for Chemical Risk Management under REACH substance under current legislation is >€ 1 million [151]. Performing in-depth testing for 100,000 existing substance is unfeasible, let alone assessing the thousands of billions of potential combinations of these substances. Regardless of financial costs, the availability of expert and laboratory resources would prove to be an equally (if not greater) limiting factor. Companies and regulators must therefore balance the level of information necessary to control a risk against the cost of generating those data. New directions in environmental risk assessment seek to meet these challenges through direct toxicity testing and monitoring ecological status [152]. The former involves testing exposures >ÃÊ ÝÌÕÀiÃÊ À>Ì
iÀÊ Ì
>Ê `Û`Õ>Ê ÃÕLÃÌ>ViÃÆÊ Ì
iÊ >ÌÌiÀÊ ÀiµÕÀiÃÊ detecting biophysical changes (e.g., biomarkers) in the environment or in humans. The cause of an observed effect (e.g., specific substance or mixture) is then identified by a process of elimination. Research is increasingly directed toward identifying and evaluating synthetic chemicals present in humans. This responds to three major issues facing risk assessors and risk managers. The first concerns testing for developmental effects in animals or humans, which cannot be identified when similar doses are administered to an adult [153]. The second issue relates to endocrine disruption, which requires measuring behavioural changes. Even when an effect is identified, establishing a causal link to an exposure is difficult. For instance, an organism may be responding to any poorly controlled independent test variable, such as the physical handling of the test system by the toxicologist. The phenomenon of ‘chemical intolerance’ is perplexing many toxicologists. A significant proportion of the general public appears to be experiencing effects from exposures to substances well-below any known or identified DNEL [154, 155]. It has been proposed that the intolerance develops via a mechanism similar to sensitisation, where a previously high exposure to an irritant triggers an immunological response upon subsequent exposure to the same substance at a lower dose [53]. Perplexingly, unlike sensitisation, chemicals that are not similar in structure can elicit the biological response [156]. 40
Literature Review As with the ‘placebo’ effect, the inter-relation between psychology and biological response may be responsible for what may be best described as ‘chemophobic’ respoNSE. The most recent ‘emerging’ form of chemical risk is from ‘nanotechnology’. It is a field in its own right, for which there is comparatively little experience with conducting risk assessments. It is therefore mostly outside the scope of this research project, and is considered by European regulators as a separate field of regulation to chemicals policy e.g., [157]. Nanotechnology comprises a single chemical molecule and/or a group of molecules assembled in structures that significantly differ from most molecular and/or solid shapes that may not be synthetically possible using environmental or biological systems. For example, one category of a ‘nanotube’ is the carbon ‘onion’ which contains concentric molecular carbon ‘buckyball’ spheres e.g., [158]. In terms of risk assessment, these molecules and materials present new challenges to toxicology, from novel surface reaction chemistry to unpredictable pulmonary particulate phagocytosis [53]. Some nanoparticles are as simple as micelles formed by fatty acids in aqueous environments, such as those found in milk.
2.2 Obstacles to Risk Management 2.2.1 Structure of Risk Management The assessment, management and communication of risk typically forms three separate (but interlinked) processes (Figure 2.4a) [159, 160]. Certain aspects of risk assessment are purely scientific (e.g., physical chemistry tests) and some aspects of risk management are only technical (e.g., product standards), but overlaps do occur (Figure 2.4b). For instance, risk management options generally determine the depth and scope of the risk assessment24 [161]. 24
For instance, a risk management decision may be based on hazardous properties regardless of exposure levels [162].
41
Framework for Chemical Risk Management under REACH
Risk Assessment
Risk Management
Risk Communication
(a) Traditional
Risk Assessment
Risk Risk Management Communication
(b) Integrated
Figure 2.4 Approaches to risk analysis An integrated approach to EU chemicals control is hindered by a linear decision-making structure. Regulatory risk management follows only the result of a hazard assessment25 [115, 163]. Decisionmaking also operates on a substance-by-substance, case-by-case basis. There are few examples of EU risk assessments or risk management measures evaluating substances according to hazard or exposure groupings even though this strategy is recommended by the OECD. The substance-by-substance approach also creates a barrier to examining potential synergistic effects that can result from the combined exposure to multiple substances (see Section 2.1.2). One reason for the linear substance-by-substance approach is that current EU risk assessment requirements depend on the amount of a substance produced or imported. As the tonnage level increases, so does the number of toxicological tests. The rationale, advantages and disadvantages26 for ‘tonnage triggers’ can be summarised as follows [164]: 25
EU chemical risk assessment and management decision-making processes appear to be separated into five or more Technical Committees and Working Groups attended by different regulators and overseen by at least three Directorate-Generals of the European Commission.
26
A more detailed criticism of the use of tonnage triggers is provided by the Royal Commission on Environmental Pollution (RCEP) in its publication Chemicals in Products [166].
42
Literature Review ‘The general assumption behind ‘tonnage triggers’ is that on average, the higher the tonnage, the higher the potential exposure(s) and eventually the higher the potential risk(s). The advantage with tonnage triggers is that it is a very simple and transparent methodology, which can be easily implemented. A drawback is that the actual risk of a hazardous substance is strictly dependent on the actual use and resulting exposure.’ The tonnage system originates from the 1990 OECD programme for the risk assessment of high production volume (HPV) chemicals (see [165]). At the international level, a tonnage system proved necessary to co-ordinate the generation and exchange of hazard data on existing substances. At the EU or national level, the use of the tonnage-based criterion is a subject of significant contention [165]. With regards to risk communication, there is little agreement on how to structure stakeholder participation in EU decision-making processes (see [143]). Regulatory decision-making becomes further complicated by different articulations and communications of risk [145, 147, 167]. Scientific interpretations of risk do not account for the value judgments that enter decision-making processes27 [168]. Adopting a concept of ‘safety’ may therefore facilitate the comparison of risk management options. It can avoid: (i) a reliance on purely scientific definition of risk, and (ii) a need for reaching agreement on hazard assessments before enacting regulatory controls. Safety can be considered as a measure of protection versus prevention. Protection aims to ensure that risk management measures are being followed (e.g., wearing the correct type of glove when handling a chemical). Prevention involves redesigning or organising industrial or societal activities to avoid certain exposure scenarios. Implementing a protective or preventive measure depends on levels of knowledge and safety communication according to industry sub-sectors or chemical use, rather than substance-specific risk assessment results. 27
To illustrate this fact, one simply needs to ask, ‘to what extent is the consequence of a child developing cancer more severe than that for an elderly person?’ or ‘how probable is exposure to a chemical product for a consumer who cannot read instructions?’ ÃÜiÀ}Ê ÃÕV
Ê µÕiÃÌÃÊ ÛÛiÃÊ VÃ`iÀ}Ê >ÞÊ ÃViVVÊ Û>À>LiÃÆÊ Ì
iÊ interpretation of data may vary significantly between individuals or societal groups.
43
Framework for Chemical Risk Management under REACH
2.2.2 Existing Chemicals Regulation In the 1980s, increased concern regarding the effects of chemicals led to an amendment to the European Commission classification and labelling Directive 67/548 that requires risk assessments of substances on the EU market. Previously, classification and labelling were based on existing data sets available to a company or in the literature, or voluntary testing of chemicals by companies. At that time, it was not possible to set legislation for the evaluation of all 100,000 existing substances that represent 99% of the total volume of chemicals on the EU market [20] due to their sheer number. As explained in the Introduction, a distinction was therefore made between substances marketed after 1981 as ‘new’ and all other substances as ‘existing’. Separate legislation, Regulation 793/93, required the detailed testing of existing substances to follow the OECD tonnage system. This meant that substances produced to a level of >1000 tonnes per year were to be evaluated first, followed by substances produced to a level of >100 tonnes per year. A ‘new’ substance would require a complete risk assessment submitted to a European Member State regulator before being placed on the EU market. By 2006, only 72 existing substances had completed the risk assessment process under Regulation 793/93 compared with about 4,300 that have been notified with test data as new substances [47]. This does not imply that these substances are not regulated or that ‘no data’ are available. Companies are required to classify all chemicals on the market based on available data [169]. This includes reviewing references available in academic articles and epidemiological studies. To assist companies in this task, an international chemical database was set up by the OECD that is frequently updated by regulators and companies across the world [166]. No ‘official’ data (20% by number)
Limited data (45%)
Base set (20%)
Level 1 (10%)
Level 2 (5%)
Figure 2.5 Available data for existing substances as percentages based on [83, 186] 44
Literature Review Based on recent surveys of publicly available existing data sets, approximately 35% of existing substances produced have at least base-set level testing (see Figure 2.5) [77, 170]28. According to these sources, some data are available to regulators and stakeholders on a further 45% of existing substances. Nevertheless, it is anticipated that many hazardous substances have not yet been classified and labelled, and that 40% of the existing substances have some form of hazardous property [171]. The percentages in Figure 2.5 may be misleading because they refer only to ‘publicly available’ data. For instance, downstream users in the detergents sector possess considerable information on exposure [172]. Evidence also indicates that in vitro mutagenicity tests are better at predicting cancer than high-dosage chronic animal testing studies [173]. This exemplifies a primary argument against a reliance on animal testing. The figures on ‘existing data’ do not account for the use of ADMET models in chemical risk assessments, which pales in comparison to that used in most pharmaceutical companies. The anti-animal testing lobby therefore proposes that companies and regulators should use animal testing only as a last-solution, and even ban its use in most risk assessments [100, 174]. Regardless of the method of hazard testing, a particular weakness of the current regulatory system is that companies lack incentives to initially carry out toxicological testing [175]! Further testing of a substance often leads to equal (if not greater) classification and labelling requirements that result in stricter regulatory control than originally foreseen [175]. In the EU, the classification of a substance as ‘dangerous’ triggers: UÊ
>}iÃÊ Ê Ì
iÊ >Li}Ê vÊ >Ê «À`ÕVÌÃÊ VÌ>} Ì
iÊÃÕLÃÌ>ViÆ UÊ i`>ÌiÊÀiÃÌÀVÌÃÊÕ`iÀÊÇÈÉÇÈÊi°}°]ÊL>ÊÊVÃÕiÀÊÕÃiÊ vÀÊ ,®Æ 28
Base-set level testing is the minimum data set that the current Existing Substances Regulation 793/93/EEC requires to carry out an assessment of risk to human health and the environment for a substance. Regulation 793/93/EEC will be replaced by REACH.
45
Framework for Chemical Risk Management under REACH UÊ VÀi>Ãi`ÊÀiµÕÀiiÌÃÊvÀÊ** ÊvÊVVÕ«>Ì>ÊÕÃiÃÊVÕ`}Ê «ÃÃLiÊLÌÀ}ÊÀÊÜÀvÀViÊi`V>ÊÃÕÀÛiÞ}®Æ UÊ ,iÛiÜÃÊvÊiÛÀiÌ>]ÊVVÕ«>Ì>ÊÀÊ«À`ÕVÌÊÃÌ>`>À`ÃÆÊ UÊ
>}iÃÊ Ê Ì
iÊ ÀiµÕÀiiÌÃÊ vÀÊ «ÀÌqiÝ«ÀÌ]Ê ÃÌÀ>}i]Ê ÌÀ>ëÀÌ]Ê>`ÊÜ>ÃÌiÊ`ëÃ>ÆÊ UÊ «V>ÌÃÊvÀÊiÛÀiÌ>ÊVÌ>>ÌÊ>`ÊÀii`>ÌÊ activities. The combined cost of conforming to regulatory requirements can result in significant expenditure for companies producing, supplying or using any substance classified as ‘dangerous’. Further testing may not always confirm initial suspicion of hazards. For example, initial studies of di(2-ethylhexyl) phthalate (DEHP) indicated that it may be a carcinogen. Additional testing confirmed it to be a non-carcinogen, which industry29 essentially hailed as a victory against over-regulation and ‘chemophobia’ [176]. Increasing evidence indicates that the phthalate causes developmental problems [178]. Industry has also contested these claims based on reproductive toxicity studies [179], but these studies did not test for postnatal endocrine disruption [180]. The only developmental toxicity and endocrine disruption tests available have been conducted by BASF in marmosets [180] and which are argued to prove inconclusive because marmosets are considered to have different developmental and endocrine systems than most mammals [181]. A further dilemma with managing chemicals under current regulation results from companies (including retailers) exhibiting a general unwillingness to change running processes or marketed 29
The term ‘industry’ refers to positions adopted by large trade associations, which comprise several companies that express potentially different views. One outcome of the ‘Public Images of Chemistry in the 20th Century’ conference held in Paris, France, in September 2004, was that industry could improve its reputation by more openly recognising the risks as well as promoting the benefits of chemicals [177].
46
Literature Review products, even if a ‘safer alternative’ exists30 [182-184]. Changing a process can result in new (non-chemical) risks or unexpected management concerns, from supply chain logistics to product performance. Nevertheless, clear market demands for the safer alternative can often catalyse rapid improvements to the environmental performance of products, processes or services [185, 186]. Risks, not just hazards, must be considered when designing or redesigning a product or manufacturing process. For instance, risks from acute toxicity can be greater than from carcinogenicity in a large single exposure [187]. Some form of comparative risk assessment is necessary to evaluate the combinations of properties that chemicals exhibit (e.g., flammability coupled with toxicity versus corrosiveness coupled with toxicity). Even a non-toxic chemical can present risks through its lifecycle by causing resource depletion, requiring energy use and transportation. Decisions based on risk assessment alone can neglect this inter-relation between multiple products and processes [188]. Life cycle assessments (LCA) that evaluate overall health and environmental burdens caused by processes or products therefore complement risk assessments. Carrying out a LCA requires considerable time and resources and, because the final results of a LCA are comparative, they do not represent risk levels. To this extent, although LCA considerations often enter risk management decision-making, regulatory measures typically target high-level risks that do not require this level of detailed information (see Appendix 2.2). Instead, other tools such as socio-economic analysis of a regulatory measure tend to be used when evaluating the inter-relation of processes and products through supply chains. The current lack of knowledge of chemical risks severely hampers the substitution of a hazardous substance by a substance or process that provides a lower degree of risk [183, 189]. Legislation has even resulted in the substitution of one dangerous substance for one that later proves to also present high-level risks. Prime examples include 30
Interested readers may want to be aware of two recent articles in Chemical and Engineering News [182, 186].
47
Framework for Chemical Risk Management under REACH the switch from tributyltin to copper-based antifouling paints, and the replacement of short-chain chlorinated paraffins with medium-chain chlorinated paraffins [190]. Apart from contention over the scientific risks of certain phthalates, delays in the regulation of DEHP, for instance, arose because many alternative substances remain untested. The increase in price per tonne resulting from toxicological testing of main groups of phthalates that occupy about 87% of the 5 billion kg-per-year plasticiser world market is considerably lower than for the already more expensive alternative substances [191]. Companies therefore appear more willing to carry out additional testing on existing phthalates than the alternatives [191]. Regulatory scrutiny targeting phthalates rather than the available substitutes has probably reinforced this industrial approach to risk management. The issue is further complicated by phthalates often being grouped together rather than differentiating LiÌÜiiÊViÀÌ>ÊV
iV>ÊÃÕL}ÀÕ«ÃÆÊ>À}Õ>LÞÊÌ
ÃÊVÕ`ÊLiÊV«>Ài`Ê with attempting to debate and regulate the risk of alcohols to human health without differentiating between ethanol and methanol. While the level of data on existing substances and the application of substitution provide contentious topics for public debate, a major failure of existing chemicals regulation is the poor use and communication of existing hazard information. Surprisingly, this gains relatively little attention from politicians, media, research or the public. A recent monitoring and enforcement exercise across several EU countries identified that only 22% of chemical preparations are in full compliance with classification and labelling requirements [192]. About 20% of the errors for preparations were considered ‘severe’, which includes the incorrect labelling of carcinogens [192]. Similar findings arose from a survey that examined substances. This study found that only 66% of substance safety data sheets31 (SDS) were available to companies and, of these, only 80% of SDS information, 75% of classification and 58% of labelling proved to be correct [193]. 31
In addition to signs at the workplace and information leaflets, safety data sheets (SDS) are one of the primary tools for communicating risk and safety information (see [164, 195, 195]).
48
Literature Review
Producer
Refiner
User Distributor
Formulator
Formulator Distributor Processor User
User
User
User User
User
Figure 2.6 A typical chemical supply chain
Given the potential complexity of many chemical supply chains32 (Figure 2.6) it is hardly surprising that some labelling is incorrect. For instance, miscommunication of information may result from language barriers experienced when tracing products across international supply chains [196, 197]. Deficiencies also arise from a company using the chemical and not internally updating supplied SDS [194, 195], i.e., the SDS gets ‘caught up’ in administration. In comparison to larger firms, ‘SME show lower awareness and knowledge of environmental issues, lack of availability of qualified personnel, lack of top management involvement, higher compliance costs and scarce financial resources’. For instance, small and mediumsized enterprises (SME) tend not to study guidance manuals unless they are given personal guidance from regulators or other institutions such as trade associations [198]. Reports on occupational health and safety lead to the same conclusions [199]. From the enforcement surveys cited above, larger businesses experience fewer problems [192]. Evidence also indicates that environmental management systems can promote a higher level of compliance with SDS [192]. 32
Useful references that consider the complexity of chemical supply chains include [149].
49
Framework for Chemical Risk Management under REACH A study in 1999 across Austria, Germany, and the Netherlands identified limited use of SDS in many SME [195]. To some extent, this results from a lack of knowledge of legislation and technical know-how to interpret risk information, as well as the userfriendliness of SDS [195]. Written from a supplier perspective, data in SDS are often poorly presented, difficult to interpret, and fail to provide practical solutions for risk management [194]. Risk reduction measures can vary considerably for different preparations, even if they contain the same amount of a particular substance. Examples range from selecting the material for personal protective equipment (e.g., gloves made of nitrile versus butyl) to controlling accidents. Because most companies in the EU are SME (Table 2.2), chemical risks in SME are of particular concern to regulators.
Number of employees at each site 1–9
10–99
100–249 250–499
500+
Sales (%)
3
10
10
14
63
Number of companies (%)
70
22
4
2
2
Table 2.2 Company size in the EU chemical industry
Of equal (or perhaps greater) concern to risks in SME are risks to professional users33 and consumers. According to the European Commission, some level of experience or supervision must be assumed for professional uses, albeit with a limited amount of knowledge on chemical safety [200]. With regards to consumers, procedures and guidelines for chemical use are extremely limited, and legislation does not require SDS to be supplied with consumer products. Even if information is available to consumers, many consumers will not read, understand or follow instructions [201, 202]. 33
‘Professional users’ refers to workers in micro-sized enterprises in a wide multitude of trades such as painters, cobblers, and cleaners.
50
Literature Review Regulators are therefore faced with not only devising systems to generate risk assessment data, but also communicating safety information effectively. Because most risk management activities occur within a company, regulators often seek to create incentives for companies to adopt standards of best practice. For instance, chemical customers generate incentives for ‘upstream’ substitution or innovation [183]. Regulators will therefore target chemical users instead of producers and interact with stakeholder groups to promote responsible chemical use through the supply chain.
2.3 The Regulatory Process 2.3.1 Regulatory Options A risk management measure to ensure or promote safe chemical use may be set-up and monitored by regulators, suppliers, trade associations, insurers and non-governmental organisations (NGO), so the degree of organisation between actors (as well as the number of actors involved) determines the efficacy and practicality of implementing regulation. A particular drawback to some regulatory measures, such as emission limits or occupational exposure, is that regulators must keep up-to-date with science and technology and adjust legislation accordingly [203]. Industrial plants can also experience technical difficulties in complying with ‘top-down’ command and control regulation due to the cost of technological investments and compliance administration [203]. Once installed, investments in technical equipment can even reduce incentives for companies searching for alternative measures to achieve risk reduction34 [204]. To overcome these hurdles, various forms of best practice guidance can be devised and disseminated [132, 205–207]. Compared with 34
A classic example of this occurred in the USA, when the 1970 Clean Air Act required electric generating companies to install emission abatement technology to limit sulfur emissions. Faced with significant investment costs, some companies switched from higher- to lower-quality coal. The result was an increase in sulfur emissions from ‘dirty coal’, offsetting any reductions from the end-of-pipe desulfurisation process [209].
51
Framework for Chemical Risk Management under REACH specific regulatory requirements, guidance enables companies and regulators to consider site-specific operational technical constraints and economic constraints, as well as variations in local environmental conditions and changes in scientific knowledge [208]. Regulators may experience particular difficulty in verifying that the relevant guidance is being communicated and followed [132]. A ban on the production, marketing or use of a chemical offers the most stringent type of regulatory control. Prohibition provides clear-cut cases for product liability and enforcement, thereby minimising exposures resulting from poor regulatory compliance or irresponsible chemical use. Product or process35 bans can be in the form of restrictions or authorisations. Restrictions allow all uses of a chemical with the exception of those decided on a case-by-case basis. Under authorisation systems, all uses of a chemical are banned unless exempted. Permitting schemes can be viewed as authorisation systems, but they usually only control production. To control the marketing of chemicals, certification schemes exist under which companies can sell only to certified, licensed, or trained customers. The costs for conforming to a certification scheme can be particularly high for SME36 [132] and certification schemes may be inappropriate in sectors where there is a high turnover of staff [132]. It is often argued that economic instruments37 provide greater flexibility to companies for achieving emission limit values or other 35
For example, EC Directives place several command and control-type obligations on employers to protect employees who may be more prone to adverse effects of chemical exposures, such as pregnant employees [210] and young persons [211].
36
If costs for training and certification are high, certification schemes may act as a preventive measure by discouraging the use of a particular product or process.
37
A straightforward tenet of the EC Treaty is that it is the polluter who should pay for pollution or harm caused by its activities or products. It establishes that responsibility for the causation of environmental pollution can be measured in monetary terms, but limitations in evaluating the effects of pollution (especially in terms of monetary value) severely hamper application of this principle. If applied fully, the ‘polluter pays’ principle would mean that prices of all products would reflect the full cost of production and consumption, including the environmental cost [220]. In this way, final consumer use of products would be the considered to be polluting activities.
52
Literature Review environmental goals than command and control measures [212–214]. Levies on discharges or industrial processes generally encourage polluters to limit their polluting activities [212]. ‘Emissions trading’ is another economic instrument [215]. Regional or local levels of pollution may be set so as not to allow environmental concentration to exceed PNEC. An increase of an emission must therefore be offset by an equal or greater decrease from another firm. A ‘cap-and-trade’ system has been suggested for the discharge of certain metals in the EU with the rate of exchange of permits based on toxicity scales for substances [216], but it remains to be seen if such a scheme will be adopted in a Member State. Setting, monitoring and enforcing economic instruments can prove equally administratively challenging as command and control regulation [216, 217] . Voluntary agreements between regulators and industry can avoid the need for regulation, but require monitoring and reporting requirements. Despite several success stories, there is limited evidence as to their effectiveness in achieving risk reduction [218, 219]. While adherence to a voluntary agreement may be difficult to monitor and enforce among the contracting parties, importers may sometimes not be part of the scheme due the difficulty in identifying the relevant companies. For example, after OECD chemical companies decided to phase-out azodyes in the 1970s and 1980s, production increased in non-OECD countries, and articles containing the dyes continued to be imported as final products [221]. In 2003, the EU decided that it was necessary to ban dyes in certain consumer articles such as [222]. Various incentive-based measures can be voluntarily adopted by companies or be required through legislation. For instance, public accessibility to pollution registers can give incentives to companies to reduce emissions. In 2003, a European Pollutant Emission Register (EPER) was launched listing 50 substances and groups of substances for which water and air are monitored for 10,000 large industrial sites38 38
In terms of dangerous substances, this register currently focuses on long-range air pollutants and prioritised substances from the Water Framework Directive. A major obstacle with devising and implementing this EU scheme arises from its dependence on harmonised monitoring and reporting methods [229].
53
Framework for Chemical Risk Management under REACH [223]. Registers have also been developed under voluntary initiatives and can equally apply to occupational risks [224]. Environmental product declarations based on the use of environmental management systems provide frameworks for companies (especially SME) to improve their environmental performance [225]. Standardised schemes such as EMAS [226] and ISO 14001 communicate best practice to companies through the supply chain [227]. In particular, eco-labelling by regulators or NGO requires a firm to demonstrate that environmental burdens have been minimised through the entire lifecycle of a product. An eco-label will not be awarded if a product contains certain hazardous chemicals and, in some cases, if certain substances have been used during production39 [228]. Insurance cover, liability and regulatory fines act as incentives for companies to comply with legislation or even adopt beyond-compliance practices [230, 231]. A new approach in improving compliance with environmental legislation comes in the form of ‘performance bonds’, which is a variation of environmental liability [232]. The amount of a bond is based on the potential future environmental damages that could be attributed against a chemical producer. If damages occur, Ì
iÊL`ÊvÕ`ÃÊiÛÀiÌ>ÊÀii`>ÌÆÊÌ
iÀÜÃiÊÌÊÃÊÀiÌÕÀi`Ê with interest or traded. Most countries operate with statutory state compensation schemes, which companies may need to supplement with other private-based ÃÕÀ>ViÃÆÊLÌ
ÊÕÃÕ>ÞÊVÕ`iÊiVVL>Ãi`ÊViÌÛiÊÃÞÃÌiÃÊ to improve chemical safety40 [233]. Environmental and product liability generally establish when civil action can be successfully taken against a company or an individual [234]. If a company is in compliance with the law, it is possible that civil action becomes directed towards the regulator in the form of judicial review for 39
It is questionable if eco-labels can provide a cost-effective alternative to controlling pollutants given the time and resources necessary to establish product-specific awards [216].
40
Insurance premiums can be based on frequency of illness, accidents, exposures, or dangerous substances handled.
54
Literature Review having set a law or failing to act [235]. Regulators must therefore avoid two extremes. If they place overly stringent regulatory demand, the administration will face political pressure and even action against it for setting legislation that is not based on science (see Section 2.4.1). If a regulatory administration does not set sufficiently strict regulation, it may be sued for compromising public health or the environment.
2.3.2 Recent Developments in Regulation Since the 1990s, diffuse sources of pollution have been increasingly contributing to synthetic chemicals present in environmental media. From a review of the literature, diffuse emissions originate from three major sources: (1) Discharges from SME outside the scope of Integrated Pollution Prevention and Control (IPPC) or ‘overlooked’ by regulators41 [236, 237]. (2) Environmental releases during professional and consumer product use [236, 237]. (3) Final product disposal (including hazardous household waste). To adopt a preventive approach to address these sources of pollution through the entire supply chain, environmental regulation has begun to target chemical products. Under product-oriented regulation, one set of standards of performance is required for all places where a product or material is being transformed, regardless of the quantities or concentrations involved. The difference between process and product-based legislation is shown schematically in Figure 2.7
41
Regulators often target larger polluters [239]. In some cases this is due to the comparative difficulty in monitoring and small revenue from non-compliance fees from smaller companies [240]. Elevated levels of hazardous pollutants have proven to come from unexpected or ‘overlooked’ dischargers such as SME textile companies, automotive garages and dental surgeries [198].
55
Framework for Chemical Risk Management under REACH
(a) Process or site-based
(b) Product-based
Figure 2.7 Process-based compared with product-based regulation [238].
Squares represent sites, installations or processes. Circles indicate individual foci of specific regulation. Arrows represent flows of materials but do not illustrate the cascading use of products. White areas within the squares represent different concentrations of chemicals regulated. Product-based legislation depends on a lifecycle approach that examines the inter-relatedness of processes and products. It facilitates pollution prevention by identifying potential risk trade-offs through supply chains. Potential advantages of product-based legislation include greater harmonisation of regulatory objectives and controls across value chains, less duplication of legislation, and, in the era of globalisation, increased consistency across political boundaries [238]. Compared with process-based controls, regulating products requires a far greater level of information on risks, whereas assessing the business effects of regulation can be far more complex. Addressing the final stage of the lifecycle of a product, extended producer responsibility and the Commission’s Prevention and Recycling of Waste Strategy [241] place demands on companies and local authorities to ‘take-back’ certain products42. This may be especially important to ensure the correct disposal of used consumer 42
Several companies have operated voluntary take-back schemes before regulation for used computers (Dell, IBM), cartridges from copying machines (Xerox) or even sports shoes (Nike).
56
Literature Review goods, and subsequent reuse, recovery or recycling. Deposit–refund schemes are sometimes used to encourage consumers returning such products [216]. With increasing production volumes of chemicals predicted for the next 20 years [242], the economic advantages of the costly application of recovery, reuse and recycling may prove limited unless regulatory controls are put in place. Chemical bans are now being enforced to ensure that material recovery and recycling targets are technically and economically achievable [243, 244]. Pollution from diffuse sources can be effectively tackled by taxing polluting products and wastes [216, 245]. Examples of chemical taxes include chlorinated solvents in Denmark and Norway [246] and phthalates in articles on the Danish market [269]. The European Environment Agency [245] and the UK Royal Commission for Environmental Pollution (RCEP) [180] suggest developing new tax bases for hazardous chemicals. The European Commission’s Integrated Product Policy also proposes taxation to promote environmentally sound practices through supply chains [247].
2.3.3 Regulatory Decision-Making Three basic principles govern national and EU regulation: subsidiarity, proportionality and precaution. These policy instruments allow regulators some flexibility for regulating chemicals, especially with respect to the scientific and technical tools used to assess and legally justify a decision. Fundamentally, it is the principle of subsidiarity that enables a Member State to set the level of protection it deems necessary within its national territories. A regulatory measure must be proportionate43: (a) it should be the option that is the least restrictive to trade [248], and (b) to also comply with World Trade Organisation rules, it must also not discriminate44 between products produced within different national territories [249]. 43
If a deviation from harmonised legislation is in place, it must be actively demonstrated to the European Commission through notification procedures (see [250]).
44
In this respect, measures to protect human health and the environment should not be confused with protectionism, which relates to trade.
57
Framework for Chemical Risk Management under REACH Any national deviation from EC legislation under national law must be notified to the European Commission, which may decide against the legislation. In these cases, the European Court of Justice may make the final decision on application of legal principles. Several Member States have taken regulatory measures beyond any enacted at the EU level. Cases include: (i) Austria’s ban on polybrominated biphenyls * ®ÆÊ®ÊÌ
iÊ iÌ
iÀ>`ýÊL>ÊÊiÀVÕÀÞÊÊÌ
iÀiÌiÀÃÆÊ>`Ê Denmark’s near-total ban on all uses of lead. In some instances, the European Commission has challenged the action of a Member State [251]. The most famous case was when Sweden banned trichloroethylene (TCE). The European Court of Justice held that Sweden was entitled to the ban because proportionality was respected through a system of authorisations [252]. Any regulatory measure must be based on scientific assessment, but there is no provision in EC or international law that requires Member States to conduct a full-risk assessment before taking legislative action [253, 254]. In other words, regulatory measures can be based on hazard if evidence supports that a negative effect on human health or the environment is occurring. In practice, a causal relationship must usually be established for substance-specific regulatory action [253]. The precautionary principle provides some opportunity to deviate from this rule because it states that regulation may be based on a substance having a ‘potentially negative effect’ [255]. A link between a substance and an effect rather than a causal effect can suffice for taking regulatory action45. The European-Commission definition of the precautionary principle should be distinguished from the use of uncertainty factors during risk assessment or margins of safety during risk management46 [255]. Assessment factors account for assumptions made during the risk assessment process, such as when deriving no-effect levels. 45
This aspect of precaution is taken as the working definition in the EU. Although three identifiable interpretations of precaution exist, they are not mutually exclusive (see Appendix 2.3). It is therefore more appropriate to consider various interpretations as ‘aspects’ rather than ‘versions’ of the application of the precautionary principle (Appendix 2.3).
46
Some regulators and scientists argue that this distinction is not justifiable!
58
Literature Review Levels of safety are further used to establish risk limits based on factors such as variations in exposure levels. With the use of assessment factors and other methods to analyse and compensate for uncertainty, risk assessment therefore aims to ‘err on the side of caution’ when assessing hazard and exposure data. By comparison, the precautionary principle uses a ‘weight of evidence’ approach to identify and control the probable cause of an observed effect. For instance, epidemiological studies may identify a statistical link between exposure to a chemical and a negative health effect that has not been confirmed in toxicological studies. In such a case, the precautionary principle may be applied, especially if the effect may be difficult to measure by toxicological testing. The precautionary principle can therefore avoid delays in enacting regulation measures resulting from timelines necessary to carry out additional testing or to conduct further studies.
2.3.4 Regulatory Approaches In a book on EU chemical risk management, Calow identified four approaches to risk management: suspicion, hazard, technical, and risk–benefit [256]. These are categorised according to the amount of scientific evidence and the amount of ‘caution’ involved in decisionmaking (Figure 2.8):
Suspicion Amount
Hazard
evidence Technical
Amount
caution
Risk benefit Figure 2.8 Approaches to risk management based on evidence and caution [256]
59
Framework for Chemical Risk Management under REACH A suspicion-based approach implies that evidence of a chemical hazard or exposure can immediately lead to regulatory action. Hazard-based approaches focus on the sources of a potential risk, attempting risk reduction by limiting the quantities of hazardous material produced and used. A hazard-based approach therefore requires only the hazard of a chemical to be identified. A technical-based approach follows the view that risks should be reduced to achieve an adequate level of safety, but does not require a socio-economic analysis of alternative risk management options. Standards are set according to available technology to ensure that certain margins of safety are not exceeded. A risk–benefit approach bases decisions on the results of socio-economic and stakeholder consultation. Figure 2.9 may not adequately represent the amount of ‘evidence’ required under each approach. For instance, identifying hazardous properties may include (or even rely) on epidemiological and environmental monitoring data. These data may be greater than the amount of laboratory-derived toxicological data used to set a technical exposure limit. As discussed in Section 2.2.2, some level-1 in vitro mutagenicity testing may be more reliable than level-2 in vivo studies for predicting carcinogenity even though level-2 data are considered to be a more ‘complete’ dataset. The amount of evidence therefore depends on how regulators ‘weigh’ different types and sources of data. Unfortunately, Calow neither defined ‘evidence’ nor provided detailed examples of the amount of data used under these approaches. Nevertheless, the general characterisation of the four approaches provides a starting point for understanding different types of chemical regulation. Reasons behind the various approaches may stem from alternative evaluative philosophies or ‘frames’ used by regulators and stakeholders when analysing chemical risks [257], i.e., the degree of confidence:
60
Literature Review 1. In the capability of mankind to fully evaluate risks. 2. In technological risk-reduction measures and the skilled behaviour of the people who manage these systems. 3. That nature can cope with the consequences of errors of humans in risk assessment and risk management. These ‘evaluative frames’ primarily relate to the scientific risk assessment process and the technical ability to manage a risk. By contrast, this book seeks to examine the impact that the different regulatory approaches have on decision-making at national and EU levels. In other words, how a regulatory action varies in the shortand long-term in terms of its (adapted from [258]): UÊ vwV>VÞ\ÊÌ
iÊ>LÌÞÊÌÊVÌÀÊ>ÊÀð UÊ vwViVÞ\ÊÌ
iÊëii`]Ê«À>VÌV>ÌÞ]Ê>`ÊVÃÌÃÊvÊ«iiÌ>Ì°Ê UÊ vviVÌÛiiÃÃ\Ê Ì
iÊ ÀiÃÕÌÊ vÊ >VÌÊ Ê ÃÕ««ÀÌ}Ê }>ÃÊ ÀÊ objectives. UÊ µÕÌÞ\ÊÌ
iÊ`ÃÌÀLÕÌÊvÊÀÃÃÊ>`ÊLiiwÌð Efficacy refers to the scientific and technical ability to control a risk. Interpretations may vary between stakeholder groups because levels of safety depend on compliance. Efficiency describes the ease of technical and administrative implementation and monitoring of regulation. It therefore depends on the resources available to regulators and stakeholders in meeting regulatory demands, which also determine costs of implementation. The effectiveness of a risk reduction describes if it achieves an objective or complements an overall risk management strategy. This will vary considerably between regulators and stakeholders according to political goals and even personal aspirations. Equity revolves around the advantages and drawbacks that regulatory action imposes on different stakeholders, such as the distribution of costs and benefits of regulation. For instance, banning a substance will be costly to the producer, but
61
Framework for Chemical Risk Management under REACH provides a market opportunity to a company that manufactures an alternative. In terms of benefits, discontinuing a hazardous substance may improve levels of environmental or health protection, but only if it is replaced with a ‘safer’ substitute. To illustrate the differences between the ‘four Es’, a hypothetical comparison of two risk reduction measures is provided in Box 2.1. The implications of both options are considered for three stakeholder groups: regulator, industry trade association (TA), and an environmental NGO. To construct the example, the ‘general approach to chemical risk management’ for each stakeholder must be clearly established and defined. Hypothetical scenario A risk assessment identifies that substance X is a non-genotoxic carcinogen and presents a risk to employees after exposure via inhalation during use of the substance for metal degreasing. The regulator, company trade association (TA), and environmental non-governmental organisation (NGO) adopt the following ‘general approaches’ to chemical risk management: Regulator: ‘Any risk-reduction measure that reduces a risk to an acceptable level is appropriate.’ TA: ‘A risk-reduction measure that does not result in a significant decrease in manufacturing potential while adequately controlling a chemical risk is acceptable.’ NGO: ‘T h e u s e o f a n y ca r cin o g en i c c he m i c a l i s unacceptable.’ Risk-reduction measures To ensure an acceptable level of risk, two possible riskreduction measures are proposed:
62
Literature Review Option A Use of local exhaust ventilation (LEV) to minimise occupational exposure via inhalation during metal degreasing. Option B Complete ban on the use of chemical X for metal degreasing. The implications in terms of efficacy, efficiency, effectiveness and equity of each option to each stakeholder are presented in Table 2.3. In terms of efficacy, risk management options ‘A’ and ‘B’ can be considered to achieve an adequate level of control for the regulator and industry TA, even though their approaches to chemical risk management differ. For the NGO, requiring the use of LEV (option ‘A’) is inadequate due to the view that the technical machinery is unreliable. The NGO may accept that option ‘A’ can reduce the risk sufficiently if it operates through a regulatory permitting and inspection scheme. Assessment of the efficiency of the measures differs significantly between the enforcing regulator and the companies represented by the TA. The regulator could prefer option ‘B’, a ban on marketing and use, if it is applied as a general rather than use-specific ban because it would be simpler and cheaper to enforce. Requiring Statements of Best Practice from metal degreasing companies could be a suitable alternative. For the TA, option ‘B’ is not economically viable. Notice that the NGO does not have a role in implementing ‘A’ but is likely to assume a proactive engagement in implementing ‘B’. In terms of effectiveness, option ‘A’ is suitable to the regulator and the TA because it adequately controls risks, but is not suitable to the NGO because it conflicts with their aim to ban all carcinogens. According to the NGO, option ‘A’ could set a bad precedence for future discussions on the bans of carcinogens. Interestingly, a ban is acceptable to the regulator and the NGO but for two very contrasting reasons. For the regulator, option ‘B’ is a matter of ensuring adequate risk
63
Regulator TA NGO
Efficacy: Use of a LEV lowers the risk to an acceptable level. Efficiency: Enforcing the use of a LEV would be an administrative burden because its implementation would be difficult to monitor. Establishing a permitting or licensing scheme may therefore be required. Effectiveness: The use of technical controls is a suitable option if it adequately controls a chemical risk. Equity: Industry can be made responsible to monitor occupational or environmental exposures.
Option B Efficacy: A ban on marketing and use would lower the risk to an acceptable level. Efficiency: Enforcing a general ban would be easy to operate through customs and excise, but ensuring that the ban is only for metal degreasing would be difficult. Statements of best practice could be a suitable alternative. There are no risk assessments for potential substitutes in metal degreasing. Effectiveness: Any measure that achieves an adequate level of control is suitable. Equity: A ban ensures safety with minimal monitoring costs to all stakeholders.
Efficacy: A LEV would achieve an adequate level of safety. Efficiency: Installing a LEV would not be very costly because it can be added-on to the existing process. Effectiveness: The use of technical controls is always a suitable option providing it does not entail significant costs. Equity: Because an LEV costs industry, but it may reduce insurance premiums and future liability.
Efficacy: A ban would adequately control the risk. Efficiency: A ban is not efficient because some companies already use LEV and achieve adequate safety. Many companies would face high costs due to the expense of using substitute substances or processes. Effectiveness: A ban is never suitable becuase the socio-economic consequences in terms of manufacturing and employment always outweigh the advantages. Equity: A ban does not grant the industry the chance to act responsibly and compromises its future competitiveness.
Efficacy: A LEV would not achieve an adequate level of safety because it is subject to technical malfunctions. Efficiency: The NGO would not be involved in implementing this risk-reduction measure. Effectiveness: Use of a LEV does not support the NGO objective of banning all carcinogens. Equity: A LEV does not support moves to the sustainability of the chemical industry. Accidents may occur during use of the substance. Therefore not all stakeholders eventually benefit in the long-term.
Efficacy: A ban on this chemical is the only means of achieving adequate risk control. The ban should be extended to cover all industrial uses. Efficiency: The NGO will raise an awareness campaign to retailers to ensure that they are not supplying products containing this chemical. Effectiveness: A ban on this chemical is a necessary step towards the banning of all carcinogens. Equity: A ban ensures that industry does not benefit from high-level risks.
Framework for Chemical Risk Management under REACH
Table 2.3 A hypothetical comparison of two risk-reduction measures from the perspective of a regulator, company TA and environmental NGO
64
Option A
Literature Review control, whereas for the NGO it supports their objective of banning all carcinogens. Option ‘B’ is considered particularly ineffective for the TA because it would go against their approach of ensuring that chemical risks are controlled without significantly reducing industrial competitiveness. A ban could set a precedence for future decisions to be made using this option. An equitable solution does not appear to reside in the position of any of the concerned parties. The impact on employment and risks predominantly lie with the workers. Unfortunately, the trade union does not appear to have been engaged in this debate. What if a workers syndicate jumps in at the last moment? It is likely to adopt a perspective that incorporates some elements of all three other stakeholder groups. This example illustrates how many factors and actors can enter decision-making at a national or EU level. It is therefore not surprising that regulators tend to adopt certain approaches to risk management. Establishing a regulatory approach can provide some degree of predictability to the position that a regulator will assume on a risk management option and the subsequent regulatory outcome [259]. In turn, one would expect that the national approach may be characterised according to the administrative infrastructure of a country [31, 32]. For instance, the involvement of multiple regulatory agencies could require attaining consensus between regulators when adopting a regulatory decision that may be avoided by having a single chemical regulator. Equally, an administrative network may promote certain stakeholder interests to the exclusion of others [260]. Figure 2.9 illustrates how risk management interlinks policy, regulatory rules, risk management practice and risk-reduction strategies. Each factor directly or indirectly influences the others. Descriptions of the four factors involved at a national level are provided in Appendix 2.4. Previous research on chemicals policy and regulation identifies the following variables as affecting each factor that defines a national approach [27, 31-33, 151, 262]: 65
Framework for Chemical Risk Management under REACH UÊ i}>ÊÃÌÀÕVÌÕÀiÃÊ>`ÊÀi}Õ>ÌÀÞÊ>`ÃÌÀ>ÌÛiÊvÀ>ÃÌÀÕVÌÕÀiÃÆÊ UÊ -âiÊ>`ÊÃ}wV>ViÊvÊÌ
iÊV
iV>Ê`ÕÃÌÀÞÊ«À`ÕVÌ]Ê«ÀÌ]Ê iÝ«ÀÌ]ÊÕÃi®ÊÊÌiÀÃÊvÊi«ÞiÌÊ>`ÊÀiÛiÕiÆ UÊ ,iÃ]ÊÀiëÃLÌiÃ]ÊÀi>ÌÃ
«ÃÊ>`ÊÀiÃÕÀViÃÊvÊÌ
iÊ>VÌÀÃÊ involved in decision-making, including historical stakeholder }ÀÕ«ÊÀi>ÌÃÊ>`ÊÃÌ>ÌiÊÌÀ>`ÌÃÆ UÊ -V>Ê >`Ê VÕÌÕÀ>Ê «iÀëiVÌÛiÃÊ ÛÛi`Ê Ê `iÌiÀ}Ê ÀÃÊ tolerability and implementing risk management measures. It is these variables that the research project investigates, and the subsequent implications that these national variables ultimately have on EU decision-making. Although risk assessment forms the cornerstone of EU decisionmaking, EU legislation comprises a mix of hazard, technical and risk–benefit. For instance, the general ban on CMR chemicals in consumer products is based on hazard47. Exception on the ban of cadmium in paints for products coloured for safety reasons follows a risk–benefit rationale [263], as does balancing technology standards with implementation cost under the Directive for Integrated Pollution Prevention and Control. Strict technical control also enters EU legislation, such as the ‘as low as technically possible’ requirement for preventing work exposures to non-threshold carcinogens [264]. If France, Germany, Sweden and the UK adopt different regulatory approaches, this research seeks to uncover which role these Member States have in the adoption of different types of EU legislation. It follows that this research must first establish if the regulatory approach of a Member State is consistent with its approach to EU decision-making. With the advent of REACH, the question now arises as to what effect these national approaches may have on REACH implementation, and what effect REACH will have on national approaches. 47
These bans appear under points 29, 30 and 31 of Directive 76/769 [265] and are limited to category 1 and 2 CMR.
66
Literature Review
2.4 REACH 2.4.1 A New Era for Chemical Control Described as one of the most ‘controversial’ pieces of legislation in European history [266], the REACH Regulation comprises >130 legal Articles supplemented by 17 Annexes and 14 Technical Guidance Documents. These documents make up several thousand of pages of ÌiÝÌÆÊ, ÊÃÊViÀÌ>ÞÊÌ
iÊÃÌÊV«iÝÊÃ}iÊ«iViÊvÊV
iV>Ê legislation devised [266]. Yet the aim of REACH is simple: to shift the ‘burden of proof’ from regulators identifying chemical risks to companies demonstrating safe chemical use [164]. In this respect, REACH closely resembles ‘product stewardship’ in the chemical industry whereby manufacturers seek to ensure that their products can be used safely and responsibly through the supply chain, including final disposal48 [267]. Policy Rules Practice Risk Reduction Strategy
Figure 2.9 Chemical risk management: the inter-relation of policy, rules, practice and risk-reduction strategies (based on [260]). Influences are indicated by arrows. 48
In 2004/2005, the Head of CEFIC’s Responsible Care programme expressed the opinion that REACH could have been avoided if the chemical industry had been more active with product stewardship programmes [267].
67
Framework for Chemical Risk Management under REACH The precise scope and detail of the legislation depends on: (i) demands of the regulatory requirements, which are subject to Ài}Õ>ÌÀÞÊÀiÛiÜÃÊ`ÕÀ}Ê«iiÌ>ÌÆÊ®ÊvÕVÌ}ÊvÊÌ
iÊ newly created European Chemicals Agency (ECHA49®ÊÊiÃÆÊ and (iii) supporting Technical Guidance Documents [47]. The core elements of the regulation have long been established. Companies must submit registration dossiers containing toxicological, use and exposure data for substances into a Registration, Evaluation and Authorisation of Chemicals-Information Technology (REACHIT) database maintained by the Chemicals Agency. It is the level of hazard assessment and the extent of mandatory data-sharing between companies necessary for registration that have been the focus of negotiation between the European Commission, the European Council of Ministers and the European Parliament [268]. As this section of the book will begin to uncover, risk management decision-making remains largely unexplored. More than 30,000 50 substances manufactured 51 or imported above the level of 1 tonne per year52 will be covered by REACH. In response to current deficiencies in the level of data contained in SDS, substances classified as ‘dangerous’ or identified as VPVB manufactured or imported by a company at >10 tonnes per year in the EU must also have detailed safety measures communicated according to specific chemical uses in an annex to SDS entitled ‘exposure scenarios’. Although exposure scenarios will be based on risk assessment results already conducted under existing legislation, such as the Chemical Agents Directive [269] and 49
This acronym distinguishes the Agency from the European Court of Auditors (ECA).
50
With the inclusion of isolated intermediates and the possibility of variations in substance definitions due to impurities or differences in composition, the number of substances subject to registration could be substantially more than 30,000.
51
The REACH legal text refers to chemical production as ‘manufacture’.
52
References to tonnages in REACH apply per legal entity. In other words, a company can reduce its legal requirements by dividing tonnage between multiple legal entities. Regulators will prioritise regulatory reviews (for evaluation, restrictions and authorisation) based on aggregate tonnage.
68
Literature Review Integrated Pollution Prevention and Control (IPPC), it must cover all ‘intended uses’ through a specified branch of a supply chain. Under REACH, the results of such ‘chemical safety assessments’ must be documented as ‘chemical safety reports’ (CSR). An exposure scenario can be as wide or specific as a company carrying out a risk assessment deems necessary. The advantage of a generic exposure scenario is that it covers a range of uses, but this may result in some over-protective risk-reduction measures being implemented. In comparison, a specific exposure scenario can clearly delineate how to control exposures from distinct uses or target potential high-exposure uses that require particular safety measures. A specific exposure scenario may therefore require a more in-depth safety assessment. REACH introduces the concept of ‘adequate control’ in EU chemical law. Traditionally, the term ‘adequate control’ has been used to refer to good practice in the workplace. REACH now redefines ‘adequate control’ in the form of risk management measures detailed in an exposure scenario necessary for the control of hazardous properties. Through a set of systematic procedures, risk management measures must be selected to reduce exposure below which adverse effects to human health or the environment are likely to occur (i.e., a DNEL, DMEL or PNEC). There is debate as to whether a concept of a safe level of exposure reduction, similar to adequate control, can apply to non-threshold carcinogens and mutagens, endocrine disruptors, persistent, bioaccumulative and toxic (PBT) or VPVB substances (e.g., [270]). Industry may need to demonstrate that exposure to these substances is always avoided or minimised, as specified in Annex I of the REACH Regulation. Under REACH, downstream users can contact the manufacturers or importers of chemicals to make uses known. A manufacturer or importer will then be required to register that use unless it
69
Framework for Chemical Risk Management under REACH decides to discontinue supply. Alternatively, the downstream user can independently notify the European Chemicals Agency of an intended use and the corresponding exposure scenario. When uses are not communicated before registration, then Chemical Safety Assessments can be carried out to ensure that the use is adequately covered by an exposure scenario or the safety measures on a SDS may be followed, as appropriate. Existing occupational and environment legislation continues to apply.
Volume of substance manufactured or imported in tonnes per year (tpy)
Approximate number of substances
General testing requirements
Phase-in periods for existing substances (years)
1–10
18000
Annex VII
10
10–100
5000
Annexes VII–VIII
10
Annexes VII–IX
5
Annexes VII–X
2.5
100–1,000
2600
q1,000 tpy or
2700
known CMR53 q1 tpy
900
known PBT q100 tpy
130
Table 2.4 Registration deadlines and general testing for existing substances
To cope with the large number of existing substances, registration deadlines for existing substances will generally apply according to 53
Approximately 6/7 of the approximately 900 known CMR are oils, tars and other crude petrochemical derivatives and will not be subject to detailed regulatory review (see [271]).
70
Literature Review tonnage (Table 2.4). Substances already identified as CMR (category 1 or 2) prove an exception, and must be registered during the first phase. As of November 2005, the European Parliament and European Council of Ministers agreed to extend this requirement to include potential PBT and VPVB substances (classified as R50/53 ‘very dangerous to the environment’) that a company produces or imports at >100 tonnes per year [271, 272]. To facilitate data exchange between companies, a pre-registration phase has been devised to run between 1 June 2008 and 1 December 2008. Once pre-registered, a substance gains the status of ‘phase-in’ where it can benefit from the staggered registration deadlines. Companies, including relevant stakeholders that possess data on a given substance, must then form a ‘Substance Information Exchange Forum’ to share data. Manufacturers and importers can then submit joint registration dossiers via a form of consortium. The objective of sharing data is to minimise the duplication of animal testing and reduce registration costs. A company can ‘opt-out’ of sharing certain confidential data or submitting information jointly if it is too costly, but sharing of vertebrate animal test data relevant for a registration will be obligatory. Member States will be responsible for reviewing the registration dossiers. Evaluation consists of checking registration data and assessing if further test or exposure data must be generated by industry in the form of targeted risk assessments54. As with current chemical legislation, testing requirements become more elaborate as the quantity of a substance produced or marketed increases, with test requirements specified in several annexes of the REACH Regulation (Table 2.4 – number of substances based on [271-274]). REACH will allow companies to carry out lower levels of testing if evidence indicates that testing is not necessary. This can be driven by data indicating that: (i) a substance is not hazardous or (ii) there is only low exposure (e.g., due to the application of safety 54
There are several exemptions or reduced requirements that include: polymers, intermediates, and substances or specific uses that have already undergone risk assessment.
71
Framework for Chemical Risk Management under REACH measures). In this way, REACH will not always mandate that a DNEL, DMEL or PNEC be established and that ‘adequate control’ be demonstrated for every single substance, endpoint, and possible population or ecosystems. Even when evidence indicates that a substance is hazardous, under certain circumstances companies may have the option of ensuring that exposure is avoided or minimised rather than having to determine a DNEL, DMEL or PNEC55. REACH presents a major change to the approach to the generation of hazard assessment data. Although the testing requirements follow the same general principle of increased data according to production volumes as under current legislation, three crucial variations exist. First, there is a focus on the use of in vitro and in silico test systems. Second, testing requirements can be ‘waived’ on the basis of predicted low hazard, negligible exposure or combination thereof. Third, REACH requires companies to demonstrate ‘why’ there is a ‘need’ for vertebrate animal testing before conducting such tests. Companies must also use a ‘weight of evidence’ approach to combine and consider multiple sources of information when interpreting data or justifying a need for further testing. The extent to which REACH will achieve these changes to existing risk assessment practices will largely depend on the development and validation of these ‘intelligent’ or ‘integrated’ assessment strategies [174], especially at the international level [273]. Any chemical substance and use that is identified as being of regulatory concern during evaluation will be subject to either an authorisation or restriction process depending on the relative priority vÊ Ì
iÊ ii`Ê ÌÊ «iiÌÊ i>ÃÕÀiÃÊ ÌÊ VÌÀÊ >Ê }ÛiÊ ÀÃÆÊ Ì
iÊ rationale is that higher-level risks will be subject to an authorisation process. The exact criteria for potential candidates for authorisation have still not yet been fully defined, but these ‘substances of very 55
Another alternative method to streamline compliance is to identify the most crticial DNEL, DMEL or PNEC for a substance (on its own or in a preparation) with regard to defining appropriate operational parameters and risk management meaures.
72
Literature Review high concern (SVHC)’ will include CMR, PBT, VPVB, sensitisers and certain endocrine disruptors. Industry will have the opportunity to comment on proposed regulatory measures. Decisions on the granting of authorisations will further divide substances according to two routes: (1) Substances that can be ‘adequately controlled’ because a clear DNEL or PNEC can be established and an application for authorisation demonstrates that exposure is below this DNEL or PNEC. (2) Substances where authorisations depends on a socio-economic justification of continued use because a safe ‘threshold’ for exposure (i.e., a DNEL) cannot be established or ensured (e.g., non-threshold CMR, PBT and VPVB substances). ‘Adequate control’ and ‘socio-economic’ routes require that a company carry out an analysis of possible alternative substances and technologies. If a ‘suitable alternative’ exists, a company must then propose a substitution plan which will influence the time period that an authorisation is granted. A wide range of stakeholders can contribute to the authorisation decision-making process by submitting comments to the European Chemicals Agency. Other than stating that substitution must be ‘technically and economically feasible’ and reduce overall risks to human health and the environment, the text of REACH does not specify the details of how the authorisation process will work or what >ÊÃÕLÃÌÌÕÌÊ«>ÊÃ
Õ`ÊVÌ>ÆÊÃÕV
Ê`iÌ>ÃÊ
>ÛiÊLiiÊivÌÊÌÊÌ
iÊ RIPs and the functioning of the Agency that must still be completed or formed, respectively.
2.4.2 Uncertainties in the New System Uncertainties in the new system and costs of registration are likely to cause a rationalisation of chemicals on the EU market. The number 73
Framework for Chemical Risk Management under REACH of discontinued substances may be as high as 20% for low-value and low-volume products56 due to costs and the human resources required for registration alone, i.e., regardless of the degree of risk that a substance may pose [275]. The corresponding number of discontinued preparations and articles may be significantly higher. Problematically, REACH does not provide a mechanism to prevent the rationalisation of safe products. For many downstream companies, conforming to REACH could become very difficult due to the high numbers of chemicals they use (approximately 50% use between 100 to 1,000 different substances each year) [276]. Some companies, such as those that do not specialise in chemical use, may not even have the resources or capacity to check whether their chemicals are registered, let alone comply with any Registration requirements [77]. Initially, it was proposed that substances contained in articles would need to be registered only 11 years after enactment ‘if during normal and reasonably foreseeable conditions of use and disposal the substance may be released in sufficiently high amounts and in such a way as to adversely affect human health or the environment’ [277]. Because their EU-counterparts may already have had to register the substance, this timeline may be particularly advantageous for companies importing articles into the EU market. The European Council of Ministers therefore proposed that registration of substances in articles follow the same phase-in periods as for substances and, for operability, that this should apply only to substances ‘intended’ on being released from articles [271]. As a separate mechanism, any substance that is identified as a potential candidate for authorisation (i.e., a ‘SVHC on the candidate list’) present in a produced or imported article will need to be ‘notified’ if >1 tonne per year57 – i.e., the Agency will need to be informed of its presence in the article. For all product types, the presence of 56
This figure corresponds to approximately 2% of all substances manufactured or imported above the level of 1 tonne per year in the EU [278].
57
An exception applies if the substance has already been registered for that particular use – irrespective of supply chain.
74
Literature Review such a substance must be communicated through the supply chain if >0.1% weight per weight58, regardless of total tonnage. The legal text also allows for full registrations of any substance in an article on a case-by-case basis. REACH has therefore become more evenly balanced in terms of its requirements for substances manufactured in the EU and substances in articles that are imported from outside the EU. Non-EU producers of articles will not need to register all the raw materials and processing chemicals used during production, as their EU-counterparts will need to do. Major questions therefore concern what constitutes ‘intentional’ release, how imports will be monitored and enforced, how many substances may be placed on the ‘candidate list’ for authoirsation as SVHC, how the European Commission may propose to adapt the legislation, and how the Agency will execute its given powers.
2.4.3 Decision-making under REACH The REACH regulation will require many risk management decisions to be made within relatively ‘short’ periods of time [379]. The brevity of periods may be very significant to industry and become a bureaucratic nightmare. As with existing legislation, decisionmaking under REACH will be largely dependent on whether existing environmental, occupational and consumer legislation is sufficient to control newly identified risks. Chemicals risks will continue to be identified through exposure monitoring data under other legislation, such as the Chemical Agents Directive, IPPC or the Water Framework Directive [280]. Under REACH, chemicals will also continue to be restricted, banned or authorised under several pieces of different legislation, including product-specific legislation such as the Toys Directive, Construction 58
A cut-off threshold of 0.1% weight/weight applies, but it currently appears unlear as to whether this concentration must be determined at the total article level or on a component or even a material basis.
75
Framework for Chemical Risk Management under REACH Products Directive and Food Contact Material Directive. Most regulatory provisions regarding environmental, occupational and consumer protection respectively fall under the legal bases of Articles 175, 138 and 153 that only provide minimum standards (see [281]). REACH falls under the scope of Article 95, which seeks to preserve the internal market while taking human health and the environment into consideration. REACH therefore applies ‘harmonising’ standards which limit the potential for Member States adopting more stringent measures according to the ‘subsidiarity principle’ (Section 2.3.3). To avoid overlaps, the REACH regulation requires that the European Commission carry out a review on the interaction between REACH and other pieces of legislation by 2012.
2.4.4 Potential Improvement to Health and the Environment There is little doubt that the identification and control of hazardous chemicals is necessary to protect human health and the environment. An assessment of the occupational health benefits of REACH estimates that improved chemical risk assessment and risk management can reduce compensation for worker-related illness by between €18 and 54 billion over a 30-year period [190]. The long-term benefits of improved environmental protection resulting from the identification of hazardous chemicals under REACH (e.g., avoided costs for carrying out environmental remediation) can readily result in savings of hundreds of million Euro per substance [190]. Based on World Bank estimates that chemicals and chemical pollution causes between 0.6% and 2.5% of diseases in ‘developed’ countries, the European Commission calculated a saving of €50 billion on health and medical care within the EU over 30 years could result if REACH can reduce the occurrence of disease by 0.1% [282]. There is the danger that REACH could become an administrative burden to companies and that the updated SDS end up being stored in a filing cabinet. Studies on the benefits to health and the environment do not consider how REACH can be implemented to provide efficient
76
Literature Review or effective risk communication or risk management. Identifying a dangerous chemical is only a small part of the risk management process, and the authorisation process will be limited to a very small sub-set of ‘high concern’ substances [36, 279]. Safeguarding human health and the environment will ultimately depend on how REACH is implemented.
2.4.5 Business Effects of REACH: Cry Wolf? A great deal of controversy surrounds the business effects of REACH. Environmental NGO argue that the chemical industry is ‘crying wolf’ over the potential negative effects that the regulation will have on EU businesses [283]. NGO support this assertion with the fact that the cost estimates for implementing the current REACH proposals represent only about 0.05% of the annual turnover of the EU chemical industry (about €400 billion) [283]. The costs to downstream chemical users represent <1% of their total turnover of around €425 billion [283]. Evidence suggests that it is not uncommon for industry to present ‘over-estimations’ of the costs for regulatory compliances on industry [283, 284]. The concerns of the chemical industry over REACH have been severely hampered by two of their own studies. First, a study commissioned by the German Chemical Industry Federation (VCI) that predicted a severe effect of the regulation on German GDP (by several percent) and employment figures (by about 100,000) was challenged by a group of economic experts gathered by the German Environment Agency for its extrapolation of micro-economic estimates (based on a number of sub-sectors) to predict national economic impacts [285]. A similar study commissioned by the French Chemical Industry Association drew comparable conclusions for the French economy [286]. But the methods used for the French industry report are confidential, and the results have been rejected by French Government sources [285]. Subsequently, the European Commission rejected the validity of both studies [285]. The predicted costs and benefits of the legislation vary to such an
77
Framework for Chemical Risk Management under REACH extent across 36 key regulatory impact assessments that, according to the Enterprise and Industry Commissioner Günter Verheugen, ‘it is scarcely possible to come to [any] conclusions’ [287]. Surprisingly, no official regulatory study has examined the variation of costs across companies or the sub-sectors of the chemical industry, let alone of other manufacturing sectors! An investment company concerned with the potential negative business impact of REACH therefore commissioned the Centre for Environmental Strategy to evaluate the company-specific costs for 19 EU chemical firms. The study found that cost–turnover ratios vary by a factor of 20 above and below the ‘averages’ cited by Environmental NGO and the European Commission59 [288]. Variations occur due to the type and quantity of a chemical product produced and its final end-market. Controversially, the European Commission does recognise that REACH will result in the collapse of several smaller chemical companies, especially those producing a high proportion of low-volume/low-value chemicals [278]. Larger companies can be expected to survive better than SMEs due to their (i) larger resource base, (ii) ability to operate economies of scale, and (iii) ability to distribute reduced product margins across wider product portfolios. National and EU ‘helpdesks’ will therefore be established to assist SMEs to comply with regulation [271]. To promote international competitiveness of the EU chemical industry and encourage innovation, the REACH proposal: UÊ VÀi>ÃiÃÊÌ
iÊVÕÀÀiÌÊÌ
ÀiÃ
`ÊvÀÊÌiÃÌ}ÊvÊiÜÊÃÕLÃÌ>ViÃÊvÀÊ 10 kg to 1 tonne. UÊ VÀi>ÃiÃÊ Ì
iÊ Ì
ÀiÃ
`Ê Ê «>ViÊ vÀÊ ÃViÌwVÊ ,iÃi>ÀV
Ê >`Ê Development (R&D) exemptions from 100 kg to 1 tonne. UÊ ÝÌi`ÃÊ «ÀViÃÃÉ«À`ÕVÌÀiÌi`Ê ÀiÃi>ÀV
Ê >`Ê `iÛi«iÌÊ (PPORD) exemptions from one to three years, up to a maximum of five years. 59
Because this business impact assessment covered large chemical companies, it was assumed that companies have most data already available for registration. Therefore costs may be significantly more, especially for SME [288].
78
Literature Review One study often quoted by environmental groups, conducted by Nordbeck and Faust (2002), argues that chemical regulation provides impetus to innovation60. It also stipulates, as does the European Commission, that the decreased requirements in EU legislation for low tonnage substances, R&D and PPORD exemptions compared with the current system for new substances will promote innovation, but this argument is flawed: 1. Data quoted by the study by Nordbeck and Faust was incorrectly sourced from a previous study conducted by Fleischer and co-workers (2000), which would otherwise indicate that regulation hinders innovation [289]. 2. The availability and use of existing substances in the innovation of new substances is not considered. 3. The resource allocation of R&D expertise to registration compliance may compromise future process or product innovation. REACH will offer companies new market opportunities61 [288]. By encouraging companies to adopt ‘value pricing’ that incorporates product supply guarantees and various other forms of service provision [79, 290], REACH may present many chemical companies with the opportunity to move away from being principally reactive to competitor prices and raw material costs [291]. A study that examines how REACH could be devised or implemented to promote innovation in the chemical sector is lacking. Decisionmaking under REACH will need to address this lack of companyand sector-specific business impact assessment data. In particular, regulators will need to account for the distribution of business management risk between companies, across sectors and through supply chains. Otherwise the financial and management hurdles 60
The study compared the rate of new substances placed on the market in EU, US and Japan.
61
For instance, one major EU company already anticipates increasing service sales revenue by about 5% during the initial phases of REACH [288].
79
Framework for Chemical Risk Management under REACH created by registration under REACH may become even more pronounced when issuing authorisations and restrictions.
2.5 Conclusions Regulatory risk management is not a purely scientific, technical and i}>ÊiÝiÀVÃiÆÊÌÊÃÊ>Ê«ÌV>Ê«ÀViÃÃÊÌ
>ÌÊÛÛiÃÊV>ÀivÕÊ>>}iiÌÊ of individual and group interests in the face of public concern over chemical risks. Regulation is an iterative process of devising and implementing regulation to attain political goals. While views of the effectiveness and equity of a regulatory option will largely depend on the political position adopted by a given actor, the roles of national regulators and stakeholder groups will determine the efficacy and efficiency of implementing a risk-reduction measure. REACH offers a potential solution to the risk assessment of existing chemicals but does not provide answers on how to better manage a chemical risk once it has been identified. REACH may compromise the ability for regulators and companies to achieve major improvements in safeguarding human health and the environment. The mammoth scope of REACH may present a complex bureaucracy to regulators, companies and stakeholders that redirect resources away from the management of high-level risks. Future regulatory decision-making will need to compensate for the limitations of the substance-by-substance and tonnage-based approaches adopted under REACH. Regulators will need to contend with the fact that while restriction and authorisation are only two regulatory control options, exactly how REACH will interact with other legislative frameworks is uncertain. Increasing international competition and poor public perceptions could jeopardise the sustainability of the EU chemical industry. REACH is a once-in-a-lifetime opportunity and the stakes are high. Whether REACH seals or opens the coffin for many EU chemical businesses will depend on the finer details of its implementation, which appears to be in the hands of the EU Member States, the European Commission and the new European Chemicals Agency. 80
Literature Review There is surprisingly little research on national approaches to chemical regulation and the effect that countries have on EU regulation. Yet, it is precisely these aspects of regulation that will affect the costs and benefits of implementing REACH. A first step to investigating REACH therefore requires examining national regulatory approaches and the implications that Member States have on EU decision-making. The next step must consider the extent that different approaches can continue to operate under REACH. As a final step, the research must establish the degree of national approach ‘harmonisation’ necessary to ensure efficient and effective risk management under REACH.
81
3
Methodology
- Learning The kind of ignorance distinguishing the studious (Ambrose Bierce, The Cynic’s Word Book, 1906)
3.1 Research Design During a European Commission Traineeship in 2002, I observed decision-making between Member States to be a process of conflict rather than collaboration. Delays resulting from a lack of consensus formation troubled me. It was not so much the costs incurred to taxpayers caused by lengthy negotiations that spurred me to research EU decision-making, it was the implications that delays in enacting regulation have on health, the environment and industry. I secured PhD funding from the UK Engineering and Physical Sciences Research Council in 2002 and set out to investigate how EU decision-making could be improved. In Chapter 2, the literature review exposed aspects of risk management under REACH that will depend on national policy, rules and practices. Therefore, in addition to tracking developments with the REACH legislative text, the methodology had to contend with changes to policy and practice at national and EU levels. A framework for EU decision-making that does not account for such variables would be rejected by Member States or the European Commission.
83
Framework for Chemical Risk Management under REACH Interviews were the primary method for data collection. In the four subject Member States, key regulatory officials and stakeholder organisation representatives were asked questions on national chemical risk management and their views of EU regulation1. When analysing national policy and practice, interview data needed to be supplemented and validated by incorporating literature references relating to (Section 2.3.3): UÊ ÃÌÀV>ÊÃÌ>i
`iÀÊ}ÀÕ«ÊÀi>ÌÃÊ>`ÊÃÌ>ÌiÊÌÀ>`Ìð UÊ -âiÊ vÊ V
iV>Ê `ÕÃÌÀÞÊ «À`ÕVÌ]Ê «ÀÌ]Ê iÝ«ÀÌ®]Ê employment, and revenue. UÊ -V>Ê>`ÊVÕÌÕÀ>Êv>VÌÀÃÊÃ
>«}ÊÀÃÊÌiÀ>LÌÞÊ>`ÊyÕiV}Ê the implementation of risk management measures. Methodologies reliant on observation, such as cultural theory, were not considered suitable for investigating the roles and relationships of such a large number of actors [292]. Lack of access to decisionmaking processes in the European Union (EU) or Member State also hindered the possibility of carrying out observational studies. A major aim of the research was to gain an understanding of past activities and events to gain insight into current and future regulatory practices. Compared with questionnaires, interviews avoided the need for respondents to commit significant time to providing answers in written form. Questionnaires were also not deemed appropriate because specific responses may require clearance from administrative hierarchies within organisations. Within the context of this PhD research project, compared with interviews, questionnaires would create particular uncertainty in terms of interviewee response rate and time. One drawback of interviews resulted from the time and travel necessary to meet participants. Even within a single country the relevant
1
A pilot study conducted between January and March 2004 confirmed that interviews could generate sufficient data to characterise the national approaches and relevant EU decision-making processes.
84
Methodology organisations were geographically dispersed, thereby restricting the possibility of conducting group interviews. Consequently, the choice of interviews as a research methodology limited the number of surveyed countries, but the selection of countries in terms of their influence on EU decision-making and regulatory styles justified the research design. Thirty-six interviews were conducted (four in France, 10 in Germany, 10 in Sweden, 10 in the UK and two EU non-governmental organisations (NGO)) with 27 organisations involved in chemical risk management (listed in Appendix 3.1). Interviews were conducted in English, French or German. Five interviews consisted of multiple respondents (2 or 3) and four interviews consisted only of short specific questions relevant to the organisation2. The use of interviewees therefore avoided the need to create detailed multilingual versions of the questionnaires. Notes on interviewee responses were summarised and categorised according to ten subject areas on national and EU chemical risk management covered by the interview questions (Section 3.5). Responses were compared and validated in a process referred to as ‘cross-checking’ (Section 3.6). To investigate the business implications of implementing EU legislation, the results from 12 interviews with Safety, Health and Environmental Managers from eight multinationals in six industry sectors were incorporated into the data set. This research was conducted in collaboration with [238] with the intention that it could be used for the purposes of this PhD project. The results from these interviews were organised according to three topics areas, which included corporate views on REACH (Section 3.5). The final objective of the research project was to develop a framework for decision-making under REACH. Applying the framework should result in a predictable set of regulatory options to control the use of 2
In these cases, interviewees expressed unwillingness to fully participate in the research because their specialist knowledge limited their ability to answer all the research questions.
85
Framework for Chemical Risk Management under REACH these dangerous chemicals. The practical operability framework was therefore tested using risk assessment results for 33 chemicals under regulatory review. The potential for adoption of the framework at the EU level was also gauged by comparing various elements of the framework with different national regulatory approaches. An overview of the research methodology is shown in Figure 3.1. A brief explanation corresponding to each stage of the project follows in Table 3.1. The strengths and limitations of the research design are presented in Section 3.8. Interviews Data set 1
Data organisation
Industry responses Data set 2 [262]
Cross-checking
National approach (Chapter 4)
Systems framework (Chapter 5)
Framework evalutation (Chapter 6)
Figure 3.1 Overview of the research methodology (analytic processes shown in grey)
86
Methodology
Process
Description
Notes
Data collected from interviews
36 interviews with 27 organisations
Data set 2
Results of interviews from research by [238]
12 interviews with 8 EU multinational companies in 6 sectors
Method to extract and summarise relevant interview data
Data organisation described in Section 3.5
Data
Data set 1
Data organisation
Results presented in an annex Technique to validate the interview data
Cross-checking described in Section 3.6
Cross-checking
Analyses
Results of crosschecking presented in an annex National approaches
Analysis of the Findings presented in regulatory approaches Chapter 4 for the four countries Development of the systems framework
Systems framework
Development of the framework described in Section 3.7 Results presented in Chapter 5
Framework evaluation
Testing of the framework using chemicals under regulatory review Evaluation of the framework based on national approaches
Evaluating the framework described in Section 3.7 Results presented in Chapter 6
Table 3.1 Descriptions of the processes comprising the research methodology shown in Figure 3.1 87
Framework for Chemical Risk Management under REACH
3.2 Analytical Framework The research project adopted a ‘soft-systems’ analytical framework to investigate national approaches. Originally developed by [293], this methodology uses engineering principles to examine social, political or economic phenomena (Figure 3.2). In engineering, a system is defined as a process in which an input is ‘transformed’ into an output. Several actors are involved in the process, as well as a number of operational constraints. Applying this approach to chemical risk management would require the identification of the factors shown below. UÊ /
iÊinput in terms of past, present and future chemical risks. UÊ /
iÊ output in terms of the consequences of risk management activities aiming to achieve chemical safety (e.g., effectiveness and equity). UÊ /
iÊ actors involved in risk management (e.g., stakeholder associations, regulators). UÊ /
iÊ constraints of science, technology and society (e.g., risk identification, efficacy of risk management control risks, tolerability of risk). Input: Risk
Actors: Roles, resoponsibilities, relations, resources
Process:Risk management decision-making & implementation
System constraints: Legal, scientific, economic, social
Output: Safety
Figure 3.2 Soft-systems analytical framework (adapted from [293]) 88
Methodology The objective of risk management can therefore be seen as process to ‘transform’ risk into safety. This process varies from implementing technical legal requirements to communicating risks through information campaigns. Based on system definitions articulated by Checkland and Scholes [293], the project formulated the following concise description of the risk management system: For the purposes of chemical risk management, the relevant regulatory authorities of a country are responsible for developing and evaluating risk reduction strategies to ensure chemical safety. The process involves careful consideration of scientific and engineering assessments, national policy, existing legislation, obligations under EC law, regulatory infrastructure, stakeholder values and concerns, and the socio-economic impact of regulation. Parallels between the soft-systems analytical framework and a constrained relativism are evident when comparing Figure 3.2 with Figure 1.1(b) in Chapter 1 – reproduced as Figure 3.3 below. According to the soft-systems approach, actors and system constraints form the ‘feedback loops’ that determine ‘reality’ in terms of chemical risk and safety (see Figure 3.3). This thesis therefore considers the impact that different national actors and system constraints have on risk management processes at national and EU levels.
‘Real World’ a
‘Observations’
‘Real World’ b
‘Observations’
‘Real World’
The real world is not directly accessible. What is ‘observed’ is a ‘world’ that is a function of the real world and an input of the observer. The resulting local feed-back loops imply a historically determined ‘reality’. Different ‘observes’ thus may live in different ‘worlds’; none of these is more true than the other.
Figure 3.3 Constrained relativism (taken from [9])
89
Framework for Chemical Risk Management under REACH In each of the four countries, regulatory administrations and stakeholder associations were selected for interviews according to their active participation in decision-making processes at national and EU level3. These actors were investigated in terms of their roles, responsibilities, relationships, and resources using the definitions shown below (adapted from Cramer [554] and Mintzberg [555]): Role:
Division of tasks for risk management between the actors.
Responsibility: Power relations between the actors with regards to risk management. Relationship:
Communication between the actors.
Resources:
Internal and external resources (technical, scientific, political) available to the actors.
The research investigated the potential for negotiation and compromise between the actors. The project also examined the influence that regulators and stakeholder groups have on the efficacy, efficiency, effectiveness and equity of implementing a risk-reduction strategy (RRS) (Section 2.3.4). Ultimately, the tolerability of a RRS depends on such group dynamics (Sections 1.2 and 2.3.4). The soft-systems analytical framework facilitated the comparing and contrasting of national approaches, as well as interviewee perspectives of EU decision-making and REACH. In total, the results of these processes formed the basis for proposing a framework for decision-making under REACH.
3
These organisations, and the relevant staff representatives, were identified from EU and national regulatory authority documentation, such as publicly available reports and consultations.
90
Methodology
3.3 Interview Selection Conducting interviews with representatives of all national stakeholder associations and regulatory authorities was unfeasible, so organisations were selected to represent the broadest areas where chemical risk assessments identify that risk occurs, specifically: UÊ ÛÀiÌÊVÕ`}Ê
Õ>Ê
i>Ì
ÊÛ>ÊÌ
iÊiÛÀiÌ®° UÊ "VVÕ«>Ì>ÊÃ>viÌÞ° UÊ ÃÕiÀÊ
i>Ì
° In each of the surveyed countries, at least one stakeholder organisation was selected from each of the following groups: chemical industry trade association, worker trade union and environmental/consumer NGO. Although other manufacturing sectors and even retailers have a significant role in many aspects of risk control, their roles and responsibilities in chemical risk management were identified from data collected from interviews and literature searches. For instance, chemical industry trade associations and environmental NGO are involved with wide cross-sectoral chemical management issues [78]. Stakeholder associations were identified from publicly available regulatory authority documentation (e.g., reports, consultations). If there was more than one possible organisation of equal relevance that could be interviewed within a given category (i.e., environment, occupational, consumer), the organisation was selected according to personal contacts or physical (geographical, temporal) accessibility. Apart from one exception4, all regulators and stakeholders contacted expressed a keen interest to participate, frequently stating that the topic of risk management warrants research investigation. 4
Only one organisation, a Swedish environmental NGO, was unwilling to participate. Emails were not returned and relevant members of staff remained inaccessible despite several attempts to make contact by telephone.
91
Framework for Chemical Risk Management under REACH
3.4 Interview Technique A semi-structured interviewing technique was used to allow interviewees sufficient flexibility in articulating their responses while ensuring sufficient topic coverage for the subsequent analysis. Interview questions acted as guides or probes, based on a ‘funnelling’ technique (see [294]): UÊ /
iÊ }iiÀ>Ê ÜÀ`}Ê >`Ê ÃÌÀÕVÌÕÀiÊ vÊ Ì
iÊ ÌiÀÛiÜÊ V
>}i`Ê according to how the individual interviewees responded during the course of the interview [295], so interviewee answers were allowed to run on when relating to another question area. UÊ /À>ÃÌÃÊ ÜiÀiÊ ÕÃi`Ê ÌÊ ÀivVÕÃÊ Ì
iÊ vÀ>̽ÃÊ >ÌÌiÌÊ onto a certain topic or issue, usually accomplished by linking the informant’s previous answer to the relevant topic of interest [296]. UÊ +ÕiÃÌÃÊ ÜiÀiÊ ÀivÀÕ>Ìi`Ê >`Ê «ÀL}Ê µÕiÃÌÃÊ ÕÃi`Ê ÌÊ elicit more detailed or specific information beyond the original response [297]. Interview questions needed to address the research questions, but were elaborated and defined after the literature review. Specifically, the questions needed to cover 10 broad topic areas relating to policy, rules, practice and experiences with official risk reduction strategy processes at national and EU levels (as described in Section 2.3.4 and defined in Appendix 2.4). These 10 topic areas were anticipated as minimising potential overlaps between interviewee reports of national and EU dimensions while being sufficiently wide to group variations in interviewee responses. I National Dimensions 1. Policy, administrative structures and decision-making 2. Actors (roles, responsibilities, relations, resources) 3. Risk management options (voluntary agreements, taxes)
92
Methodology 4. RRS process 5. EU decision-making (implications at the national level) 6. Responses to REACH (preparing for REACH) II EU Dimensions 7. Commonalities and differences between Member States 8. Obstacles and opportunities for future EU regulation 9. EU RRS process 10. Responses to REACH (recent developments, future decision-making). Interview questions and technique were tested during a pilot study with two regulators from Germany, two from Sweden and one from the UK. A copy of the findings from the pilot study was sent to German interviewees and the results discussed at a subsequent meeting. The response from the regulators to the pilot study was very positive and did not identify weaknesses in the research methodology. The pilot study identified necessary revisions to the interview questions, as shown below. UÊ `i«Ì
Ê µÕiÃÌÃÊ Ê LÕ`}iÌ>ÀÞÊ ÀiÃÕÀViÃÊ >`Ê ÕLiÀÊ vÊ staff available within an organisation should be avoided. The remit of practically every organisation interviewed covers specific aspects of risk assessment and management that are not directly pertinent to the research study (e.g., pesticides, detergents). Questions on resources resulted in total budgets and staff numbers for pertinent and non-pertinent activities, limiting the potential for country comparisons. Respondents also required a disproportionate amount of time in answering these questions. UÊ +ÕiÃÌÃÊ ÌÊ Ài}Õ>ÌÀÃÊ Ê ÃiviÛ>Õ>ÌÛiÊ «À>VÌViÃÊ Ã
Õ`Ê LiÊ limited because interviewees responded with reference to specific
93
Framework for Chemical Risk Management under REACH obligations and duties under national and EC legislation that is otherwise easily accessible. UÊ ÌiÀÛiÜʵÕiÃÌÃÊÃ
Õ`ÊvVÕÃÊÊÌ
iÊÃÌÊ«iÀÌiÌÊÃÃÕiÃÊÀiiÛ>ÌÊ to that organisation, as identified in the literature review. Less directly relevant questions should be left to the end of the interview. The final list of questions and guiding probes is in Appendix 3.2, corresponding to a typical interview lasting between 45–75 minutes. In general, interviews began with questions relating to the general roles and responsibilities of the national actors. Interviewees were then asked questions on the general resources and relationships between the various actors. Finally, the interviews investigated decision-making at the EU level. Questions were carefully constructed so as not to influence interviewee response. Interviews should be a one-way process because researcher bias can be more easily introduced in a two-way process of communication [298]. Interviews were also constructed and conducted so as to avoid confrontation and challenge, allowing the interviewees to respond freely. Most interviews were conducted between January 2004 and >ÀV
ÊÓääxÊ>ÌÊÌ
iÊÀ}>Ã>̽ÃÊ
i>`µÕ>ÀÌiÀÃÆÊwÛiÊÌiÀÛiÜÃÊÜiÀiÊ conducted over the telephone when scheduling meetings proved difficult. Interviewee responses over the telephone tended to be shorter and therefore less detailed, even when using follow-up probes. The research was conducted in accordance with the Statement of Ethical Practice of the British Sociological Association with regard to informed consent, anonymity and confidentiality. Data collected from the interviews were presented as general statements resulting from the analysis.
3.5 Initial Analysis Although soft-systems analysis provided an over-arching framework for organising and interpreting data, it was also necessary to follow a methodology to extract and validate information from the interview
94
Methodology data sets. The methodology needed to be reproducible and easily verifiable. Not only is this necessary for checking the accuracy and validity of the findings, but other researchers should be able to follow the methodology (e.g., if other countries were to be added or compared with the analysis). Interviews were recorded on audiotape but not transcribed due to time and cost considerations. Instead, interviewee responses were summarised into descriptive phrases and paragraphs. In many instances, interviewees provided detailed or convoluted descriptions of risk management processes that they deemed necessary to communicate an idea or statement during the interview. This high level of data facilitated the summarising and grouping of responses, but is not detailed in this thesis. Notes taken during the interview were later supplemented and reviewed by taking further notes when re-listening to recordings of the interviews. To facilitate interpretation and presentation of data, interview responses under each of the ten topic areas listed in Section 3.4 were sorted into sub-categories (i.e., ‘subjects’). In some cases, subV>Ìi}ÀiÃÊÜiÀiÊÕµÕiÊÌÊ>ÊVÕÌÀÞÆÊÊÌ
iÀÊV>ÃiÃ]ÊÌ
iÞÊ>««i>Ài`Ê as recurrent themes across all four countries5. This facilitated comparisons between interviewee responses. At this point, results from interviews conducted as part of research by [238] on implementing EU environmental legislation were inserted into the data pool. These interviews were done in spring and summer 2003 with 12 corporate EHS staff from eight multinational companies spanning six industry sectors (chemicals, pharmaceuticals, electrical engineering, mechanical engineering, food and drink, public domain products). This data set covered many of the same issues covered by the research project, and was organised according to corporate responses to current EU regulation, REACH and future EU regulation. 5
Techniques of discourse analysis that measure the frequency of words used or evaluate the association between ideas expressed by interviewees were not suitable due to variations in interviewee expertise and knowledge, as well as interview structure (e.g., [299]).
95
Framework for Chemical Risk Management under REACH
3.6 Interviewee Response Validation – Cross-Checking To achieve a high degree of validity, data collected from several sources were compared and cross-checked [300, 301]. Answers from respondents were checked against: £°Ê "Ì
iÀÊÌiÀÛiÜÊ`>Ì>Æ 2. Sourced references produced by the respondents during interviews ÌÊÃÕ««ÀÌÊÌ
iÀÊÃÌ>ÌiiÌÃÆ 3. Background information on the actor organisation (e.g., background documentation, policy documents, position statements) and previous research studies. If an interviewee expressed uncertainty as to the validity of his/ her response, the data were omitted from the research findings. Information provided by the interviewees that was outside the scope of the study was excluded, for example: UÊ vÀ>ÌÊ«iÀÌ>}ÊÌÊÀÃÊ>ÃÃiÃÃiÌÊÌÊ`ÀiVÌÞÊÀiiÛ>ÌÊ ÌÊÌ
iÊÀiÃi>ÀV
Æ UÊ vÀ>ÌÊÊV
iV>ÃÊÕÌÃ`iÊÌ
iÊÃV«iÊvÊÌ
iÊÃÌÕ`ÞÊi°}°]Ê pesticides). If discrepancies occurred following any of the procedures described above, either (a) the notes were checked by listening to the interview recordings again, (b) the subject matter was specifically examined in further interviews with other interviewees, or (c) the research study results made these existing differences explicit. Although interviewee responses were frequently reviewed using the tape recordings and several points needed to be clarified during subsequent interviews (e.g., relationship between certain actors), no problems in the methodology resulted from the cross-checking exercise. Any perspective that was expressed by only one or two interviewees or which could not be supported by references in the literature was explicitly stated in the research findings6. 6
For the purposes of the PhD viva voce, the complete set of interview notes was made available to the examiners as a separate annex to the thesis.
96
Methodology
3.7 Systems Framework A four-step methodology was devised to develop a framework for EU risk management decision-making under REACH (Figure 3.4).
Data collection for developing framework Step 1
Framework evaluation Step 4
Step 2
Step 3
RRS* TGD**
REACH legislation and RIPs***
RRS reports
Interviews and restrictions Identification and categorisation of elements necessary for framework
Criteria and decision-making processes according to categories
Criteria and decision-making processes according to categories
Testing framework categories and decisionmaking processes
Actor roles and responsibilities for national and EU risk management
Relevance and implications of actors for EU decision-making
Identification of methods for supporting actor responsibilities
Necessity for national versus EU action
Problems with current regulatory processes
What processes have been followed, why the problems may have occurred
Identification of how regulatory processes may change under REACH
What outcomes could result from using the framework
Problems anticipated under REACH
Potential obstacles with implementing the current TGD under REACH
Potential solutions to obstacles
Avoidance of potential obstacles and improvements to the TGD
Improvements to national & EU risk management
Identification of structural changes to the TGD
Identification of decision-making structures
Opportunities for streamlined decisionmaking
Figure 3.4 Overview of the step-wise process for developing and testing the framework *RRS = Risk-reduction strategy **TGD = Technical guidance document [28] ***RIPs = REACH implementation projects 97
Framework for Chemical Risk Management under REACH After following the general research methodology for data collection and analysis (step 1) a framework for EU decision-making under REACH was developed using the current Technical Guidance Document on the Development and Selection of Risk Reduction Strategies [28] (step 2). This technical guidance document (TGD) is a tool that Member States had previously already agreed upon but predates REACH. Being >100 pages long, it is rather cumbersome to use and has not been revised since its publication in 1998. Part of developing the framework therefore involved producing a new and updated streamlined version of this guidance document. Because a soft-systems approach was used as a structure to analyse EU risk management (i.e., inputs, outputs, actors, constraints and feedback loops), the framework for decision-making under REACH is referred to as the ‘systems framework’. Interviewee responses to questions on the current TGD and official EU RRS process proved fundamental to developing the systems framework. In particular, the research focussed on problems experienced with existing EU decision-making structures that could remain under REACH. Decisions taken under Directive 76/769 [302] were incorporated into the framework by establishing chemical risk criteria that previously warranted EU restrictions (step 2). Guidelines from the RRS TGD were also extracted and condensed according to categories of chemical risk criteria. Interviewee responses were then compared with the regulatory requirements of REACH and relevant draft reports or guidance from the REACH implementation projects (RIP) to identify potential limitations in the European Commission legislative proposal (step 3). Finally, the operability of the systems framework was tested using 28 completed EU RRS reports7 and a further five chemicals also under regulatory review 8 (step 4). 7
Summaries of the risk-reduction strategies (RRS) for 28 substances published in the Journal of the European Union were supplemented by risk assessment reports [13, 303-310].
8
A further five chemicals (bisphenol A, deca-bromodiphenyl ether, medium-chained chlorinated paraffins, phthalates, trichloroethylene) were included due to information provided by regulators during interviews or available in referenced literature [304, 311-313].
98
Methodology The systems framework was also evaluated by considering how it meets actor demands and system constraints of EU decisionmaking. The process for developing and evaluating the framework is described in Chapters 5 and 6.
3.8 Strengths and Limitations of this Research The rationale of the research project presented in Chapter 1 explained the selection of France, Germany, Sweden and the UK (Section 1.6). Evidence suggests that France, Germany and the UK may seek future co-operation on chemical matters. Immediately before the legislative proposal for REACH in October 2003, the governments of France, Germany and the UK issued a joint communiqué to the European Commission stressing the importance of maintaining the international competitiveness of EU industry under REACH [314]. Preparatory activities for the implementation of REACH further support the relevance of the choice of countries. Regulators from France, Germany, Sweden and the UK were four of the seven countries9 to have participated in a Strategic Project on REACH Testing10 (SPORT) of the Registration and Evaluation processes of REACH [315]. In short, these countries appear to be some of the most active in past, present and future chemicals policy in the EU. The major exception appears to be the Netherlands, which has developed a very detailed national chemicals policy of its own [316] which it is implementing in parallel to REACH [317]. The Netherlands will probably use its policy to influence the development of the REACH regulation. Aspects of the Dutch policy have therefore been contrasted to REACH during presentation of the research findings (Chapter 5).
9
The other principal countries involved in SPORT were Finland, Italy and the Netherlands.
10
SPORT focussed on risk assessment evaluation and did not examine regulatory risk management decision-making for restriction and authorisation.
99
Framework for Chemical Risk Management under REACH Arguably, Italy could have replaced France as the fourth country. Like the Netherlands, Italy has a strong chemical industry and may therefore be active in EU decision-making processes. Italy also holds an equal number of votes as France, Germany or the UK in the qualified majority voting system, twice that of the Netherlands. Although Italy offers the advantage of representing the regulatory style of southern countries, language barriers would have been problematic. Not a single publicly available chemicals policy position paper from the Italian government appeared to be available at the start of the research project in October 2002. Sweden’s approach to chemicals policy closely resembles that of other Nordic countries, especially due to the strong role of the Nordic Environment Council [261], but the environmental policies of France and the UK have been placed in the same categories as southern EU countries (see Section 1.6). Countries that have recently joined the EU have very small chemical industries (Section 1.6) and regulatory concerns on the administration and enforcement of REACH of France, Germany, Sweden and the UK may be anticipated to be more pronounced for these countries (Chapter 4). The chemicals policy of Germany may prove particularly influential on eastern European countries due to its geographical proximity coupled with its large production of chemicals [318]. France, Germany, Sweden and the UK are expected to be representative of most EU Member States. This proves particularly important when evaluating the potential for the systems framework to be adopted at the EU level (Chapter 6). Due to time constraints, only four interviews could be conducted in France but, given the amount of data collected on EU decisionmaking during interviews in Germany, Sweden and the UK and made available by [238], this would probably not compromise the ability to develop a framework for EU decision-making. While conducting the interviews, it became apparent that many regulators were becoming increasingly busy preparing for REACH.
100
Methodology This limited the options available for evaluating the systems framework. There was little recompense to offer regulators or stakeholder representatives in return for completing questionnaires on various aspects of the framework or participating in post-framework interviews. Interviewees already presumed that the research findings would be circulated among certain Member State regulators and officials of the European Commission. Similarly, organising a small workshop to discuss the research findings was deemed too challenging, especially as hosting such an event would require additional funding. The research findings were disseminated to the interviewees, certain members of the European Commission, and the relevant European Parliament committee reporters. The systems framework was formatted and submitted as a short document to the RIP responsible for developing TGD for restriction and authorisation. Responses received were considered in terms of the overall contribution of the research project. A major limitation of the methodology arose from restricted access ÌÊ>Ì>Ê>`Ê 1Ê`VÕiÌÃÆÊ, ÊÃÊ>Ê
}
ÞÊÌ«V>ÊÃÃÕiÊÌ
>ÌÊ is under constant discussion and development. I found that meeting with regulators and key stakeholder representatives helped keep the research ‘up-to-date’ with the latest regulatory developments. As the research findings conclude, regulatory attention continues to focus on the risk assessment rather than risk management processes of REACH (Chapters 4 and 5). Targeting the risk management aspects of implementing REACH has therefore enabled the findings to contribute to future policy, technical guidance and the functioning of the future Chemicals Agency (Chapter 7). A summary of the research limitations is presented in Table 3.2.
101
Countries Interviews
Selection of Countries
number of interviewee
Potential Relevance to Results
Justification for Methodology
No ‘southern’ or ‘eastern’ European countries
A different approach to chemical risk management and enforcement issues
- F, D, S and UK demonstrate the four different styles of policy evident in the EU15 (note that southern states fall under the same groupings as UK) - Eastern Member States have very small chemical industries - D policy may prove particularly influential on eastern countries due to its geographical proximity and its large chemical production - Limited language barriers: knowledge of English, French and German
The Netherlands
Highly developed national chemicals strategy
- Dutch chemicals strategy available in English - Dutch strategy considered when devising the Systems Framework (Chapter 6)
A gap in towards the end-use of the supply chain
- Retailers are playing an increasing role in chemicals policy - Chemical distributors may have specific views on chemicals policy and risk management
- Geographical and temporal accessibility to interviewees - Choice of ‘major actors’ in chemicals policy - Environmental and consumer NGO are likely to be involved in the activities of retailers - Regulators and industry trade associations are likely to be aware of the role and responsibilities of chemical distributors
4 interviews in F
Potential ‘unbalanced’ view of F
- Insufficient time to perform the interviews in France
Potential Gaps
Framework for Chemical Risk Management under REACH
102
Constraints
Resources of organisations
Risk assessment not investigated
Assumes harmonised risk assessment process
Limits the ability for the research to examine the detailed inter-linking of risk assessment and management activities
- The research considers aspects of risk assessment that are relevant to risk management ÊÊ/
iÊVÕÀÀiÌÊvVÕÃÊvÊ, ÊÃÊÊÀÃÊ>ÃÃiÃÃiÌÆÊ>ÞÊ>ëiVÌÃÊvÊ risk management appear neglected
No postframework questionnaires or interviews
Limits the potential to evaluate the framework for decision-making in terms of its potential to be adopted at the EU level
- Insufficient resources and limited accessibility of regulators to conduct post-framework questionnaires or interviews - Desk-based evaluation considers potential for adoption at EU level - Operability of the framework tested using 33 chemicals under regulatory review - Research findings disseminated to interviewees and key regulators
Framework Evaluation
103
Table 3.2 Overview of research limitations: F = France, D = Germany and S = Sweden
Methodology
Interview Questions
Focus
Interviews
- The Pilot Study identified that specific questions on resources were difficult to collect and compare - General resources were considered to depend on the resources of Ì
iÀÊÀ}>Ã>ÌÃ]ÊÌÊÕÃÌÊÌ
ÃiÊ«>ÀÌV«>Ì}ÊÊÌ
iÊÀiÃi>ÀV
ÆÊ information was therefore collected on the general resources available in a Member State
May create difficulty in comparing regulators and stakeholder organisations
Framework for Chemical Risk Management under REACH
3.9 Conclusion The soft-systems analytic framework facilitated organisation of the research project. To provide robust research, the research design used several techniques to check the validity and reliability of the methodology and subsequent research findings. Interview questions followed directly from the literature review, but collection and organisation of data focussed on developing answers to the three main research questions. The structure of this book presents the final research findings according to these questions: Chapter 4 ‘National approaches’ UÊ ÜÊ`iÃÊV
iV>ÊÀÃÊ>>}iiÌÊ`vviÀÊLiÌÜiiÊÌ
iÊ Member States and why? Chapter 4 ‘National approaches’ and Chapter 5 ‘Systems framework’: UÊ ÜÊ `Ê Ì
iÊ `vviÀiÌÊ >Ì>Ê >««À>V
iÃÊ >vviVÌÊ 1 decision-making? Chapter 5 ‘Systems framework’ and Chapter 6 ‘Testing framework’: UÊ 7
V
ÊvÀ>iÜÀÊvÀÊV
iV>ÊÀÃÊ>>}iiÌÊiiÀ}iÃÊ from the research?
104
4
National Approaches
- Influence In politics, a visionary ‘quo’ given in exchange for a substantial ‘quid’ (Ambrose Bierce, The Cynic’s Word Book, 1906)
Introduction France, Germany, Sweden and the UK vary considerably in terms of the size and structure of their chemical industries, regulatory authorities and stakeholder associations. Consequently, there are few similarities between the roles and relationships of the actors involved in chemical risk management across the four countries. Data collected from interviews with regulators and stakeholder association representatives indicate that many inefficient and ineffective regulatory practices at the national level may continue under REACH. Many interviewees expressed the opinion that REACH could intensify current regulatory inefficiencies while creating new hurdles to managing chemical risks at the national level. Investigating national approaches revealed several ways to improve chemical regulation at national and EU levels. A critical conclusion is that each country could benefit from exchanging and learning from other national approaches. The findings presented in this Chapter lay important groundwork for developing a framework for decision-making under REACH.
105
Framework for Chemical Risk Management under REACH The Chapter begins with a description of the chemical industry of each country (Section 4.1) and an overview of the national approaches (Section 4.2). Exploring how and why national regulatory authorities and policy styles differ (Sections 4.3 and 4.4) enables an analysis of the strengths and weaknesses of each approach and the effect that these Member States have on EU decision-making. The research identifies that understanding how a country will implement REACH requires in-depth examination of the social and cultural contexts of national chemical regulation (Section 4.5). Only after carrying out this investigation can we draw conclusions on the effect that REACH is having on national approaches and how national approaches may change in the future (Sections 4.6 and 4.7).
4.1 Chemical Landscapes Contributing approximately 17% more to national gross domestic product (GDP)1 than the other three countries, Germany is only one of a few EU countries in which the value of chemical exports exceeds imports2 [40, 319]. Most other EU countries, including France and the UK, have approximately equal trade balances [320, 321]. Although the chemical industry in Sweden accounts for about 25% less of the national manufacturing value-added than it does in France, Germany or the UK, its contribution to national GDP is approximately the same as in France and the UK. Sweden has a consistent slight trade deficit in chemicals3 [40, 322, 323]. The population of Sweden is approximately 1/10 that of Germany and 1/7 that of France or the UK [324].
1
The value added generated by the chemical industry contributed 1.4% to the GDP in Germany compared with 1.2% in the other three countries – when excluding pharmaceutical products [320-322, 325].
2
Other EU countries with a largely positive trade balance of chemicals in 2002 were Belgium, the Netherlands, and Ireland. France and the UK generally only have small positive balances [40, 321] and the UK has a deficit when including rubbers and plastics [40].
3
In 2002, Swedish (exports – imports)/(exports + imports) equalled 8%. This is much lower than the trade deficit for Cyprus (85%), Latvia (72%) and Greece (62%) [40].
106
National Approaches Many chemical small and medium-sized enterprises (SME) operate in France, Sweden and the UK 4. The approximately 10,000 enterprises5 operating in the UK chemical sector is therefore about 15% greater than the number in Germany even though its turnover is much smaller. The UK also has 25% more industrial sites listed on the European Pollution Emission Register (EPER) than Germany (2,397 and 1,836, respectively) [229]6. These numbers are higher than the approximately 8,000 enterprises operating in France and 1,277 facilities listed on the French EPER. With one of the smallest average numbers of employees in the EU, the Swedish chemical industry comprises about 2,250 companies, and there are only 200 EPER-listed sites. Even though there is a major variation in the number of companies operating in the chemical sector between the four countries, national percentages of the manufacturing workforce frequently handling dangerous substances (and infectious agents) are broadly comparable. Sweden and Germany are a few percentage points lower than the EU15 average (16%), whereas percentages are slightly above the EU average for France and the UK [55]. Workers frequently inhaling chemical vapours and fumes in each country are also all around the EU15 average of 22% [55]. Total exposures of the workforce to carcinogens appear consistent across the EU15 at around 20% [326].
4
The average number of employees in chemical enterprises is about 20 in Sweden. This is similar to the figure for Italy, Cyprus, Poland and Slovenia. The average is about 80 in Germany, 50 in France, and 40 in the UK [40].
5
This number of enterprises includes chemical distributors and separates larger companies into their operational sub-divisions based on facilities.
6
The EPER covers larger companies across manufacturing sectors with high emissions of certain pollutants (Section 2.4).
107
Framework for Chemical Risk Management under REACH
Basic organic
Basic inorganic
Specialty and fine
Trade balance
France
High
Low
Medium
Approximately equal
Germany
High
Low
Medium
Export
Sweden
Medium
Medium
Medium
Import
UK
Medium
Low
High
Approximately equal
Table 4.1 Qualitative descriptors of national chemical industry profiles (based on data from [40, 232, 237-239])
A qualitative comparison of the activity of the chemical industry of each country is given in Table 4.1, which broadly compares turnovers and trade balances. France and Germany primarily produce basic organic chemicals whereas UK manufacturing is based on higher value specialty and fine chemicals. Sweden’s industrial activities are spread across the three sub-sectors. Due to a rising demand for their exports and a greater pricing power in basic chemicals than most specialty and fine chemical products7 [330, 331], the French and German chemical industries have increased turnover and value added in the last decade8 (based on [49, 319-321, 327, 332-336]). Swedish and UK industries have not been performing ÃÊÜiÆÊÛiÀ>Ê}ÀÜÌ
Ê
>ÃÊÃÌ>}>Ìi`ÊÛiÀÊÌ
iÊ>ÃÌÊwÛiÊÞi>ÀÃ9. All four countries face increasing international pressure, which has already
7
Profit margins for producers of specialty and fine chemicals are often squeezed between price increases in basic chemicals/raw materials and large industrial/consumer customer demands for low prices [331].
8
Data from the Organisation for Economic Co-operation and Development (OECD) or Eurostat on turnover and value added are insufficient to compare annual percentage changes over the last decade across the countries due to missing data points or differences in definitions of the chemical industry.
9
This fact is often masked in published statistics by the inclusion of the pharmaceutical sector in aggregated data reports.
108
National Approaches resulted in a general decline of the growth rates in the EU chemical industry over the last decade [49, 319, 337]. As a result of the number of companies operating in the chemical sector, one would expect that it is relatively easy to co-ordinate, harmonise and regulate chemicals across Sweden, but is difficult to do so in the UK. This may be true, but each national regulatory approach also differs due to historical and cultural reasons. Ultimately each approach seeks to safeguard human health and the environment while sustaining the international competitiveness of its national chemical and manufacturing industries. There are several advantages and drawbacks to each approach that depend only partly on the size and structure of the industry.
4.2 Regulatory Approaches France, Germany, Sweden and the UK exhibit contrasting approaches to chemical risk management. France and Germany predominantly operate through prescriptive regulation based on technical standards (France – [338-340]). Results from interviews10 indicate that the wider political debates on chemicals policy in France and Germany are thereby generally avoided, as is the need for conducting detailed socio-economic analyses of risk management or policy measures. In comparison with a technology-driven approach, Sweden’s risk management strategies focus on pollution prevention, especially by advocating the substitution principle [262, 341]. The analysis of interviewee responses found that by adopting a hazard approach to risk management, Sweden also avoids the need for carrying out socioeconomic analyses. Although assessments of the costs of alternative risk management measures enter the regulatory equation at the plant-
10
To create a complete and in-depth understanding of national policies and practices, interviewee responses have been combined with publicly available literature, such as policy documents and research studies. Summaries of the interviews are included in a separate annex, available upon request. Overviews of the national policies in Appendix 4.1 provided a further source of material that has been incorporated and referenced in the research findings.
109
Framework for Chemical Risk Management under REACH level in Sweden [189], interviewees reported that these are hardly present when setting Swedish chemicals policy objectives. In stark contrast to the other three countries, the UK takes a risk–benefit approach that relies on socio-economic analyses of alternative policy or risk management options [342, 343]. The following sections of this Chapter will explore national regulatory approaches and how these approaches have evolved. Contrasts between the national approaches are most evident by the fundamental differences in the roles of the national actors (an overview of the findings are shown in Table 4.2). The resulting relationships between regulators and the stakeholder groups surveyed in this research will also be examined (indicated by shading in Table 4.2).
Regulators
Industry
Trade unions Environmental/ Consumer NGO
France
Germany
Sweden
UK
Technical regulation aimed at protection through comitology
Technical regulation aimed at protection through comitology
Technical regulation aimed at prevention
Risk–benefit regulation through consultation
Consultation
Technical support
Technical support
Consultation
Technical support
Cost–benefit approach through consultation
Technical support and policy consultation
Cost–benefit approach through consultation
Consultation
Inform consumers
Influence companies and inform consumers
Consultation
Table 4.2 Primary roles and relationships of national actors.
Shading qualitatively represents relationships between organisations and regulatory authorities, or between regulatory
110
National Approaches >ÕÌ
ÀÌiÃÆÊ `>ÀiÀÊ Ã
>`}Ê `V>ÌiÃÊ >Ê V«>À>ÌÛiÞÊ
}
iÀÊ level of conflict. France - Technical Regulation based on Statistics Before REACH, France is the only one of the four countries that does not appear to have had any national chemicals policy debate. A centralised French government guides all aspects of policy and decision-making. French authorities carry out in-depth risk assessments that feed into defined administrative responsibilities for highly structured technical implementation. Founded within strong social security and industrial insurance schemes, statistics on occupational health and industrial accidents guide risk management activities. After direct risks to humans resulting from industrial activities or accidents, consumer protection is placed second in terms vÊ«ÀÀÌÞÆÊiÛÀiÌ>Ê«ÀÌiVÌÊÀÊ«ÕÌÊ«ÀiÛiÌÊvÀÊ industrial or product emissions holds relatively little weight. In fact, it was only in 2005 that an environmental charter was introduced into the French constitution11 [344]. Historically, consumer regulation has focussed on direct risks to the general public, so the environmental and indirect human health consequences of product use have been regulated only through waste regulation. Environmental protection is otherwise controlled through industrial and hazardous waste regulation, which can easily neglect the contribution of environmental emissions from products and smaller companies – especially if these manufacturing sites are not included within statistical data collecting schemes. Although the French government is highly centralised, interviewees described how a considerable level of decentralisation is present for regulatory enforcement. This provides regulators with some flexibility for a more risk–benefit approach at local levels based on technical and economic considerations for industrial activities and waste management. The 11
The charter establishes that citizens have the right to live in a healthy environment and, together with the State, have a responsibility to protect ecosystems in addition to public health.
111
Framework for Chemical Risk Management under REACH risk–benefit approach at a local level is further facilitated by the use of statistics on industrial accidents, occupational safety and environmental emissions to guide broad policy objectives. Germany - Technical Regulation based on Standards Of the four countries, Germany takes the most legalistic and technical approach to regulating chemicals. Preference for command and control measures in German environmental regulation are often explained by the desire to secure a ‘Rechtsstaat’ ( ‘law-based’ state) after World War II and the dominance of lawyers in German public administration [345]. Federal technical standards are seen as a tool for achieving a level playing field across the Länder [345]. Interview responses indicate that the availability of technical and scientific resources from predominantly large chemical companies and large insurance institutions facilitates setting regulatory standards 12. A focus on charges on chemical emissions [346] and insurance schemes [230] then provide the incentives for companies to adopt beyond-compliance (i.e., preventive) practices. In addition to product standards, best available technology, emission and occupational exposure limit values, Germany uses technical regulations to establish alternative chemical and working practices13 [206]. German regulators interviewed often stressed that the legal obligation for substitution under the European Community Chemical Agents Directive [269] features strongly in German risk management. By limiting the application of substitution to the workplace, Germany has avoided the complex issue of assessing and weighing the lifecycle impacts of alternative chemicals in industrial processes and products14. More recently, German regulators have 12
With about 150 companies each employing >500 employees, German chemical companies have the largest average employee size in Europe [40].
13
For example, rules 602 to 619 of the Technical Rules for Hazardous Substances (TRGS) published by the Federal Ministry of Economics and Labour in the Federal Labour Gazette concern ‘restriction on use, substitutes and substitution of processes or technology’.
14
Technical guidance specific to substitution of chemicals in the German workplace incorporates environmental hazard classifications as part of its selection criteria (TRGS ÀÕiÊ{{äÆÊÀiviÀÊÌÊ«ÀiÛÕÃÊvÌÌi®°
112
National Approaches experimented with creating a list of ‘recommended’ dyes for the use in the textiles industry selected according to hazard criteria rather than full lifecycle assessments (LCA). Although the product listing resulted in a very positive response from chemical users, recent tight budgetary constraints have limited the future development of similar schemes. Sweden - Hazard Regulation based on Bans A history of trans-boundary pollution sets the background to Sweden’s chemical concerns (see [347] for a brief overview). Emissions arising from international chemical production continue to be of concern, but Swedish chemicals policy is increasingly focussed on preventing diffuse emissions and direct health risks arising from (mostly imported) consumer products [348]. Interviewees asserted that the combination of a strong environmental government and powerful trade unions has resulted in Sweden adopting a precautionary approach to chemical risk management in the workplace. Since 1970, unlike their German or British counterparts, Swedish trade unions have been empowered to stop dangerous work practices pending decisions from the Labour Inspectorate [349]. Results from the interviews indicate that action on banning a chemical is far more easily enacted in Sweden than France, Germany or the UK, even if an official risk assessment indicates a need for further testing. Although a hazard-based approach acts as a guiding principle behind Swedish chemical risk management, exposures and risk characterisations enter the risk assessment process. In some cases, a hazard combined with a certain use can suffice for identifying a need for immediate and stringent regulatory action. If a carcinogen in an article presents exposure to consumers and an alternative less hazardous substance that carries out the same function is available, Swedish regulators would certainly aim to ban the use of the carcinogen. In such a case, Swedish regulators do not see the need for conducting an in-depth evaluation of the number of persons that may contract cancer or the resulting business impacts of implementing a restriction. Arguably, avoiding just a handful of cancer cases would outweigh the costs to industry which, in any case, can substitute the 113
Framework for Chemical Risk Management under REACH chemical. In other words, Swedish regulators view such a decisionmaking context as not necessitating a quantitative cost–benefit analysis. By prioritising regulatory actions on chemicals that present imminent risks, the implication is that the effect of such bans on the performance of manufacturing processes or final products should be negligible when compared with the overall benefits to health or the environment. Despite its hazard-based approach, Sweden sets occupational exposure limits for most dangerous chemicals, even if they are extremely hazardous to human health. Sweden also regulates through setting environmental quality standards for hazardous substances even though this is seemingly contradictory to its valued goal of establishing a ‘non-toxic’ environment, free of ‘man-made’ chemicals, by 202015 [350, 351] (Appendix 4.1). UK - Risk–Benefit Regulation based on Negotiation Comparatively, the research identifies that UK regulators and companies are given the highest degree of flexibility in considering local environmental, social and economic conditions. Each regulatory administration may develop its own chemicals policy (e.g., [352, 353]). It is argued that the range of specialists and administrations, with a mix of disciplinary skills and experiences, involved in UK environmental policy creates networks for a wide range of organised societal interests to express their voice [354, 355]. To accommodate different approaches, preference for objective rather than prescriptive regulation underpins the UK risk–benefit approach (see also [356]). With multiple regulatory agencies at local and national levels, responsibilities for implementation are further divided across the devolved administrations of England, Northern Ireland, Scotland and Wales. Choosing environmental quality standards above rigid centralised emission limit values has historically emerged from 15
Swedish chemicals policy recognises the ability of organisms to degrade or detoxify ÃÌÊ
>ÀvÕÊV
iV>ÃÆÊÌ
iÊÌiÀÃʼÌÝV½Ê>`ʼvÀiiÊvÊ>>`iÊV
iV>ýÊ>ÀiÊ not given detailed definitions.
114
National Approaches its decentralised regulatory structure coupled with profiting from opportunities for the dilution of chemicals released into rivers when they reach the sea [357]. Divided responsibility and shared decision-making among the many UK regulatory administrations tends to require detailed assessments of all the implications of a given national policy or regulatory measure [358]. Recent experiences with risk management appear to have reinforced the UK risk–benefit approach. For example, a UK regulator explained that the need for conducting an extensive indepth evaluation of the use of certain brominated flame-retardants resulted from a clash between detailed regulatory requirements under chemicals and fire-protection law, which were not found under other EU national legislation16. Socio-Economic Analysis and Stakeholder Consultation The four national approaches vary considerably in terms of how the regulatory authorities evaluate the socio-economic consequences of regulation, particularly with regard to stakeholder consultation. The two French regulators interviewed expressed the view that their colleagues have little experience in carrying out socio-economic analysis on chemical matters. In comparison, German regulators described avoiding socio-economic analyses because the results are always based on many assumptions and therefore easily subject to political attack (see also [359]). To avoid quantitative regulatory impact assessments at the EU level, Germany contributed a method for carrying out trend analyses to the technical guidance document (TGD) that qualitatively and semi-quantitatively weighs the socio-economic implications of control options [360, 361]. This German approach closely resembles Swedish evaluations of the impacts of chemical regulation except that, according to interviewees, regulators from the Chemicals Inspectorate (KemI) rarely see potential business impacts of 16
The UK therefore deemed it necessary to balance the risks to health from the brominated flame-retardants with the prevention of fires, even though alternative technologies or flame-retardants were available. Today, penta- and octa-BDE are banned in the EU, although the flame-retardant deca-BDE is still subject to ongoing investigation [311].
115
Framework for Chemical Risk Management under REACH regulation in Sweden as negative. Sweden’s chemical industry lobby explained during the interviews that it lacks the political influence or resources that German and UK industry associations inherit by being larger chemical producers. One stakeholder representative described the Swedish trade unions as taking a more long-term perspective than other EU countries, such as the UK, which tend to focus on short-term regulatory compliance costs rather than long-term benefits. In terms of stakeholder consultation and participation in regulatory chemical risk management, the UK is perceived by many interviewees as the most progressive of the four countries, perhaps even in all of Europe. Since 1999, a Chemical Stakeholder Forum has been directly advising the UK government on chemicals policy and risk management strategies [362]. Although structured stakeholder participation in Swedish and Germany chemicals policy appears to be limited, certain stakeholder groups are heavily involved with developing and implementing chemical risk management measures through various platforms. For instance, Swedish Environmental nongovernmental organisations (NGO) are particularly involved with eco-labelling [363] and trade unions provide in-house safety support to SME [364]. In Germany, insurance institutions and industry trade associations set standards with regulators through technical institutes and advisory committees [206, 365-367]. French stakeholder consultation and participation is dictated by the French state. Interviewees explained that organisations are selected by the state through a process of ‘invitation’. Evidence from the interviews indicates that French trade unions and trade associations follow, conform with, and facilitate political decision-making. For instance, French industry and trade unions gather statistics that can be used by regulators. This fits a ‘corporatist model’ that prevails in the French government where stakeholder associations serve to relieve the state of administrative burdens [368].
116
National Approaches
4.3 Regulatory Administrations National administrative structures tend to embody and strengthen the regulatory culture of each country. The extent that regulatory administrations separate, or integrate, policy and science appears to be a key differential. Sweden is the only country with a single central chemical administrator (KemI) that holds primary responsibility for policy and regulation. In France and Germany, chemicals policy lies within the domain of the ministries, with federal technical and scientific institutions providing scientific and technical support. Policy therefore guides scientific standard-setting at arm’s length in France and Germany, while Sweden combines prescriptive technologybased legislation and objective policy approaches through a single regulator. For the UK, decentralised government departments, agencies, and other institutions are each responsible for developing and implementing independent policies. Objective-based legislation therefore enables UK-competent authorities to use their own devices to achieve set targets. Sweden was the first to recognise that the combination of many separate agencies (water, nature conservation, air protection) into a central environmental agency can provide efficient and effective environmental protection [369]. It was seven years after the creation of the Swedish Environmental Protection Agency in 1967 that a central environmental agency was created in Germany [370] and at least two decades before one was set up in the UK17 [371, 372]. France does not have an equivalent to the environment agencies found in Germany, Sweden or the UK. Decision-making and policy development stem directly from Inter-Ministerial Committees (IMC). The division of the roles and responsibilities between chemicals policy and risk management science forms the most prominent distinction between the administrations of the four countries. 17
Arguably, creation of the HM Inspectorate of Pollution in England and Wales, and Scotland, resulted from the need to control environmental releases to different media through a central regulator and dates to 1987 [372-374]. It was not until 1995 that Environment Agencies were established across the UK to bring together a wider set of environmental controls to air, water and land than under the Pollution Inspectorates [375].
117
Framework for Chemical Risk Management under REACH France In France, the Ministry of Ecology and Sustainable Development (MEDD) organises chemical regulation activities, whereas the Ministry of Economics, Finance and Industry (MEFI) co-ordinates roles between the various ministries. The Ministry of the Interior (MI) decides on the legal aspects of market restrictions on consumer products 18 and responsibility for the protection of the public from environmental emissions ultimately lies within Ministry for Employment and Social Affairs19 (MESA). All political and regulatory decisions occur via IMC, where the clear division of roles minimises the potential for conflict. Responsibility for decisions is thereby also divided equally between the ministries, but transparency and accountability for decision-making also diminishes. Occupational and consumer protection are seen as technical exercises in France. The National Institute of Research and Security (INRS) and the Institute for the Industrial Environment and Safety (INERIS) may be aptly described as state apparatus to control costs to the state of poor regulation. INERIS plays little part in the development of risk-reduction strategies (RRS) because consumer protection and environmental protection follow directly from the risk assessment process. INERIS feeds relevant information directly to the MEDD. Because MESA acts only as a statistical and economic branch of occupational protection, responsibility for the development of occupational RRS is left to the INRS. During implementation, roles are clearly divided through a hierarchical structure in France. Regional Environment Departments (DIREN) and Regional Departments for Industry, Research and the Environment (DRIRE) execute decisions of the ministries which are chiefly communicated via the MEDD. At this stage, the INERIS and INRS provide the technical support to the Regional Departments 18
The Ministry of the Interior otherwise focusses its chemical-relevant activities on market surveillance and the recall of dangerous products, which generally falls outside the scope of Directive 76/769.
19
Within MESA, responsibility is located in the Directorate-General for Consumer Health (DGS).
118
National Approaches and their constituent local authorities. Implementing the restrictions aimed at consumer protection resides directly with the Ministry of Consumption (MCC), which has its own team of inspectors. Based on the interviews, there appears to be little scope or need for examining relationships between the regulatory administrations in France. The remit of each administration is clearly defined and no individual administration is responsible for developing policy. There are therefore no overlapping roles or responsibilities of the regulatory administrations. There appear to be no subsequent confrontation or conflicts between administrators. Germany In Germany, federal institutions support the Ministries and Länder only on scientific and technical matters. Two ministries are responsible for chemicals policy and decision-making: the Ministry for the Environment (BMU) and the Ministry of Economics and Labour (BMWA)20. Day-to-day aspects of chemical risk management revolve around the Federal Environment Agency (UBA) and the Federal Institute for Occupational Health and Safety (BAuA) 21, although the ministries provide legal support to companies and regulators for compliance issues. Responsibilities are shared. The Ministry for the Environment co-ordinates RRS activities of the BAuA which otherwise functions under the competence of the Ministry of Economics and Labour. Because the Environment Ministry covers poisonings, it has central responsibility for consumer protection22. 20
The Ministry for Labour and Social Affairs and the Ministry of the Economy merged in 2003 to form the Ministry of Economics and Labour. Before this move, both ministries were involved in EU decision-making.
21
For example, the federal institutions are involved during the development of risk reduction strategies (RRS). The Federal Institute for Risk Assessment (BfR) and the Federal Institute for Materials Research and Testing (BAM) may also be involved in chemical risk management processes.
22
Although decision-making concerning marketing and use restrictions aimed at consumer protection involves the Ministry for Consumer Protection (BMV), it does not have a major role. The chemical management activities of its supporting Institute for Risk Assessment (BfR) have recently been transferred to the BAuA.
119
Framework for Chemical Risk Management under REACH Co-operation amongst the Länder, and between the federal institutions and the Länder, appears successful. While the Länder and municipal authorities are responsible for implementing decisions made by the federal ministries, they play little part in the decision-making process. Increased decision-making at the EU level does not appear to result in significant discontent among the Länder over their diminishing role in chemical risk management decision-making, probably because it is seen as creating a level playing field across the EU. Regulation, monitoring and enforcement between the Länder is chiefly communicated and co-ordinated by the federal UBA and BAuA. Sweden Since the creation of the Swedish Environmental Protection Agency in 1967, Sweden has continued its centralisation relating to chemical issues. Created in 1991, KemI directly advises the small Ministries of the Swedish government on chemicals policy and the Environment Ministry23 on marketing and use restrictions. In collaboration with the Swedish Work Environment Authority (SWEA) and the Swedish Environmental Protection Agency (SEPA), KemI develops and implements RRS. Neither the SWEA nor SEPA appears to be involved in selecting RRS. Instead, KemI represents Sweden at the European Commission Working Group for Risk Reduction Strategies and the Limitations Working Group24. Aided by the Swedish Consumer Agency for enforcement, KemI also holds primary responsibility for consumer protection from chemicals. A particular feature of the Swedish regulatory infrastructure is KemI’s Product Register, which lists all chemicals produced or marketed in its territories. This register is a predecessor to the one that will be created under the REACH regulation.
23
In January 2005, a Ministry of Sustainable Development replaced the Environment Ministry, combining it with the Energy Ministry.
24
The Commission Limitations Working Group develops the specific proposals for EU marketing and use restrictions.
120
National Approaches UK The UK has the most complex regulatory structure of the three countries, even when compared with the German federal structure. Although the devolved UK administrations have a negligible role in terms of risk management decision-making, a regulator interviewed explained that implementing EU legislation must account for potential differences between the administrations, as well as separate legislative systems across UK territories. Regardless of the fact that central government delegates significant responsibility to its departments and agencies, informal inter-ministerial committees and each minister responsible for the departments (referred to as ‘Whitehall’) make the final decisions on marketing and use restrictions. Together with the Department for Trade and Industry (DTI), the Department for the Environment, Food and Rural Affairs (Defra) provides the major input to decision-making at this level25. The Health and Safety Executive (HSE) and the Environment Agencies of the devolved authorities (EA) also have major roles during the development and selection of RRS during national and EU decision-making. Enforcement of restrictions on marketing and use is divided between the HSE, DTI and Trading Standards. The Department of Health is involved only in consumer risk assessment activities. The principal regulatory authorities of the four countries are summarised in Table 4.3. To facilitate comparison, administrations have been divided according to their roles in the development and selection of risk-reduction measures, and their responsibilities with regard to implementing risk-reduction measures. Sweden and the UK present opposite extremes of centralisation. KemI is present in every activity listed in the table, whereas the roles and responsibilities of UK administrations are distributed across 25
Although the DTI is primarily responsible for consumers and holds some remit on marketing and use restrictions, Defra appears to be involved with most risk reduction strategy decision-making.
121
Risk management activities
France
Germany
Risk-reduction strategies
Marketing and use restrictions
Developing
Selecting
Implementing
Decisionmaking
Implementing
MEDD, INRS
IMC
DIREN, DRIRE
MEDD, IMC, MI
DIREN, DRIRE, MCC
Länder, local authorities
BMU and BMWA
Länder
SEPA, SWEA, local authorities
Ministry for Environment
UBA, BAuA
SEPA, SWEA Sweden
KemI
UK
Defra, EA, DTI, HSE
‘Whitehall’, Defra, HSE, DTI
EA, HSE, local authorities
‘Whitehall’
HSE, DTI, Trading Standards
Table 4.3 Principal roles and responsibilities of the national regulatory authorities
Framework for Chemical Risk Management under REACH
122
Regulatory Roles and Responsibilities
National Approaches these categories. In comparison, France and Germany clearly allocate specific roles and responsibilities according to activity. Descriptions of the French regulatory processes appeared to be wellunderstood by all the interviewees in France. With the exception of restrictions on consumer products, enforcement ultimately occurs at the local level through the Regional Departments and social insurance inspectors. Statistics provided by the French industry trade associations helps prioritise enforcement activities while guiding political goals at the ministerial level. Reports from German regulators indicate that distinguishing between political and scientific activities creates a good working environment, where everyone is fully aware of the others’ remit. An even distribution of resources between the authorities enables each authority to fulfil its responsibilities without depending on another for further input. Shared structures allow for a close work environment with good communication and co-ordinated prioritisation between administrations. Two recent administrative reorganisations have increased German centralisation. The first is straightforward, involving transferring responsibility for consumer risk management evaluations from the Institute of Risk Assessment (BfR) to the Institute for Occupational Health and Safety (BauA) in 2003/2004. The rationale behind this shift is that occupational and consumer risks can be evaluated simultaneously during the substance-by-substance approach to EU risk assessments and RRS. Regulators also explained that the second reorganisation was in response to high unemployment. In 2003, the Ministry of Labour and Social Affairs (BMAS) merged with the Ministry of the Economy (BMW) to form the Ministry of Economics and Labour (BMWA). Typically, the Ministry of Labour and Social Affairs generally supported any strict policy or management measures proposed by the institutes or other ministries, whereas the Ministry of the Economy objected if costs to industry were deemed to be high. German regulators expressed the view that these discussions now occur internally within the new ministry, where the old voice of the
123
Framework for Chemical Risk Management under REACH Ministry of the Economy resonates louder, meaning that the Ministry for the Environment has lost a strong partner. Sweden’s centralised chemical authority means that there is never doubt as to where or how to address relevant queries or concerns. Although good working relations between the agencies exist, responses from the interviewees indicated that co-ordinating the relevant chemical risk management activities of KemI, SEPA and SWEA appears strained at times, resulting from different priorities within the administrations. For instance, KemI does not need to consider the concerns associated with the ergonomic risks of occupational settings or physical safety issues arising from consumer concerns. As a result of differences in prioritising risk management activities, the implementing agencies have little resources available to input into the official risk-reduction process. Not sharing regulatory power means that the implementing agencies have very little influence over KemI’s decision-making processes26. It is therefore perhaps not surprising that a regulator complained that companies do not always co-operate with SEPA on chemical issues. More co-ordinated prioritisation between the authorities would improve the efficiency and effectiveness of the existing administrative structure. The very decentralised administrative structure in the UK could also benefit from more communication and co-ordination. Interviewees noted that a lack of systematic procedures for communicating on current activities has caused duplication of work within the administrations. There are so many chemical-related activities across the various administrations, including the Department for Health, the Food Standards Agency and the Health Protection Agency, that it was difficult to gain an overview of the chemical risk management decision-making processes or whom to contact with respect to a particular query when conducting the research. One regulator was Õ>LiÊÌÊÃÌÊ>ÊÌ
iÊÀiiÛ>ÌÊÀÃÊ>>}iiÌÊ>ÕÌ
ÀÌiÃÆÊÀi}Õ>ÌÀÃÊ generally appeared uncertain as to the precise role of customs and 26
SEPA and SWEA work under strict mandates from the relevant ministries, whereas KemI has a legislative structure that grants it ‘free rein’ so long as it meets political goals set by the Ministry of Sustainability.
124
National Approaches iÝVÃiÆÊ>`ÊiÊÌiÀÛiÜiiÊ`iÃVÀLi`ÊÌ
>ÌÊ>ÊiLiÀÊvÊÌ
iÊ}iiÀ>Ê public would probably be redirected via individual members of staff following a specific query. For internal and external communication of roles and activities, the UK may therefore want to consider the use of a matrix system similar to those used in corporate safety, health and environment management. Working relations in the UK appear to be excellent. Regulators expressed the view that a high level of trust exists between the authorities and that there is a well-rounded, although general, view on the primary responsibilities of each agency. There is no lack of UK Policy-makers seek to attain consensus
Policy-makers react to problems
Policy-makers anticipate problems Sweden
Germany
France Policy-makers seek to impose decisions
Figure 4.1 Countries placed according to dimensions of policy style (adapted by Colebatch [375] from Richardson [376])
high-level expertise within the EA and the HSE, but one regulator complained that the political and administrative layers dilute tremendous amounts of exceptional work produced or commissioned by the UK authorities.
125
Framework for Chemical Risk Management under REACH
4.4 Policy Styles and Implications for Risk Management 4.4.1 Policy Styles Analysing national approaches allows for evaluation of the policy styles of the regulatory authorities. Because policy style influences how authorities set and implement regulation [377], they determine the efficiency and effectiveness of chemical risk management. An analysis of policy style also provides a basis for examining why the relationship between the national actors varies to such a large extent between the four countries. The overall position of each national approach is shown schematically in Figure 4.1 [376, 378]. The results of the interviews summarised here indicate that KemI imposes decisions on stakeholders and other regulators. This ‘topdown’ approach to regulation is also reflected by the importance that the Swedish regulator interviewees attached to several chemical authorisation schemes independently operated by KemI, SEPA and SWEA27. KemI simultaneously anticipates environmental problems, which is consistent with a hazard-based approach. Sweden therefore depends on application of environmental management principles, such as pollution prevention, substitution and precaution. In comparison, the UK is clearly based on negotiation, thereby seeking to attain consensus, forming the basis of its risk–benefit approach. France reacts to problems and imposes decisions based on in-depth risk assessment with activities prioritised according to industrial statistics on accidents, occupational safety and 27
Several of the substances subject to ‘phasing-out’ in Sweden must be authorised by KemI by issuing general exemptions (e.g., mercury) or on a case-by-case application basis (e.g., trichloroethylene). The use of several carcinogenic, mutagenic or reprotoxic (CMR) and sensitising substances must also be authorised on an individual basis by SWEA [379]. Although all EU countries operate permit schemes under Integrated Pollution Prevention and Control (IPPC), the regulator interviewed from SEPA noted the importance of achieving chemical control through this piece of legislation, and other interviews reported it being more strictly operated in Sweden than other Member States. By contrast, German regulators see IPPC as promoting a level playing field.
126
National Approaches environmental emissions. Germany’s general approach appears to Ài>VÌÊÌÊ«ÀLiÃÆÊ>ÊÌiV
V>ÊÃÕÌÊV>ÊLiÊ>««i`ÊÞÊViÊ a risk has been identified. Although KemI can identify risks, most Swedish interviewees indicated that it appears limited in its ability to incorporate these principles in a systematic way that would enable it to establish practical and realisable goals. Swedish regulators and industry described problems that KemI encounters with setting prescriptive-based regulation to achieve general objectives. Specifically, phasing-out substances requires detailed scientific and engineering evaluations, which often only follow regulatory bans. Interviewees often reported that, as the net result of not involving chemical companies in decision-making, KemI finds itself having to take company concerns into account during actual implementation. At this point Swedish risk management closely resembles Germany’s technical approach, but is open to individual interpretation as to where to draw the regulatory line between prescriptive and objective. All Swedish interviewees agreed that its national approach has instilled a strong chemical safety management culture in companies, but whether Sweden successfully prioritises its regulatory activities is a recurring issue. In addition to the consumer association representative reasoning that other household risks are more prominent and the SWEA regulator expressing greater concern on the inhalation of ‘non-hazardous’ powders than some chemicals, several Swedish interviewees described the selection of eight substances and five groups of chemicals28 for ‘phase-out’ as more political- than riskbased [380]. While interviewees agreed that there was a need to limit exposure to these chemicals, several questioned whether the level of resources dedicated by regulators and industry is justifiable. These thirteen ‘unwanted chemicals’ have been a regulatory priority for over 15 years [341]. For instance, the constant need for reviewing case-by-case authorisations for the use of trichloroethylene is reported by the industry representatives as administratively burdensome while 28
Methylene chloride, trichloroethylene, tetrachloroethylene, lead compounds, mercury, cadmium, organotin compounds, chloroparaffins, phthalates, nonyl phenol ethoxylates, arsenic, creosote, brominated flame-retardants.
127
Framework for Chemical Risk Management under REACH not achieving risk reduction [380]. The targeting of carcinogens over mutagens and reprotoxins29 may explain why the volume of carcinogens used in Sweden decreased in the late 1990s whereas the volume of other dangerous substances appears to have increased significantly (based on data from [381]). Instead of general policy aims, UK’s ‘objective’ regulation involves achieving targets. Prioritising risk management activities based on statistics holds many weaknesses, especially with respect to creating a ‘culture of safety’. Risks that are difficult to identify or measure, such as reproductive toxicity, can easily be neglected30. The use of statistics may also shift attention from knowledge, understanding and control of chemical risks within companies to performance based on narrow benchmarking activities or ‘league tables’ [382]. One regulator explained that conflicts between defining strategies for attaining objectives and allocating resources occur when objectives are developed from a ‘top-down’ approach while attaining the objective follows from the ‘bottom-up’. Unless evidence indicates that statistics can promote holistic approaches to chemical risk management and address the general lack of understanding of fundamental chemical safety issues in SME, it is likely that the recently proposed ‘Hampton Review’ risk-based strategy for conducting industrial inspections will make matters worse [383]. According to this over-arching policy, enforcement activities should also be prioritised according to statistics. The UK is also experiencing a dilemma with regards to its consensusseeking policy. While the UK is the only of the four countries to have a Chemical Stakeholder Forum, UK Environmental NGO have complained that the forum is ‘more talk than action’ [384] and, 29
A Swedish regulator’s statement that Sweden has focussed on carcinogens is supported by the fact that the lists of substances subject to SWEA use-specific authorisation [379] and subject to phase-out contain more carcinogens than mutagens and reprotoxins by volume.
30
For instance, as reported by two interviewees, current Health and Safety Executive (HSE0 priorities are based on statistics for carcinogenicity, respiratory illnesses and skin disease to the exclusion of mutagenic or reprotoxic effects.
128
National Approaches according to one regulator, these members have also asserted that the government ‘does not follow Forum advice’. In response to the latter criticism, the regulator stated that some stakeholder members have misunderstood their role. In short, the Forum advises government but represents only one input to decision-making. It is also difficult for government or companies to follow stakeholder action when the Ã>iÊV
iV>ÃÊ>ÀiÊLi}ÊÃÕLiVÌÊÌÊ 1ÊÀi}Õ>ÌÀÞÊ`iVÃ>}ÆÊ variation in national legislation would have to be broadly concurrent to a forthcoming EU directive. In this respect, it is unfortunate that the Forum has selected the same priority substances currently undergoing EU evaluation. An interviewee reported that Friends of the Earth, Greenpeace and the World Wildlife Fund anticipate withdrawing from this advisory body. The legitimacy of UK consensus building is questioned by a further four observations. First, officially published UK consultations are often conducted after major steps in the EU decision-making process have been completed31. Second, the UK government recently presented an unsatisfactory response to recommendations from the Royal Commission on Environmental Pollution (RCEP) for changes in UK chemicals policy, administration and regulation [385]. Third, a major industry trade association representative has expressed the view that the UK government is not following industry advice or heeding industry’s concerns. Fourth, data from the interviews expose two examples when incorrect scientific data was used to justify politically predetermined courses of action 32. Evidence 31
Publication of a proposal for a Directive by the European Commission follows general Member State support of the legislation in Risk Reduction Strategy and Limitations (i.e., restrictions) Working Groups. Therefore, an example of the UK running a consultation after the Commission had proposed a Directive on 28 April 2004 is its consultation on restrictions of toluene and trichlorobenzene beginning in September 2004 and ending in December 2004 (see http://www.defra.gov.uk/corporate/consult/tcb-toluene/index. htm). Another example is the Defra consultation following the Commission proposal for restrictions on nonylphenol and nonylphenol ethoxylates (http://www.defra.gov.uk/ environment/consult/nonyl/index.htm).
32
Several interviewees described the UK position on dichloromethane as mired by industrial business interests, which involved delaying regulation by referring to erroneous data. A single UK interviewee reported the other case, describing an occupational health committee delaying regulatory action for >10 years due to (non-declared) business interests interfering with the decision-making process.
129
Framework for Chemical Risk Management under REACH therefore indicates that despite the claims of open consultation and consensus building, the UK needs to establish a method to improve the transparency of decision-making at Whitehall. The government should make clear how it weighs different views and perspectives on a given policy or measure. Exclusion of French and German environmental and consumer NGO from setting technical standards as a result of their lack of expertise or resource availability does not appear to cause dissatisfaction. In France, interviewees reported that this is because the NGO are formally ‘invited’ to the multitude of committees that provide decision-making support to the ministries. REACH has triggered what one interviewee described as ‘mandatory participation’ of NGO in two newly created French chemical Working Groups33. Given the political nature rather than technical aspects of current REACH debates (see Chapter 6), interviewees explained that environmental NGO can participate more than in the previous more technically oriented discussions. An interesting phenomenon is that German environmental and consumer groups appear to rely on measuring and reporting technical standards for chemical contents in consumer products34. This information serves as a data source for retailers to withdraw products35. In comparison, Swedish Environmental NGO are very active in eco-labelling activities rather than testing individual products outside of the eco-label award [363]. As a Swedish Consumer NGO representative explained, it is then not uncommon for the general 33
In response to REACH, two Working Groups have been created in France: one examines the political, legal and technical requirements of REACH, the other reviews the sustainability of the French chemical industry.
34
While several consumer magazines are published in each of the four countries, the German Öko-Test appears the most prolific for testing the chemical contents in products. For instance, in 2004, the magazine published tests for >700 DIY, construction and household products, 950 health and fitness products, 750 child products and 1300 more general products (see www.oekotest.de). An interview also reported that German Environmental NGO campaigns on chemicals have focussed on specific product risks.
35
Documented personal communication from regulatory managers of a consumer product manufacturer and an electronics manufacturer in January 2006.
130
National Approaches types of products on retail shelves to be monitored by local consumer groups. According to one interviewee, UK NGO are now moving towards increasing the availability of information on chemical contents in consumer products through UK consumer publications, partly in response to growing consumer requests to the NGO for such information. One interviewee indicated that UK Environmental NGO appear to be at a loss as to how best to proceed.
4.4.2 Strengths and Weaknesses It is hard to imagine Germany adopting an anticipatory approach while it faces short-term economic concerns. Contaminated land remediation has been a focus of German chemical regulatory activity for some time, and presents Germany with a mammoth task, especially after reunification [386, 387]. No interviewee doubted Germany’s strong economic dependence on its chemical industry. In addition to its current economic crisis and high unemployment, Germany’s position in the EU allows for easy relocation of industry into the recently enlarged EU [388]. Although interviewees expressed the view that Germany may be reluctant to legislate on environmental matters, interviewees also reported that technical measures are immediately set and implemented to control imminent occupational health concerns. Problems regulating chemical use in SME occur across the EU proves no exception to these four countries (see Section 2.2.2). A common response of UK, German and Swedish interviewees on this issue is that SME often require a ‘hands-on’ approach to information provision and technical support whereas French interviewees tended to perceive risks as a result of insufficient regulatory monitoring and enforcement [389]. Several interviews confirmed that controlling chemicals in SME in Germany has been facilitated by branch regulation and technical standards [390]. Similarly, two Swedish regulators reported positive results from sector-specific guidance. These Swedish regulators 131
Framework for Chemical Risk Management under REACH expressed dissatisfaction that only a few such documents have been produced in Sweden. Swedish interviewees also described the guidance providing an important tool for facilitating public discussion on chemical use at the local level. German technical bodies and insurance institutions assume considerable responsibility in supporting downstream SME users with recommendations for chemical uses and establishing product standards [206]. German regulators also stressed the important role of mandatory training or licensing schemes for the supply of certain substances to SME and professional users. In Sweden, according to a trade union representative, Regional Safety Representatives36 provide a crucial link in communicating chemical safety information to SME. this includes ‘problem solving’ for occupational health issues [393], see also [394]. Swedish companies also have a greater reliance on personal protective equipment (PPE) suppliers providing information door-to-door than their European counterparts, which may improve the accessibility and transfer of expert chemical safety knowledge to SME. The production and import of all substances must also be registered in Sweden, and uses of certain substances must be authorised by either KemI or SWEA. Such stringent and case-by-case approaches to regulating individual firms do not exist in the other three countries. One interviewee described that employer opposition to the Worker Safety Advisor scheme has halted its development across all industry sectors in the UK [395, 396]. French interviewees described several safety representative state-sponsored schemes in France, but the resources appear far more restricted than in Sweden. There was no evidence from the interviews or literature review to indicate that Germany has such a scheme. Evidence from the interviews indicates that such schemes in France and Sweden link to their strong social health care system, with very high union membership distinguishing Sweden from France. 36
Regional Safety Representatives visit many ‘micro-enterprises’ (fewer than 10 employees) [391]. This activity is made possible by high state provisions heavily supplemented by trade union contributions [392].
132
National Approaches Although considerable effort is made in the UK to provide information to SME, particularly by means of regulatory or trade association publications, the effectiveness of this strategy is questionable. Communicating intricate details of chemical risk management involves complex issues regarding the selection of appropriateness of measures (e.g., gloves made of butyl versus nitrile compounds) that are difficult to convey through general guidance documents. One UK occupational health officer expressed a strong concern that the UK approach may be ‘dumbing down’ chemical safety, and avoids companies reviewing if the use of a hazardous substance or process is necessary in the first place. Even if detailed or direct advice is available, companies may be reluctant to contact regulatory-related (although independent) services in fear of repercussion for noncompliance with existing legislation. The UK system of management appears to send mixed messages to stakeholders. The combination of lengthy stakeholder consultations, in-depth cost–benefit analyses, complex bureaucracy, and final decisions made at a political level appears to limit the incentive for companies taking beyond-compliance action or seeking innovative new solutions. A further result is that much of the responsibility becomes focussed on regulators and industry rather than a wider set of stakeholder associations (e.g., trade unions, occupational insurance companies). Inherent fallibilities in the reliance of statistics present France with some of the same challenges as the UK. Interviewees reported that French regulation has particularly neglected environmental emissions arising from SME. While the reliance on occupational health statistics is necessary for compensation through the state social security system, an interviewee explained that the focus on prioritising regulation on industrial accident statistics was strengthened after a large explosion at a Toulouse chemical plant in 2001. The lack of decision-making powers by French statutory bodies limits their ability to directly promote organised interests in the ‘bottom-up’ manner described above for the UK [368].
133
Framework for Chemical Risk Management under REACH All French interviewees expressed concern at the current levels of occupational protection and the costs to the state, which is catalysing a move towards a more risk–benefit approach to regulation. Several interviewees specifically referred to a problem with ‘victim associations’ in France. These associations target industry activities or specific substances that are believed to have damaged occupational health and which frequently appear in the press. Because dispensation of state benefits to the affected individuals depends on official recognition of negative health impacts as a result of occupational exposures, the victims and their associations have a financial incentive to ensure that they succeed in achieving their goals.
4.4.3 EU Decision-Making At the EU level, the reliance on in-depth risk assessment by the British and French is illustrated by the Anglo-French joint evaluation of deca-BDE, which took 10 years under the Existing Substances Regulation 793/93. The ‘completed’ risk assessment concluded that further data are needed before proceeding with regulatory measures, whereas Sweden and Germany had already banned its production and use in many products [397]. Similarly, when short-chain chlorinated paraffins (SCCP) were identified as posing unacceptable risks to the environment, the UK found it necessary to carry out detailed risk assessments on medium-chain chlorinated paraffins (MCCP) despite their similarity in structure to SCCP [398]. Long before the conclusion of the EU-wide risk assessments conducted by the UK, Germany had achieved a near complete ban in metal working applications based on technological assessment of their substitution [398]. Similarly, all chlorinated paraffins were part of Sweden’s programme for phaseout. Regulatory action is now being taken across the EU to control SCCP and MCCP, but these measures are not as stringent as under German or Swedish national regulation [399]. Sweden has long recognised that its use of epidemiological evidence differs from many other Member States. It even organised an ‘ad hoc’ EU risk assessment meeting in 1999 to discuss the use of 134
National Approaches different sources of data for classification and labelling of potential carcinogenic, mutagenic or reprotoxic (CMR) substances (e.g., [400]). Another example of differences between the countries at the EU level is Germany’s reduction of all workplace exposures to substances toxic to fertility to the lowest technical limits possible. Currently, EU directives mandate this technical approach to only nonthreshold carcinogens and mutagens [401]. Such a technologically driven approach to regulating non-threshold substances in the workplace was originally opposed by the UK which proposed ‘as low as reasonably practical’ rather than ‘as low as technically possible’ exposure control standards for these substances in 1992 [402]. The different regulatory approaches can complement each other at the EU level. Germany has proven instrumental in developing EU standards [32], including the development of test methods for a recent ban of azocolorants in textiles [403]. If it were not for the UK, Environmental Quality Standards would not have appeared in the Dangerous Discharges Directive [265, 404], thereby missing important diffuse sources of pollution arising from SME or consumer products. Sweden’s precautionary approach has increased EU regulatory attention and scrutiny of several dangerous chemicals [405]. A summary of the contrasting approaches to chemical risk management is presented in Table 4.4. To test the robustness of the analysis, the application of two methodologies to interpret the data have been used and found to yield similar findings37. The next section 37
The use of the different analytical frameworks results in different formats to present the research data. For comparative purposes, the framework of analysis selected by the research methodology and the resulting comparative descriptors appear to be just as relevant as any other. Approaches to regulatory decision-making may be broadly V
>À>VÌiÀÃi`Ê >VVÀ`}Ê Ì\Ê £®Ê >ÕÌ
ÀÌÞÊ >`Ê iÝ«iÀÌÃiÊ vÊ Ì
iÊ >VÌÀÃÆÊ Ó®Ê iÝÌiÌÊ vÊ «ÌV>ÊyÕiViÊÊÃViÌwVÊ`iVÃ>}ÆÊήÊiÝiVÕÌÛiÊ>ÕÌÞÊvÊVÌÀÃÊÊ `iVÃ>}ÆÊ{®ÊViVÌÛÌÞÊvÊ`iVÃ>}ÊvÜ}ÊQ{äÈ{änR®°Ê iÃVÀ«ÌÃÊ of the national approaches are sufficient to create an overview of the results that could be drawn from applying this framework of analysis. Another method for describing organisations combines the types of tools and processes used to support decision-making [409]. To enable this more management-orientated approach to organisational analysis, decision-making is described in terms of its flexibility to incorporate other actors (i.e., degree of authority and collectivity). A report on the analyses is included in the original version of this thesis, as published in 2007.
135
Framework for Chemical Risk Management under REACH of this chapter examines how the evolution of national approaches involves complex socio-economic and cultural phenomena. Germany
France
Sweden
UK
Risk management approach
Technical
Technical
Hazard
Risk-benefit
Policy style
Reacts to problems
Reacts to problems
Anticipates problems
Consensusforming
Regulatory style
Prescriptive
Prescriptive
Objective (policy)/ prescriptive
Objective (target)
Regulatory tools and policy instruments
Administrative structure
Application Technology In-depth risk of policy >ÃÃiÃÃiÌÆÊ >ÃÃiÃÃiÌÆÊ princistandard setÃÌ>ÌÃÌVÃÆÊ ples (e.g., Ì}ÆÊÌÀi`Ê stakeholder prevention, analysis consultation substitution)
In-depth risk assessiÌÆÊÃV economic >>ÞÃÃÆÊ ÃÌ>ÌÃÌVÃÆÊ stakeholder consultation
Central/deCentralised centralised Integrates Separates policy and science and science implementa- through cention tralisation
Decentralised Integrates policy and science at each decentralised level
Central/decentralised Separates policy and science
Table 4.4 Regulatory chemical risk management in France, Germany, Sweden and the UK
136
National Approaches
4.5 Social and Cultural Contexts 4.5.1 Legal Systems and the Role of Experts The legal systems and the role of experts provide insight on how and why different regulatory approaches have developed in each country. A distinction exists between common law found in the UK38 and the civil ‘codal’ law39 of France, Germany and Sweden. Founded on national custom, common law operates through the establishment of judicial precedence [410]. Answers to legal questions in civil law are usually sought by cross-referencing coded legal text. Because any code requires deciphering, the legal systems of France, Germany and Sweden predispose these states to rule by legal interpretation [410]. Consequently, implementing German and Swedish chemicals law requires particular scientific and technical expertise. The reliance on statistics rather than regulatory standards enables France to escape relying on this inherent property of codified law (Sections 5.1 and 5.2). German and Swedish codal law revolve around principles40 ([189, 411, 412), as evidenced by the development of the precautionary principle in Germany since the 1970s [413, 414] and the substitution principle in Swedish chemical law dating to 1985 [415]. In France, some environmental principles were referenced in legal texts in 1997 following the Rio Conference on Sustainable Development, but could be fully introduced only after environmental protection entered the constitution in 2005 [416]. The lack of legal principles in British civil law has seen to cause the comparatively ad hoc and uncoordinated development of its legislative system.
38
Although Scots Law in Scotland is a form of civil law, centralised government in England and modern statutes has introduced many new laws that are identical for both countries [410].
39
Codification can be traced to Roman and Napoleonic Laws [410, 411].
40
The German and Swedish interviewees frequently referred to policy principles, especially in terms of the obstacles and opportunities for their application during regulatory decision-making or implementation. A Cartesian view of the world and human actions fits this codal system for legal governance, as does the Kantian view of 12 physical principles by which the universe functions [412].
137
Framework for Chemical Risk Management under REACH Several UK interviewees described responsibility for chemical management as being divided between regulators and industry, where regulators provide information to enable companies to meet regulatory obligations but companies are responsible for being informed and adopting relevant practices. In Germany, regulators described regulatory standards as necessary to ensure that chemical safety can be understood and followed by the less technically able. The near opposite occurs in Sweden, where companies must be technically adept to interpret and implement regulatory demands. In France, ministerial regulators set broader standards that enforcers and stakeholders must achieve through their own devices. Several Swedish interviewees described KemI as lacking technical experience in industrial settings, a feature that distinguishes it from SEPA and SWEA. Interviewees described the fragmentation of a previously centralised agency41 as causing this divide in expertise. In France, two regulators reported a lack of economic expertise within the regulatory administrations. According to the regulators, this may soon change because France has started to focus on reducing social benefit compensation following occupational illness by conducting occupational protection impact studies [417]. Several interviewees reported that the UK regulatory administrations contain a wide mix of experts and resource skills. While this may promote a more multi-disciplinary approach to regulation than in other countries [354, 418], inclusion of a wider set of experts in decision-making presents challenges. Experts must communicate the intricacies and relevance of their specialisation to other experts. Simply increasing the number of inputs in decision-making creates a more complex task for regulators. In turn, this may reduce the propensity for immediate regulatory action, as evident in the government’s delayed response to bovine spongiform encephalopathy (BSE) [354]. For instance, a lack of long-term epidemiological evidence caused British scientists to initially challenge the USA toxicological-based restriction of the pesticide aldrin [419]. More 41
One Swedish regulator described collaboration between the administrations as hindered by regulators ‘fearing’ that KemI, SEPA and SWEA might be re-combined.
138
National Approaches recently, British scientists have countered evidence on chemicals causing endocrine disruption in human populations, arguing that the subject requires a wider assessment into potential causes and effects [420], whereas German regulators officially stated that sufficient links emerge from scientific reviews to warrant regulatory action [421]. Similarly, the UK Food Standards Agency contests whether ‘phthalates’ may be reprotoxic to humans or present risks to the public [422], contrary to indicative results from a number of scientific sources on several of the most commonly used phthalate substances (e.g., [178, 180, 423, 424]). Because EU legislation is codal and relies on principles, one would expect that implementing EU regulation is easiest in Germany and Sweden, but the level of comitology between the various committees of the many Directorate-Generals of the European Commission involved in EU chemical risk management bears a closer resemblance to administrative structures in the UK or France. These observations begin to explain why each of the four countries frequently struggles with understanding and following EU regulation, as will be further examined in Chapter 5.
4.5.2 Future Socio-Economic Concerns and Chemical Safety High investment capital in basic chemicals production can limit a company’s willingness to change a product or adopt alternative process technologies (Section 2.3), so the average smaller-sized Swedish and UK companies may be more adaptable to changing regulation than the typically larger French and German firms. The net effect is that there may be limited scope for applying pollution prevention once large scale chemical production has begun in France or Germany [425]. If international competition becomes increasingly pronounced in basic organics during the next two decades, as predicted by the OECD [374], France and Germany may need to examine the sustainability of their chemical industries. Economic pressure on French and German organic chemical companies may become particularly pronounced
139
Framework for Chemical Risk Management under REACH if bio-based organics replace traditional petrochemical derivatives [336], as predicted in recent UK chemical business news articles (e.g., [426]). In the short term, maintaining high levels of industrial production in France and Germany may depend on the development of extended producer responsibility. In the longer term, achieving a smooth transition away from a reliance on petrochemicals towards bioproducts is important [427]. Fatalities per 100,000 workers 1996
1997
1998
1999
2000
2001
2002
France
9
7
2
0
2
11*
1
Germany
8
6
14
3
5
6
5
Sweden
0
1
0
0
0
0
0
UK
0
1
3
2
1
1
0
EU average
3
3
4
1
2
3
1
Lost time injury frequency rate converted to 3 days 1996
1997
1998
1999
2000
2001
2002
France
8
8
5
5
5
6
6
Germany
10
9
9
9
8
8
7
Sweden
2
2
2
2
2
2
1
UK
4
4
3
3
3
3
3
EU average excl. DK)
9
9
8
7
7
7
6
Table 4.5 Occupational health statistics in the chemical industry [327] (*31 people died in the 2001 Toulouse accident)
The high operating temperatures and pressures often required for petrochemical basic organic manufacturing [428] could explain the high proportion of incidents in France and Germany (Table 4.5). With the exception of the Toulouse explosion in 2001, the data in Table 4.5. indicate that France has achieved a marked improvement in reducing fatalities over recent years. Injuries in the French industry have
140
National Approaches remained comparatively level, which may be the result of regulatory activities focussing on reducing accidents in larger industrial sites [321] rather than SME. Sweden exhibits a particularly good performance in terms of avoiding fatalities and injuries in this sector. These statistics could support the view that a culture of safety exists in Swedish companies, but chemical risk management in Sweden is arguably simpler for regulators and stakeholders due to fewer total chemicals produced and used within its territories (Section 4.1). This was exemplified during the interview with Swedish chemical industry representatives (paraphrased): A journalist contacted us to ask our opinion on the risks of PFOS, perfluorooctane sulphonate, and the need for regulatory action. We had never even heard of this substance. After some investigation, we found out that it is neither produced nor used in Sweden! Nevertheless, the statistics presented in Table 4.5 must be interpreted with caution. As the evidence in Section 4.4.1 suggests, the comparatively low frequency of fatalities and low lost injury time in the UK and Sweden may be attributable to regulators in these two countries prioritising more readily identifiable effects of exposure to certain hazards (e.g., carcinogenicity or sensitisation) over less apparent effects (in terms of existing epidemiological studies or occupational health incident reports) that result from exposure to other hazards (e.g., reprotoxicity).
4.5.3 Public and Political Catalysts Given the prominent role of the chemical industry in Germany42, it is not surprising that the German public holds one of the highest perceptions of the benefits of the industry in the EU [429, 430]. Several interviewees expressed similar views on the positive public opinion of the industry and described German companies as experiencing less pressure from environmental NGO than the other
42
Bayer Leverkusen is the only European league football team named after a company!
141
Framework for Chemical Risk Management under REACH countries43. Two German interview respondents also explained that relatively low public concern over chemical risks and lack of NGO engagement follows recent reductions in environmental concerns amongst the public (see [431]). Based on an interview with the German Federation of Consumer Associations (VZVB), German NGO take a fundamentally different approach to chemical risks than in the other countries. German NGO focus on informing the public on product choice and physical actions to reduce exposures ( Section 4.1). The German Housewife Association, which organises classes on how lifestyle affects prenatal exposures to common chemicals, provides another example of this pragmatic approach. In stark contrast to generally positive German perceptions of the industry, the public in France and Sweden exhibit the most pronounced negative view [432]. According to French and Swedish industry representatives, potential chemical risks are headline news, but miscommunication or errors made by the media or NGO hardly gain press attention44. While the media propagates the ‘victim association’ phenomenon in France (Section 5.3), the industry representatives described the 2001 explosion in Toulouse as triggering a major negative public response to the industry. By contrast, Sweden appears to have a long history of media campaigns fuelling confrontation between regulators, industry, NGO and the general public (Appendix 4.2). The British public does not exhibit a particularly strong or pronounced opinion of the chemical industry, but its view does tend to be negative [433, 430]. Media attention has recently been drawn to UK NGO studies identifying synthetic chemicals in human blood. One such biomonitoring survey was reported by the WWF-UK interviewee as 43
Ironically, the loss of power for the German Ministry for the Environment (BMU) resulting from the creation of the Ministry for Economics and Labour now means that, according to two interviewees, the BMU needs environmental NGO stakeholder support to achieve its policy objectives. The main Consumer NGO (VZBV) appears to fill some of the gap in chemical issues created by the environmental NGO, but it has closer work-relations with the Ministry for Consumer Protection than the BMU.
44
Organisations and institutions provide networks for risk communication, but the media holds a key position in shaping public attitudes to risk [436].
142
National Approaches attracting the greatest public response the NGO has ever received [71]. Attention drawn by the media may have impacted the UK chemicals policy debate, but individual members of the public also reacted by contacting the WWF-UK to ask for practical solutions to reduce consumer exposure. The WWF-UK appeared unprepared vÀÊÌ
ÃÊÀiëÃiÆÊÌÊVÕ`ÊÌÊ«ÀÛ`iÊÌ
iÊÃ>iÊiÛiÊvÊvÀ>ÌÛiÊ advice that is, for instance, readily available in Germany. With respect to industrial relations, an interviewee described German trade unions as abandoning political support for stringent regulatory actions that may have a pronounced negative economic consequence. This is evident by the chemical industry trade union (IGBCE) and trade association (VCI) joining forces to ease the regulatory burdens of REACH [434]. Exceptions may occur when measures significantly benefit worker health [435], exemplified by three major downstream manufacturing trade unions (IG BAU, ver.di and IG Metall) calling for registration to sufficiently cover occupational safety [434], demands that are unlikely to significantly affect chemical production. Until recently, Swedish environmental NGO and trade unions have been extremely active in the development of chemicals policy [437]. Historical affiliations between these groups and two of the stable six-party political structure explains the strong influence that these stakeholder groups can exert on Swedish politics [438]. Even though the Swedish public has a generally poor perception of the chemical industry, three interviewees reported that public concern on chemical matters has recently decreased45. Interviewees explained that a social change towards a more ‘individualistic’ society as also being 45
Most Swedish interviewees described that the environment holds a ‘personal’ dimension that may not be present in France, Germany or the UK, especially as Swedes often spend time in the countryside. One interviewee further explained that the parents of many of this generation were once farmers. In addition to having a large percentage of open land per inhabitant where ‘everyman’s right’ permits a general public right of way on private property (see [440]), environmental education is very pronounced in Sweden [441]. Consequently, 92% of Swedish children believe that solutions to environmental problems exist [441]. Previous regulation and state-sponsored awareness campaigns have also focussed on the importance of individual action in achieving local and global environmental goals [442].
143
Framework for Chemical Risk Management under REACH responsible for declining trade union memberships. Consequently, Sweden has begun to adopt a pro-industrial competitiveness approach to incorporating socio-economic analyses into political decisionmaking, as described by two interviewees. Apart from the chemical industry, the chemical manager organisation (FECCIA) appears as the most engaged stakeholder group in REACH policy debates in France. Such employer organisations have been described as ‘state-made partners in policy-making’ (see [439]) and research suggests strong ties between regulators and industry46 are present at a personal level [443] which is entrenched in a traditional preference for the state to deal directly with individual patrons [444]. Fragmented trade unions generally have a negligible role in centralised French policy-making [445, 446] and, based on the interviews, state inspectors linked to the social security system overshadow union representatives (see also [447]). Similar to France, UK trade unions are decentralised and tend to avoid engaging with the government [448, 449]. Research suggests that shared economic concerns by employer and employee organisations across the UK have halted a swing from working class to industrial interests [450], which has pushed environmental issues into the background [425, 451]. An interview with the trade union Amicus and correspondence exchanged with the British Trade Union Confederation indicate that their input on chemicals policy is limited to discussion on the potential negative costs to industry.
4.6 Propensity to Change One way of evaluating the propensity of each country to adapt to REACH is to consider the changes that the new EU chemicals policy 46
Educated in one of a few well-established engineering schools, entrance into governmental regulatory (environmentally relevant) positions is limited to graduates of these select schools, where new recruits are quickly ‘taught’ the informal rules of the administration [443]. Close ties in the French industrial setting are further mediated through an unofficial ÞiÌÊ Ü`iÞÊ ÀiV}Ãi`Ê «À>VÌViÊ ÀiviÀÀi`Ê ÌÊ >ÃÊ «>ÌÕy>}iÆÊ Ü
iÀiÊ iÝiVÕÌÛiÃÊ Ê >À}iÊ industries exchange positions with high-level regulatory officials [452].
144
National Approaches debate has already caused. REACH appears to be enabling Sweden to transfer its technical and political problems to the EU. Thanks to REACH, KemI may escape from the limited future ability for Sweden to take stringent action caused by: (1) its joining of the EU, (2) recent decline in support for KemI’s approach from many stakeholders, its own regulators and the general public, (3) Sweden’s move towards carrying out socio-economic analysis of regulation. Any action, or lack thereof, at the EU level under REACH could be disguised as driven by the ‘EU’ rather than ‘Sweden’. Achieving a more integrated approach to chemical regulation, in which SEPA and SWEA have an equal share of responsibilities, may be hampered by the further concentration of decision-making at the EU level. KemI will be the sole representative at the EU Chemicals Agency. Unless the administrative structure of the European Chemicals Agency (or the European Commission) provides a platform for incorporating national regulators that represent occupational or wider environmental legislation, SEPA and SWEA may continue to lose power. France exhibits the most clear-cut case of REACH catalysing change in a national approach. REACH has forced the government to evaluate the future of its chemical industry. In turn, this has enabled the incorporation of a wider set of actors into chemicals policy debate47. Therefore, although France may be a ‘fence-sitter’ when it comes to EU environmental policy development (Section 1.6), it has the propensity to ‘leap off’ the fence and provide a structured response at the national level. In Germany, discussion on REACH has exposed the current inefficiencies in stakeholder negotiation and consultation. There is no evidence to suggest that REACH has sparked a wider debate on the sustainability of the industry. The German chemical industry will need to adopt a leading role in promoting such debate because, without its support, reforms are unlikely to succeed. Germany’s administrative 47
Actors may otherwise not have been included due to (i) their lack of scientific or technical expertise or (ii) the decentralised parts of its regulatory structure.
145
Framework for Chemical Risk Management under REACH division of science and policy may continue to give rise to confusion as to where to draw the line between the technical and political, which may not be resolved by a wider debate on chemicals policy. Several interviewees expressed the opinion that UK has been moving towards a more precautionary approach to regulation. This may help avoid delays in regulatory action caused by carrying out in-depth cost–benefit analyses. Such activities may be offset by a continued reliance on statistics in the UK regulatory culture following the appointment of a statistician to the Cabinet Office seeking to implement strategies similar to those proposed under the Hampton Review. The HSE is also engaged in promoting responsibility in the industry through statistic-based benchmarking [224] without evaluating the effects this may have on corporate or regulatory risk management activities outside the scope of the scheme. In other words, it does not appear to have evaluated alternative means of achieving the same objective. From the interview responses, British retailers appear more active in recent chemicals policy formulation than the other surveyed countries (see also [453]) reacting to REACH and NGO campaigns (e.g., [454]). It remains to be seen whether these relatively new actors in chemicals policy sustain their commitments and lobbying once REACH has been enacted.
4.7 Preparing for Implementing REACH All regulators and stakeholders interviewed agreed that some form of REACH is necessary. In particular, the current process of risk assessment needs to be reformed. With regards to risk management, regulators held a common view that REACH presents a general approach without clear objectives. As a result, a reoccurring opinion expressed by most interviewees is that REACH will be (i) burdensome to industry, (ii) require significant resources from Member States, and (iii) generally bureaucratic to all stakeholders involved. 146
National Approaches Although Sweden has been the primary catalyst for the adoption of the new EU chemicals policy, interview results indicate that the UK government exhibits the most developed policy positions and perspectives on the new REACH legislation. This appears to be a result of UK experiences with its Stakeholder Forum, highly detailed chemicals policies separately developed by several regulatory administrations and the RCEP, as well as the strong involvement by a wide range of stakeholders in many public chemical debates. Apparent similarities between Swedish regulation and REACH must be interpreted with caution. Responses from the interviews indicated that REACH is likely to be far more prescriptive than Swedish regulation. In this respect, a particular limitation of REACH may be the lack of flexibility that it provides to companies and regulators for the implementation of risk management measures at the shop floor. Many interviewees expressed concern that the implementation aspects of risk management, enforcement and compliance under REACH are being neglected. Interviewees often perceived the EU regulation as ‘top-down’. Subsequently, the danger is that resources will not be sufficient to cope with ‘bottom-up’ implementation, especially if REACH does not successfully address current deficiencies in compliance with existing chemicals legislation. Despite many improvements in environmental management, a recent review by the European Commission on the general overall quality of the European environment reports that further improvement is constrained by cases of regulatory non-compliance [455]. While new legislation such as REACH may be required, the focus needs to be on implementation [455]. For instance, it has been proposed that better co-ordination or harmonisation of the risk management priorities of the European Commission and the Member States should be achieved [456]. From a top-down perspective, implementing REACH does not appear to require major changes to administrative structures for Sweden or Germany. Sweden is the most efficient of the three countries for inputting information to EU decision-making, albeit limited to KemI. 147
Framework for Chemical Risk Management under REACH One interviewee explained that this is why KemI has not been recombined with SEPA and SWEA. Although Germany is considering forming a central chemicals agency, its existing structure appears fairly well prepared for REACH, especially as regulators foresee ministry and institute officials being able to attend relevant decisionmaking meetings together (exchanging lead roles according to the political or scientific nature of discussions). Through the operation of its IMC, France has the best structure for inputting multiple perspectives to REACH decision-making processes. It can be anticipated that France will have one such committee dedicated to co-ordinating decision-making in the European Chemicals Agency. French bureaucracy performs efficiently for ‘topdown’ regulation so, while France appears to be the best equipped for enforcement and high-level decision-making under REACH, existing decentralised structures may compromise its ability to collect data from the ‘bottom-up’. Another picture presents itself for the UK. Regardless of REACH, the UK should re-examine its existing historic structures, such as having consumer protection as a responsibility of the DTI instead of the Environment Agency or the Department for Health. The RCEP recently proposed that a central chemicals unit should be established in the UK [457]. A central chemicals unit should act as an instrument of authority for prioritisation of resources and centralised decisionmaking structure, rather than just co-ordinating the activities of other administrations (as proposed by the RCEP). Experience from Germany would suggest that having individual units within one agency operating under the supervision of another ministry could create a good working environment. The UK could also consider official inter-ministerial committees for increasing the transparency and accountability of decision-making. 148
National Approaches
4.8 Conclusions 4.8.1 Discussion and Conclusions National approaches to risk management are strongly established within France, Germany, Sweden, and the UK. The national approaches have evolved over many years involving complex socioeconomic and cultural phenomena. In turn, the regulatory culture is reflected in the country’s administrative infrastructures. Historical and social contexts, such as the legal systems and industrial relations, provide some explanation as to how and why differences occur in national approaches to chemicals policy and risk management. Each regulatory approach holds several strengths and weaknesses for achieving chemical safety. There is no single best option. French and German technical-based regulation often resembles the prescriptive elements of implementing Swedish policy, but lacks over-arching goals. In France, this appears to be a result of statistics guiding ‘topdown’ regulation while regulators carry out in-depth risk assessments. In Germany, this appears to be the result of separating political from scientific regulatory responsibility, whereas Sweden has a central regulator responsible for both. By comparison, a decentralised administrative structure enables the UK to operate a risk-based approach where objective goals supported by stakeholder dialogue translate to statistics and ‘bottom-up’ implementation. Overall, France and the UK appear to be the most apt for achieving changes in their regulatory approaches. Germany is deeply entrenched in its attempt to bridge the divide between science and policy. Sweden appears to be headed for an end to the ‘glory days’ of a powerful and ÌÀÕ«
>ÌÊ«ÕLVÞÃÕ««ÀÌi`ÊiÆÊÌ
iÊÞÊiÃÊ`Ã>««Ìi`Ê>ÞÊ be idealistic environmentalists who have been impressed by Sweden’s ‘tough’ regulatory approach. If policy serves to make organised activities stable and predictable, then the current legislative proposal for establishing the new EU 149
Framework for Chemical Risk Management under REACH chemicals policy appears to have failed. In particular, REACH does not have sufficiently specific objectives for risk management decisionmaking. Regulators are faced with an arduous task of responding to current weaknesses in their regulatory administrations and decisionmaking processes while preparing to implement a totally new system of EU chemical risk management. Careful application of control instruments combined with a strategy for dedicating regulatory and stakeholder resources can enable separate national objectives to be attained without negatively affecting EU decision-making processes. For instance, reductions in occupational exposure to a substance, reductions in industrial emissions or even limitations in the contents of a substance in consumer products can be achieved at a national level through various measures and initiatives independently from EU legislative measures aimed at the prohibition of that given substance. Compared with banning specific uses of a substance, the process of setting and meeting objectives through other control instruments – such as voluntary commitments, certification and training schemes, and workplace authorisation schemes – has a greater reliance on the involvement of industry and other stakeholder associations at a national level. In terms of stakeholder relations, a lack of attention to ‘bottomup’ REACH implementation may create further distance between stakeholders and regulators, which could be particularly problematic for Germany and Sweden. Several interviewees reported that increased decision-making at the EU level is already shifting the attention of some stakeholder associations away from national issues, catalysed through European Commission funding of EU associations. In other cases, stakeholders interviewed explained that individual national organisations are sometimes finding themselves having to adopt common positions. As a result, working in partnerships and achieving voluntary agreements between organisations at a national level could suffer. In particular, Sweden may find that its industry is no longer ‘so voluntary’ in establishing voluntary agreements for the phase-out of certain substances. Germany may experience further difficulties with its relations to national consumer and environmental NGO. The potential and 150
National Approaches implications of such events, especially with regard to trust in regulators at a national level, require further investigation. The functioning of the central European Chemicals Agency will have a profound impact on national regulatory administrative structures >`Ê ÃÌ>i
`iÀÊ Ài>ÌÃÆÊ vÕÌÕÀiÊ 1Ê Ài}Õ>ÌÀÞÊ `iVÃ>}Ê should consider how to accommodate and support the wide range of stakeholder activities, especially by promoting diversity rather than empowering centralised European networks or organisations. In this respect, the strengths and weaknesses of the regulatory administrations of France, Germany, Sweden and the UK may provide insight into how the Chemicals Agency should be structured and managed. The Chemicals Agency could provide a platform for coordinating, communicating and exchanging regulatory experiences between Member States.
4.8.2 Recommendations Member States could benefit from exchanging experiences with devising and implementing regulation. In particular, France, Germany and Sweden could complement each other in the development of criteria for the rapid enactment of technical EU restrictions. Such activity must be paralleled by the exploration of instruments at national levels that can be used in conjunction, in support or as exceptions to EU restrictions (e.g., licensing or certification schemes). In turn, the level of information needed for various decision-making contexts must be clearly delineated in a way to accommodate risk–benefit approaches present in other countries. For example, epidemiological evidence warranting immediate action at the EU level must be differentiated from instances when in-depth socio-economic analyses are necessary only at a national level. Because Germany’s approach to chemical risk management closely resembles that of France, the former should consider the creation of inter-institutional fora to facilitate political discussion. Although experience with a Chemical Stakeholder Forum in the UK has failed to 151
Framework for Chemical Risk Management under REACH meet the aspirations of NGO, it has provided a platform to examine supply-chain issues. This aspect of risk management appears to have been largely neglected in France, Germany and Sweden. In Germany, this would support a wider political debate on defining and achieving ‘sustainability’ for its chemical industry and provide a platform for co-ordinating existing programmes. Much like Germany, France needs to develop a national policy on chemicals. Unlike Germany, France is progressing to achieve this through its mandatory REACH Working Groups. Within a more immediate timeframe, France and Germany should seek to develop a wider set of instruments that can support current production levels while limiting the potential for trans-boundary pollution or exposures resulting from product use and disposal in other countries. For Sweden, a national stakeholder forum could improve the flow and exchange of information between regulatory administrations and stakeholders with KemI, which currently appears unidirectional. Sweden should also examine how to prevent KemI from compromising the authority or regulatory power of other Swedish regulatory administrations. The UK needs to evaluate the transparency and accountability of its final decision-making. Without creating a formal structure for weighing various inputs into decision-making inputs and without establishing criteria for different types of risk management action, it is difficult to see how transparency in governmental regulatory decision-making can improve. The flexibility of the UK’s risk–benefit approach offers many positive attributes that should be retained. A risk–benefit approach should be even further promoted at the EU level to balance (or avoid) potential over-prescription of technical rules. Simultaneously, the UK must recognise that in some cases technical and hazard-based approaches offer efficient and effective means of regulation. Because France lacks policy development and guiding principles in its approach to chemical risk management, it faces similar consequences as the UK in terms of its neglect to create a culture of safety in
152
National Approaches SME. In this area, France and the UK may have a lot to learn from Sweden’s use of policy objectives with respect to promoting a culture of chemical safety in industry and through supply chains. For France, adopting a more hazard-based approach to risk management could limit the occurrence of ‘victim associations’. Developing technical methods for substitution may support the effort to improve knowledge and understanding of chemicals safety in French and British SME. Maintaining a strong reliance on statistics for guiding regulatory activities may otherwise neglect fundamental risk management practices in industry or by professional users. Devising structures for communicating shop-floor experiences and regulatory implementation concerns to policy-makers in France and UK could also assist in promoting chemical safety awareness in businesses. Country-specific and overlapping recommendations are summarised in Table 4.6, organised according to preventive and protective measures. A framework for risk management under REACH must consider these existing strengths and weaknesses of chemical management at national and EU levels. The next Chapter exposes how without careful attention to particular issues during implementation the REACH legislative text is unlikely to deliver the recommendations identified in Table 4.6. Several examples are presented in Chapter 5 that illustrate how the strengths and weakness of the national approaches tend to hinder rather than facilitate EU decision-making. The proposed systems framework for decision-making under REACH then seeks to rectify these potential shortcomings of the legislation.
153
Framework for Chemical Risk Management under REACH
Prevention
Protection
Provide support for safety management in SME
France
Germany
Develop methods to implement substitution, precaution and pollution prevention through supply chains that supports industrial competitiveness
Develop methods to standardise and promote best-practice at the EU level
Engage with stakeholders
Develop rapid methods to set EU product standards
Develop methods to improve ‘bottom-up’ substitution Consider the use of taxes
Incorporate mutagens and reprotoxins into regulatory priorities
Improve the co-ordination of activities between regulatory administrations Increase transparency and accountability in decision-making Sweden
UK
Streamline socio-economic analysis by targeting specific sectors or limiting its application to the national level Develop benchmarking schemes for environmental standard-setting at the EU level
Incorporate mutagens and reprotoxins into statistical studies
Provide support for safety management in SME
Table 4.6 Recommendations for national regulatory administrations
154
5
A Systems Framework to Implement REACH
- Refinement Drinking whiskey out of a champagne glass (Ambrose Bierce, The Cynic’s Word Book, 1906)
5.1 The Need for Regulatory Reform Risk Assessment Risk Reduction Strategy 141 Total ? Draft Reports
Implemented
131 Draft Reports 72 Conclusions agreed & published
28 Conclusions agreed & published
?
Figure 5.1 Lack of progress under current EU chemicals legislation since 1993
The need for fundamental change to EU regulatory decision-making is illustrated by Figure 5.1. Since 1993, only 28 risk-reduction strategies (RRS) out of the 141 priority existing substances being assessed under Regulation 793/931 had been finalised and published
155
Framework for Chemical Risk Management under REACH by 20062 [47, 303]. Evaluating the ability of the current system to control chemical risks is particularly difficult because it is not clear how many of the completed RRSs have been implemented under the various national or EU legislative frameworks. While REACH will generate significant data on chemical hazards, interviewees expressed a sincere concern that slow and inefficient regulatory decision-making processes will persist. Since the publication of the EU chemicals policy in 2001, there has been little exploration of implementation scenarios for managing chemical risks under REACH. Proposals from Member States concentrate on mechanisms for sharing hazard data [271] and the role of the European Chemicals Agency during hazard assessment [459, 460]. Similarly, business impact assessments have examined generating and collecting hazard data, rather than processes for enacting efficient and effective regulation through exposure assessments or risk management measures [77, 276, 461]. Trial runs of registration and evaluation identified these potential Ã
ÀÌV}ÃÆÊÌ
iÊStrategic Partnership on REACH Testing (SPORT) and the Piloting REACH on Downstream Use and Communication in Europe (PRODUCE) stress the need for exposure data and the involvement of actors outside the chemicals industry in its generation ([315, 462], respectively). With the exception of a research article from Germany on the interaction of different legislative frameworks during the development of three recent EU RRSs [463], published research on regulatory risk management decision-making is lacking. Many of the obstacles currently encountered by regulators may continue under REACH, an opinion expressed by several interviewees. 1
Regulation 793/93 on the evaluation and control of the risks of existing substances [458] will be replaced by REACH.
2
Data based on information available at time of original writing of the thesis – i.e., before May 2007.
156
A Systems Framework to Implement REACH
5.1.1 Risk Assessment Risk-reduction stretgy Risk assessment Identification and characterisation of risks where hard assessment is the first step in the risk assement process
Development and selection of risk-reduction strategies to control or limit risks Decisions on controls through various legislative frameworks Stakeholder consultation/dialogue
Risk communication & implementation Implementation of risk management measures through various EU legislative frameworks, wider policy initiatives or voluntary actions
Figure 5.2 Existing structure of EU decision-making for chemical regulation (adapted from [564]) The focus on hazard assessment stems from the linear basis of EU chemical regulation (Figure 5.2). An obstacle encountered during hazard assessment delays the completion of a risk assessment (see Section 2.2.1). Because hazard classification depends on complex scientific evaluations of toxic potency and dose–response curves, regulatory measures can be postponed even if epidemiological, health surveillance, environmental monitoring or other risk characterisation data indicate that exposures to a given substance are causing significant risks to human health or the environment. Carrying out in-depth risk assessments also proves extremely complicated due to the vast number of exposure scenarios for >Ê Ã}iÊ ÃÕLÃÌ>ViÆÊ Ì
ÃÊ ÃÊ Ü
ÞÊ Ì
iÊ "À}>Ã>ÌÊ vÀÊ VV Co-operation and Development (OECD) chemical programme shifted from a risk to hazard approach to data collection in 2001 (see [166]). Although the range of exposure scenarios across the EU is more limited than across the world, the large contribution of hazard data to the OECD programme by EU countries may be responsible for the current EU focus on hazard assessment. Another possible explanation is that risk–benefit approaches to risk management dominate over hazard or technical approaches at the EU level (discussed in Section 6.3).
157
Framework for Chemical Risk Management under REACH Under REACH, hazard assessment requirements will continue to be placed on EU chemical manufacturers and importers based on ÛÕiÊÃiiÊ-iVÌÊÓ°Ó°£Ê>`ÊÓ°{®ÆÊ, Ê`iÃÊÌÊiÃÌ>LÃ
Ê>ÊVi>ÀÊ method to target chemical uses that can result in high exposures (e.g., consumer spray applications, grouting agents directly injected into environmental media, professional uses during manual applications that often result in skin abrasions). A particular limitation with the legislation is that it does not detail a mechanism to collate and consolidate existing exposure and risk characterisation data from stakeholder groups such as occupational insurers or poison centres. Regulators will be required to generate such data or otherwise base decisions on incomplete risk assessments. Apart from the creation of the authorisation process, which the interviewees typically anticipated to be limited to some 5–10 substances per year, REACH does not introduce new mechanisms for evaluating risks or enacting regulatory measures. In effect, although REACH responds to a need to fill data gaps on chemical hazards on a substance level, interviewees held a common opinion that it lacks a clear approach or framework for regulatory risk management.
5.1.2 Risk-Reduction Strategy (RRS) Risk assessment identifies a need for limiting risks Consult with stakeholders Identify risk-reduction measures Identify options for implementation Propose risk-reduction strategy Consult with stakeholders Refine Strategy
Figure 5.3 Existing EU ‘risk-reduction strategy’ process (adapted from [464-466])
158
A Systems Framework to Implement REACH After the completion of a risk assessment, EU decision-making takes place during the official EU ‘RRS’ process. Unless changes are made to the REACH legislation or new approaches are proposed in the technical guidance documents (TGD), then Member State regulators will continue to follow a process similar to the current Technical Guidance for the Development of Risk Reduction Strategies shown in Figure 5.3 [28]. The existing guidance document, which is lengthy and cumbersome to use, describes a very general procedure for decision-making that requires input from various legislative frameworks to evaluate chemical control options. Several of the regulators interviewed reported that attempts have been made to structure the EU RRS process and revise the TGD accordingly. These efforts have yielded little success. Hampered by the substance-bysubstance approach presenting limited opportunities for regulators to discuss structural issues, agreement on change to the process remains difficult, especially given the large number of Member State regulators involved. Although the interviews with regulators from France, Germany, Sweden and the UK indicated that progress on reaching agreements at the EU level has been made amongst regulators involved in the RRS process, there does not appear to be a coherent or transparent procedure for EU decision-making. Most countries have not produced official RRS reports3, so interviewees frequently reported that there was a general lack of experience on chemical risk management decision-making at the EU level. Germany and the UK are two of the six countries to have produced more than one EU RRS report4. Sweden has been proactive by publishing at least one draft RRS before 3
At time of writing, countries having completed the 28 risk-reduction strategy (RRS) reports were: Austria (1 report), Denmark (3), Finland (2), Germany (5), Ireland (1), the Netherlands (5), Spain (2), the UK (6). The total number of 141 reports is distributed as follows: Austria (5 reports), Belgium (3), Denmark (8), Finland (4), France (14), Germany (38), Greece (1), Ireland (3), Italy (3), Luxembourg (0), Netherlands (27), Norway (2), Portugal (0), Spain (4), Sweden (4), UK (25).
4
Taking a risk–benefit approach, the UK has finalised the most official reports, as well as commissioning the most chemical-specific studies on the advantages and disadvantages of market restrictions. By contrast, Germany’s technical approach is reflected in not having conducted official socio-economic analyses of its RRS.
159
Framework for Chemical Risk Management under REACH completion of a risk assessment5. France has not yet completed a RRS and was described by several regulators as experiencing particular difficulties in conducting the socio-economic aspects of alternative risk reduction measures. Surprisingly, there is no formal structure for developing or selecting RRS in the four countries surveyed. With the exception of companies producing the chemicals subject to regulatory review, stakeholder organisations were reported as providing little or no input to national or EU decision-making. Several interviewees expressed the opinion that existing practices do not permit the voices of other manufacturers, including downstream users, to be heard. With the exception of the regulators directly involved in producing RRS reports, interviewees had little understanding or knowledge of the national or EU process. Previous decisions were described by regulators as being neither systematically evaluated nor compared at national or EU level because responsibility for implementing decisions under various frameworks fall under various Member State regulatory authorities. The existing structure therefore leaves few opportunities for ‘learning from doing’. Evaluative factors: -Information needed to assess each risk level factor
Risk level factors: -Socio economic impacts -Risk assessment -Social values
Decisionmaking
Practical factors of alternative risk controls: -Implementation timeline -Efficacy, efficiency, equity -Enforceability
Political factors: - Tolerability of the regulatory response to alter risk level factors
Figure 5.4 The regulatory decision-making ‘wheel’ 5
Although Sweden is attempting to avoid the linear dimension of decision-making, the RRS report will not be finalised before the risk assessment report.
160
A Systems Framework to Implement REACH Many risk, evaluative, practical and political factors enter each stage of the RRS process (Figure 5.4). In addition to the socio-economic impacts of a regulatory action varying across the EU, the interviewees described that the level and detail of information necessary to justify a decision politically varies significantly between Member States, confirming the analysis of the different national approaches of France, Germany, Sweden and the UK in Chapter 4. Due to the inter-dependency of the factors shown in Figure 5.4, `iVÃ>}ÊV>Êi>ÃÞÊLiViÊ>Ê«ÀViÃÃÊvÊ`>Ì>ÊÀiiÛ>Õ>ÌÆÊ a change in information on a given chemical risk and risk reduction strategy can require each factor to be reassessed, creating the ‘wheel’ effect illustrated in Figure 5.4. Two examples of how this can occur are presented in Table 5.1. Although hypothetical, the two examples are based upon experiences described by regulators during the interviews. The decision-making ‘wheel’ results from the unstructured nature of the RRS process. Information is often unavailable or arrives only after a provisional agreement has been reached. Interviewees explained that this is because the Member States and Commission Directorates-General use a wide range of policy and legislative frameworks to input information into decision-making (see also [467]). In particular, interviewees complained that timelines for implementing specific risk management measures through existing legislative frameworks are often unavailable or not easily accessible (see also [463]), a concern echoed by the manufacturing companies surveyed by [468]. Decision-making is further complicated by test methods necessary for compliance and enforcement taking several years to standardise (see also [136]) and a lack of analytical substance reference materials necessary for enforcement (see also [136]). Finally, decision-making must also account for substantial differences in Member State national legislation even when implementing European Union (EU) directives (e.g., [208, 469]), as well as the levels of compliance (e.g., [470, 471]).
161
Framework for Chemical Risk Management under REACH
Political
Product standards are not politically acceptable to some Member States because previous consultation has addressed only the option of emission limit values. It may therefore be necessary to use a voluntary-based programme, which would affect the risk factors and require a re-evaluation of the potential socioeconomic impact.
Evidence of occupational illness has caused the need to reduce exposure levels to a given substance below those previously agreed.
Risk
Previously agreed emission limit values must be lowered but could be too costly for many SME to achieve. Therefore, a more cost-effective option could be to regulate the sale of the product. This would however require agreement on product standards.
Example 2Description
Due to the evidence of occupational illness, the substance may need to be reclassified, which could also cause a change to risk levels during the use of professional and consumer products.
Evaluative
Evaluative
The classification of the substance requires a reassessment of exposure `>Ì>ÆÊ«ÀiÛÕÃÊiÛ>Õ>ÌÃÊ relied on estimates worstcase scenario models where there were high margins of safety.
Factors
Because of the political scrutiny, there is a need to have a well-conducted risk assessment. In particular, the feasibility of reducing occupational exposure must include a technical assessment of respiratory equipment and workplace local
Practical
Risk
Data have just been made available that justify a need to change the hazard classification of a substance from toxic to very toxic.
Practical
Example 1 Description
Political
Factors
The re-evaluation process has identified that only occupational exposure to a certain size range of particulate matter needs to be strictly controlled. Further risk data must be collected and the socioeconomic analysis revised accordingly. Product standards should be set, but will require harmonised testing methods for enforcement across the EU.
Table 5.1 How further data can require re-evaluation of a risk-reduction strategy 162
A Systems Framework to Implement REACH Three recent cases of obstacles to EU risk management are presented in Box 5.1. In retrospect, complex decision-making processes could have been avoided by enacting immediate decisions through a more formal structure. The case histories in Box 5.1 illustrate how differences between national approaches become evident, prominent and may even be amplified during EU decision-making. The case of dichloromethane (DCM) highlights the contrast between German technical and British risk–benefit approaches. Arguably, the UK risk–benefit approach delayed decision-making but Germany’s insistence on technical product regulation is as much to blame. After almost two years of decision-making, the following compromise was eventually achieved: targeting specific uses for technical product regulation and allowing licensing at a national level for other uses. In the other two case examples, France and Sweden needed to respond to occupational protection due to unique national political pressure. Sweden went ahead with immediate action, whereas France waited for an outcome from EU decision-making. While interviewees described the delay as having fuelled political pressure in France from occupational victim associations, the immediate regulatory response in Sweden appears to have appeased trade union concern. Because neither case requires product standards or involves trans-boundary pollution to protect health or the environment, exposure reduction for these substances does not warrant a harmonised EU approach. EU chemicals policy reform should begin by reviewing the RRS process. First, exposure scenarios and the corresponding regulatory controls dictate the necessary level of harmonised or co-ordinated EU action. Second, many aspects of risk management, communication and implementation can operate in parallel to risk assessment processes, providing a basis for more efficient and effective (i.e., integrated) decision-making (Section 2.2.1). Third, regulatory administrators appear unable to contend with the limitations of the current process due to shortages of expert staff and resources. In terms of regulatory and research analyses, national and EU RRS procedures remain unexplored. Consequently, all interviewees anticipated that many of the current obstacles to EU decision-making will continue under REACH. 163
Framework for Chemical Risk Management under REACH Box 5.1 Three EU risk management anecdotes based on interviews 1. A political priority for 2-(2-butoxyethoxy)ethanol (DEGBE) and 2-(2-methoxyethoxy)ethanol (DEGME) In France, politicians faced pressure over mounting evidence of occupational risks from the potential reprotoxic properties of glycol ethers. The French government responded by making DEGBE and DEGME priorities in the EU risk assessment process, even though other Member State regulators were of the opinion that risks arising from some other highly dangerous substances warranted more immediate attention. Restricting the use of DEGBE and DEGME at a national level in France was thereby delayed until the outcome of the EU regulatory process. Several years later, official EU risk-reduction strategies (RRS) for DEGBE and DEGME were produced, but discussions of possible restrictions largely focussed on applications outside industrial settings relevant to France. France therefore decided to adopt much lower occupational exposure limits than those decided at the EU level, essentially banning most industrial uses of the two substances. 2. Preventing death and serious incidents from dichloromethane (DCM) To protect occupational and consumer health from DCM, Germany set technical standards on DCM sales. National measures would not have been practical for regulating the trade and marketing of DCM in the EU. Action to restrict the marketing of dangerous substances to protect human health or the environment must comply with Article 95 of the EC Treaty relating to the internal market. Objecting to immediate adoption of EU standards to ensure adequate risk reduction, the UK – a major producer of DCM – declared that it was necessary to conduct full risk assessments of alternative substances, as well as a detailed socio-economic analysis of
164
A Systems Framework to Implement REACH alternative RRSs, even though DCM had been substituted in some countries (e.g., Sweden) without evidence of increased health risk from alternative substances. After lengthy regulatory evaluations, a document was published detailing the need to select risk-reduction measures from various risk management options [472]. German regulators said that immediate action could have prevented the death of two workers in Germany resulting from DCM use. To facilitate the decision-making process, standards could have been established for container size and evaporation rate which would have to be used unless sold to certified or licensed users. 3. A national response to trichloroethylene (TCE) Before completion of an EU risk assessment on TCE, Sweden decided that epidemiological evidence on the hazardous properties of the chemical was sufficient to warrant strict regulatory control in the workplace (through the application of Article 138 of the EC Treaty). Sweden therefore began to implement a company use-specific authorisation system. Sweden’s action was later challenged by the European Commission as causing unnecessary barriers to trade, but a European Court of Justice ruling upheld the Swedish system. Several years later, EU risk assessment data confirmed TCE as a carcinogen, but a RRS has yet to be completed. TCE is a commonly used solvent that has several applications that may not easily be substituted, so several Member States have adopted different national regulatory controls for TCE [246]. Achieving future consensus on EU action to control this substance may therefore prove particularly difficult. Evidence suggests that, with the possible exception of setting (minimal) occupational exposure Limits, developing and implementing RRS to protect occupational health should remain at the national level for this substance.
165
Framework for Chemical Risk Management under REACH
5.2 REACH - A Critical Analysis Stakeholders Inputs Industry Registration Risk Management Risk Assessment Evaluation
Classification & Labelling Restrictions Authorisations Other Legislative Frameworks
Risk Communication &Implementaation Information Publication Classification & Labelling Restrictions Authorisations Other Legislative Frameworks
Figure 5.5 Regulatory decision-making under REACH (differences from the current regulatory structure are indicated in grey)
The current draft REACH legislative proposal follows the same linear structure as current regulatory decision-making: a hazard assessment must be completed before decision-making on risk management or communication. The differences from the previous regimen, indicated by grey shading in Figure 5.5, are: (a) a more extensive requirement for industry to generate chemical hazard and use data which it must then put into a central database managed by the newly created European Chemicals Agency, (b) an authorisation procedure for a small subset of ‘highly’ dangerous substances, (c) a structure for stakeholder input into restriction and authorisation procedures, and, (d) a mechanism for responding to stakeholder requests for chemicalspecific risk information, as well as a requirement to communicate
166
A Systems Framework to Implement REACH substances of very high concern on a Candidate List for Authorisation through the supply chain. With this linear approach to regulatory decision-making, a danger is that the input of hazard and use data on 30,000 substances will cause the REACH system to become overloaded. In particular, if new information continues to enter decision-making at various points, the ‘wheel effect’ described in Section 5.1.2 will continue (Figure 5.4). Requiring the completion of in-depth hazard assessments may create a bottleneck to further risk management, risk communication and implementation of risk reduction measures. Altogether, the mammoth creation of REACH could result in slow and unpredictable regulation. By focussing on filling the gaps in hazard data for chemicals on the EU market, the European Commission White Paper for a future EU Chemicals Strategy completely overlooked four fundamental questions, developed in Box 5.2. 1. Who defines chemical risks? 2. How should chemical risk information be communicated? 3. Which regulatory structures facilitate management and promotion of chemical safety in businesses? 4. How should EU decision-making be structured to facilitate Member State authorities reaching agreement on regulatory processes or outcomes?
167
Framework for Chemical Risk Management under REACH Box 5.2 Fundamental questions not addressed by the EU Chemicals Policy (based on interviewee responses) 1. Who defines chemical risks? More often than not, stakeholders disagree with each other on scientific risk assessments. While a company may conclude that a chemical does not present an unacceptable risk to human health or the environment, a regulator may identify the need to reduce a risk [473]. Consumer or environmental non-governmental organizations (NGO) may push for regulatory action beyond that which regulators deem adequate to reduce a risk. With risk assessment unlikely to be fully harmonised within the foreseeable future, the regulators interviewed expressed the opinion that difficulties in reaching consensus on scientific and technical evaluations of risk levels will continue to occur between Member States and industry (see also [38, 474]). Under REACH, regulators will retain the burden for evaluating risk assessments and risk management measures recommended or implemented by industry. A structured method for incorporating data from stakeholders outside the immediate chemical industry or companies using chemicals will be limited to web-mediated consultations during the restriction and authorisation procedures. Regulators will continue to hold responsibility for determining risk levels, especially for risks at the EU level resulting from multiple emissions and exposures to chemicals. Decision-making under REACH will therefore have to contend with no clear methodology for carrying out risk assessment under consumer legislation (see [475]). Otherwise, industrial associations or private enterprises specialising in compliance standards will have to continue setting levels of protection for consumer products when restrictions do not exist, even though this practice lacks democratic legitimacy and legal protection [476].
168
A Systems Framework to Implement REACH From an implementation and enforcement perspective (as opposed to legalistic interpretation), it is not always clear if restrictions take precedence over product-based legislation such as ‘CE marking’ [477]. Regulators have also identified gaps in existing chemical safety requirements across Member States and even differences in interpreting legislative requirements under provisions such as the Safety of Toys Directive [478, 479]. REACH must therefore clearly establish that restriction and authorisation take precedence over all other legislation.
2. How do you communicate risk information? A recent report by the European Policy Centre identified a need for improving regulatory risk communication [143]. While recommendations by the European Policy Centre involve wider aspects of risk communication across EU legislative frameworks, REACH must contend with consumer and environmental NGO demands for increased communication of chemicals present in consumer products. A systematic procedure for listing the chemical contents of products according to categories of product use needs to be established before implementing REACH. This can take the form of technical guidance for the operation of the newly created Chemicals Agency. In terms of the practicality of implementing REACH, no chemical safety report (CSR) appears to exist at the time of regulatory enactment of REACH. The concept of a CSR has therefore never been tested. The extent to which a substance-specific CSR can cover more than one specific use or be incorporated into safety data sheets (SDS) for preparations (i.e., chemical mixtures) needs development.
3. Which regulatory structures promote chemical safety in businesses? A common image held by most interviewees is that of REACH being a complex piece of ‘top-down’ legislation (Chapter 5).
169
Framework for Chemical Risk Management under REACH The current legislative proposal does not appear to meet general downstream user demands for a simple and predictable set of regulations (e.g., [192]). The Commission’s statement that REACH will avoid overlaps with other legislation can be misleading [48]: most of the 40 directives and regulations that REACH replaces are amendments to Directive 76/769 restricting the marketing and use of dangerous substances. Unless changes are made to decision-making structures, many regulators and stakeholders interviewed expressed the view that future legislation and policy may continue to prove difficult for companies to track and predict (see also [480]). Because the authorisation process of REACH will apply only to a small subset of substances presenting ‘very high’ risks to human health or the environment6, achieving risk reduction will continue to rely on implementing measures through a wide number of legislative frameworks or policy instruments, with marketing and use restrictions at the EU level anticipated only as a final option [164]. In addition to clarifying the inter-relation of restrictions with product-based legislation, improving consistency across legislative frameworks should focus on harmonising procedures for demonstrating compliance. Several companies interviewed reported that difficulty in achieving regulatory compliance does not necessarily stem from different standards across legislative texts or across different implementation of EU directives, but from the internal (corporate) administrative structures necessary for demonstrating compliance. CSR should be constructed so as to make it easier for a company to demonstrate compliance with other legislation such as the Chemical Agents Directive, the Dangerous Discharge Directive [265] or Integrated Pollution Prevention and Control (IPPC). Such harmonisation could also improve the standardisation of environmental reporting for regulatory administrations (following [481]). For the chemical industry, this 6
Ê
iV>ÃÊÀiµÕÀ}Ê>ÕÌ
ÀÃ>ÌÊÜÊLiÊ`iÌwi`Ê`ÕÀ}ÊÌ
iÊ«iiÌ>ÌÊ«ÀViÃÃÆÊ the number of actual chemicals and uses subject to the process will depend on the availability of resources within the European Chemicals Agency [279].
170
A Systems Framework to Implement REACH could also promote innovation by reducing institutional barriers necessary for demonstrating regulatory compliance [482]. Regardless of the extent of integration or harmonisation of REACH with other legislative frameworks, companies will face complex supply-chain management issues under REACH [483]. Implemented over q11 years, detailed information on chemical risks for most chemicals will not be available until after registration, as will decisions on future restriction and authorisation. Future guidance should consider how companies may establish acceptable chemical uses, otherwise companies may opt for chemicals that have already gone through the system, regardless of risk levels. Finally, regulators must consider the enforceability of REACH. No evidence was obtained from the research interviews to suggest that current levels of non-compliance with chemical legislation ÜÊ«ÀÛiÊÜÌ
Ê, ÊvÜ}ÊQ£ÓRÆÊ-«iÌâ]Ê£®°ÊÛiÊ the technical difficulties in monitoring and enforcing chemical products, considerations of enforceability must be incorporated into the demands for registration. REACH may otherwise fail to create a level playing field between EU and non-EU products.
4. How do you structure regulatory decisions? Apart from registration, authorisation and CSR, the new REACH legislation does not make significant changes to the current process of EU risk management. The use of wider tools such as voluntary agreements, producer take-back of (un)used products or sector-specific guidance for achieving risk reduction have not been integrated into the current legislative text. Determining the necessity to take EU action will be left to Member States, the Chemicals Agency or the European Commission during procedures for restriction and authorisation. All interviewees agreed that unless radical changes are enacted in the legislation or proposed in future technical guidance documents (TGD), many existing risk management dilemmas will persist.
171
Framework for Chemical Risk Management under REACH Overall, resolving the potential shortcomings of the REACH legislation will depend on the extent to which future TGD can address the issues discussed above. In this respect, it is worthwhile considering alternative chemical regulatory systems that have been proposed by the Dutch Ministry of Housing, Spatial Planning and the Environment (VROM) [316] and the UK Royal Commission on Environmental Pollution (RCEP) [165]. VROM and the RCEP recognise the need for clear and predictable decision-making processes based on the categorisation of chemicals and chemical uses. A set of ‘decision-making rules’ forms the central aspect of the Dutch Strategy on Management of Substances (SOMS) (Figure 5.6a) which is organised according to a hazard and use matrix. In comparison, the RCEP’s proposed approach concentrates on the use of monitoring data for decision-making through a structured approach to using stakeholder inputs and performing reviews of previous decisions (Figure 5.6b). The RCEP also proposes a mechanism for identifying many chemicals of regulatory concern through a screening process before full risk assessment evaluation, the results of which would be published to provide risk information to stakeholders and businesses at an early stage of regulatory decision-making (Chemical Screening and Listing: Figure 5.6b). This approach differs to the incremental identification of a limited number of ‘substances of very high concern’ under REACH. The Dutch SOMS and the UK RCEP systems generate risk information on a prioritised ‘need-to-know’ basis, avoiding the strong reliance on hazard assessment and volume triggers that REACH proposes. Under SOMS, minimal hazard data requirements are established >VVÀ`}ÊÌÊV
iV>ÊÕÃiÃÆÊvÊVÀÌiÀ>Ê>ÀiÊÌÊiÌ]ÊÌ
iÊVÀÀië`}Ê uses are banned. For the RCEP, risk assessment and management would take account of all relevant monitoring data. The RCEP would therefore also avoid the linear dimension of decision-making under REACH: information on exposure and risk characterisation would enter the decision-making process independently to data generated or supplied from industry. This information could, in turn, trigger risk management monitoring and review which feed back into the risk
172
Industry Data
Other Legislative Frameworks
Classification & Labelling
Risk Communication & Implementaation
Industry Data Stakeholder Inputs
Other Legislative Frameworks
Phase-Out
Risk Communication & Implementaation Classification & Labelling
Risk Monitoring and Review
Other Legislative Frameworks
Phase-Out
Classification & Labelling
Risk Assessment & Management
Chemical Screening & Listing (Risk Communication)
(b)
Figure 5.6 Alternative chemical regulatory systems to REACH: (a) Dutch Strategy on Management of SubstancesÊ-"-Ê>`>«Ìi`ÊvÀÊQΣÈRÆÊL®Ê1Ê,
*ÊChemicals Strategy (adapted from [166])
Other Legislative Frameworks
Classification & Labelling
Risk Management
Risk Assessment & Decision-Making Rules (Risk Management, Communication & Implementation)
(a)
A Systems Framework to Implement REACH
assessment and management processes. The RCEP screening process would also rely on existing hazard assessment data to make full use of in silico (computational chemistry) methods for identifying and prioritising substances for regulatory review.
173
Framework for Chemical Risk Management under REACH The question of delineating responsibility for defining and communicating risk levels begins to be addressed in the SOMS and RCEP systems. By establishing a set of decision-making rules, the Dutch SOMS can be seen as promoting the communication of risk and safety: a company would know the primary criteria under which a chemical may be used or be subject to further regulatory controls. The RCEP method of chemical listing sets out a similar approach to risk communication by making data from monitoring studies and hazard evaluations publicly available. Chemicals with hazardous properties may be subject to further regulatory review or may even be phased-out if detected at elevated concentrations in humans, marine mammals or top predators. In terms of regulatory risk management, the SOMS does not detail rules for EU action corresponding to each hazard and use categorisation. Similarly, while the RCEP presents a wide and detailed set of recommendations to make increased use of a wide range of risk management and policy instruments, it does not present these as a formalised structure for EU decision-making.
REACH
SOMS
RCEP
Prioritisation of decisions
×
Yes
Yes
Decision-making rules
×
Yes
×
Chemical listing
×
×
Yes
Structured stakeholder inputs
×
×
Yes
Monitoring and review
×
×
Yes
Table 5.2 Comparison of REACH, SOMS and RCEP
174
A Systems Framework to Implement REACH Although many aspects of the SOMS or RCEP system can be readily integrated with REACH, neither appears to have been particularly influential in the European Commission legislative drafting (Table 5.2). The systems framework, which is presented in the next section of this Chapter, contains the key elements of the SOMS and RCEP proposals shown in Table 5.2.
5.3 Systems Framework This section presents the framework of the system approach proposed in this dissertation. Compared with the current REACH legislative proposal, the framework contains elements of the Dutch SOMS and the UK RCEP chemical strategies (Section 5.2; Table 5.1). Specifically, the framework would incorporate: (i) a set of decision-making rules that includes chemical listing, (ii) a method for extending stakeholder inputs into all parts of decision-making, and (iii) a mechanism for monitoring and reviewing regulatory outcomes. It also proposes to develop a method of target-setting to facilitate the co-ordination of regulatory activities under other legislative frameworks with REACH. Figure 5.7 highlights the structural differences of the systems framework compared with the current REACH legislative proposal (refer to Figure 5.5). First, before registration and evaluation, information on chemicals used in certain consumer products would be made immediately accessible to stakeholders and the general public (listed uses). At this stage, decision-making rules would enable companies to identify safe and permissible uses that would then be subject to less rigorous regulatory review. Immediately following REACH enactment, companies across the supply chain would be required to follow a basic set of regulatory recommendations for controlling chemical risks, regardless of registration timelines. All these processes could continue to operate in parallel to the implementation of the REACH regulation, serving as a guide for companies and regulators.
175
Industry Registration
Industry Data
Decision-Making Rules (Risk Management, Communication & Implementation)
Evaluation
Risk Management Classification & Labelling Restrictions Authorisations
Stakeholder Inputs
Risk Communication & Implementaation Classification & Labelling Restrictions Authorisations Targets
Targets
Risk Monitoring and Review Figure 5.7 The proposed systems framework (structural differences between the framework and REACH are highlighted in grey)
Framework for Chemical Risk Management under REACH
176
Risk Assessment
A Systems Framework to Implement REACH During evaluations, inputs from stakeholders outside the chemical industry would be reviewed together with the data submitted in company registration dossiers. A wider set of monitoring and epidemiological data would thereby be systematically incorporated into the risk assessment process, as well as information submitted by stakeholders on the efficiency and effectiveness of various risk management measures. Following the evaluation procedure, EU or Member State legislative frameworks outside the immediate scope of REACH would only be used to achieve or supplement decisions under REACH, often co-ordinated by a target-setting mechanism. REACH would not depend on inputs from other EU legislative frameworks for risk management decision-making. Finally, a formalised mechanism would be established for carrying out retrospective analyses of regulatory outcomes and revising the decision-making rules. Although the framework can take the form of an official TGD for industry and regulators to use when implementing REACH, any structure for decision-making will be pertinent to the functioning of the European Chemicals Agency and the European Commission. The corresponding details of the framework are therefore presented in the following three sections: decision-making rules, technical guidance and administrative structures.
177
Framework for Chemical Risk Management under REACH
5.3.1 Decision-Making Rules Listed uses Criteria for listing of potential chemical components of certain consumer products before registration
Safe uses
Permissible Authorised Restricted uses uses uses
Tolerable uses
Exceptions Low priorto evaluaity for tion evaluation Excluded from restriction and authorisation
All uses Some uses Overall banned banned, risk unless all other reduction authorised uses althrough Possible through lowed EU (timeEU or as under targetlimited) national REACH setting exceptions processes where to restricDoes not selection tion and Member apply to of approauthorisaState occupa- priate risk tion permitting tional ex- manageschemes posures in ment factored industrial measures into settings only apREACH ply at the decision- Covers all national productmaking level based legislation (e.g., toys)
Figure 5.8 Categorisation under the systems framework
Decision-making rules would be used to organise chemical uses according to listed, safe, permissible, authorised, restricted and tolerable (Figure 5.8). Substances contained in preparations or in articles with high consumer exposure potentials would be listed7 on the REACH-Information Technology (REACH-IT) website, 7
As with the current REACH legislative proposal, a company would be able to request a derogation from the Chemicals Agency on grounds of confidentiality.
178
A Systems Framework to Implement REACH which will be maintained by the European Chemicals Agency. The availability of such information should facilitate the registration process and also facilitate several actors prioritising their risk management activities, as discussed below. Safe and permissible uses would establish general exceptions from prioritised regulatory decision-making procedures anticipated under REACH. While safe and permissible uses would streamline regulatory processes for EU decision-makers, the identification of these uses should help provide companies with predictable regulatory outcomes. In particular, a safe or permissible use would not face a ban under restriction or authorisation. Based on minimal sets of hazard and exposure data, companies could identify safe uses for which the substance may be marketed for manufacturing and professional applications. Information for making choices on alternative products will thereby be available to downstream users before registration. Permissible uses would be limited to manufacturing processes that may require the use of dangerous substances for safety or performance functions8. All substances intended for use in professional and consumer products would also need to follow a set of regulatory recommendations. By adhering to the recommendations, a downstream user would not need to communicate the corresponding uses upstream during registration. The recommendations would later serve as a template for banning chemical uses under restriction or authorisation. For predictable EU decision-making following evaluation, chemical uses requiring regulatory risk reduction would be categorised as restricted, authorised9 or tolerable. The categorisation process for the proposed systems framework is shown in Figure 5.9. A substance or group of similar substances (referred to as a ‘chemical grouping’) would enter the systems 8
Companies would need to apply the substitution principle under the Chemical Agents Directive when selecting substances identified as meeting the criteria safe or permissible uses.
9
Authorised uses would be subject to authorisation under REACH or other Member State permitting schemes.
179
Framework for Chemical Risk Management under REACH
Authorised Uses Regulatory Risk Reduction Decision-making
Restricted Uses Tolerable Uses
Professional and Consumer Recommendations
Manufacturing Permissible uses
Guiding Implementation & Streamlining Regulatory Decision-making Professional and Manufacturing Safe Uses
Consumer Listed Uses Substance or Chemical grouping
Figure 5.9 Flow process for the systems framework
vÀ>iÜÀÆÊ >ÞÊ V
iV>Ê }ÀÕ«}Ê ÜÕ`Ê LiÊ L>Ãi`Ê Ê ÌÀÃVÊ profiles10 of a substances identified through Reach Implementation Project (RIP) 3.311 on Information Requirements. The systems
10
Intrinsic properties of a chemical consist of physico-chemical properties (e.g., solubility, vapour pressure), fate properties (e.g., (bio)degradation, partition coefficients), ecotoxicoligical properties (e.g., toxicity to aquatic or terrestrial organisms) and toxicological properties (e.g., acute toxicity, irritation, genetic toxicity) – see Section 2.1.
11
In addition to the use of (quantitative) structure–activity relationships and ‘read-across’ (see Section 2.1.1), ‘category approaches’ for grouping substances according to structural activities are being developed under RIP 3.3.
180
A Systems Framework to Implement REACH framework would focus on chemical uses to avoid an over-reliance on a substance-by-substance approach. Risk management activities based on use can also then help identify, assess and regulate synergistic effects that exposure to multiple substances can have on health and the environment (Section 2.2.1). As a first step, relevant consumer uses would need to be listed (listed uses). Safe and permissible uses would then be removed from subsequent EU regulatory risk reduction processes12. While companies must follow recommendations, regulatory risk reduction may deem it necessary to supplement these with more prescriptive duties. Listed Uses Examples of the types of substances contained in consumer products subject to listed uses would include: aerosols intended for indoor use, personal deodorants, volatile substances contained in household adhesives, plasticisers with high mobility incorporated into the outer matrix of toys, and colouring agents loosely bound on the surface matrix of clothing textiles intended for direct prolonged contact with skin. The product categories could include the criteria necessary for companies and retailers to identify and follow the regulatory recommendations. Regulators and stakeholders (such as retailers, academic researchers and material science institutes) would collect this data before registration through various voluntary and government/EU programmes. The data sets could then be updated and reviewed during registration. Obligations to communicate the presence of substances of very high concern on the Candidate List for Authorisation in products, as required by the REACH Regulation, would operate separately. While the communication obligations for Candidate List substances covers a wider set of product groups than listed uses, these REACH requirements are limited to substances where regulators agree on hazard assessment results and the need for authorisation. Therefore, these communication requirements under REACH may overlook 12
Unless criteria for establishing safe and permissible uses need to be reviewed.
181
Framework for Chemical Risk Management under REACH many substances where hazard is not currently evident or has not been subject to regulatory scrutiny. The extent to which evaluations under REACH would need to cover consumer products could be limited to the listed uses unless registration data indicate a particular concern (e.g., endocrine disruptors, very persistent and very bioaccumulative (VPVB) substances, or genotoxic carcinogens). Listed uses would also facilitate the prioritisation of enforcement activities aimed at consumer goods. In some cases, regulatory requirements for product labelling (e.g., substances meeting certain hazardous profiles) for the listed uses could be required under product specific legislation, or as part of REACH restriction and authorisation. Safe or Permissible Uses
Safe uses Intrinsic properties - no evidence of non-threshold toxicological effects and - low volatility and low mobility and - no evidence that the substance is easily absorbed via the skin and - no evidence that the substance causes cracking of the skin Exposure scenarios (for substances with the intrinsic properties listed above) 1) industrial brush applications requiring gloves or spray applications requiring gloves and respiratory protective equipment 2) industrial systems not discharged or released directly into the environment – e.g., filtered, recycled or effluent treated as waste water 3) incorporation into articles not intended to be in direct contact with humans during use
Table 5.3 Examples of possible safe uses
182
A Systems Framework to Implement REACH Establishing a process for identifying safe and permissible uses before, or in parallel with, Registration is expected to reduce uncertainty in chemical supply chains resulting from REACH. Downstream users may otherwise have to wait for registration phase-in periods before making choices of alternative products (i.e., identifying and selecting safe uses). The systems framework would therefore require Member States and the European Commission to define minimum in vitro and in silico toxicological test data, following the registration requirements13 for low exposure manufacturing applications that would correspond to safe uses. The safe uses would be limited to substances where there is no evidence of non-threshold toxicological effects (see Section 2.1.1) and the physico-chemical properties of the substance enables exposures to be controlled (Table 5.3). The safe uses would still be subject to registration or, if any relevant new risk data become available, regulatory review. A sample set of safe use criteria is provided in Table 5.3 that avoids chemical use scenarios with the highest exposure potentials following current risk assessment TGD [114]. A further set of criteria would be devised to identify permissible uses for chemicals that provide certain safety or utility functions – e.g., colours in warning signs, materials in automobile airbags, materials in parachutes and professional sports equipment, and processing chemicals for pharmaceuticals used in closed-systems. The chemical use would need to fulfil certain requirements that demonstrate that substitution of the substance may present other risks by compromising the safety or performance of a final manufactured product or service provision. Following TGDs developed in the RIPs or issued by the agency, a company could identify if certain uses meet these demands before registration and even exchange existing data during the pre-registration period (e.g., for substances with later deadlines).
13
Companies could supplement these data with additional information that does not involve animal testing. Companies would also need to take account of whether existing data or other information indicates a particular hazardous property.
183
Framework for Chemical Risk Management under REACH A simplified set of registration data requirements would be demanded for safe or permissible uses14 using a basic ‘tick box approach’. Safe uses would not be subject to further risk management under REACH unless evidence indicated that registered data are incorrect. Permissible uses would not be prioritised for evaluations or other decision-making, and would be issued with time-limited exceptions during restriction and authorisation. Recommendations Based on previous restrictions on the marketing and use of chemicals under Directive 76/769, a set of recommendations could be detailed in TGDs or as an official European Commission recommendation. The regulatory recommendations would immediately apply to all professional and consumer products on the EU market 15. Responsibility for following the recommendations would apply directly to individual companies throughout the supply chain, which therefore holds implications for product, supplier and user liability. A company would need to possess a minimal amount of hazard and exposure data to meet a given recommendation. If evidence indicates that a hazard exists, including data from academic research articles, industry would be responsible for establishing that the relevant professional or consumer chemical exposures are below derived noeffect level (DNEL) thresholds or otherwise minimised or avoided (Section 2.1.2). Companies would be responsible for establishing and documenting DNEL based on generic guidance available from regulators that is being developed in the relevant RIP16. Companies would work within DNEL boundaries by noting similarities and 14
In such an instance, the company that has previously identified a safe use would be liable for risks resulting from incorrect information that it had knowingly used to identify the safe use.
15
Although professional uses include chemical exposures resulting from occupational use, the systems framework proposes to make a clear distinction between ‘professional’ and ‘industrial’ applications, as explained in subsequent sections.
16
These TGD focus on scientific literature reviews of assessment factors and how to develop exposure scenarios but do not consider the extent that an exposure may fluctuate above or below a DNEL or in what situations regulators, or other stakeholders, may disagree on a DNEL.
184
A Systems Framework to Implement REACH differences with previous examples, for instance by using metabolism studies in conjunction with biomonitoring studies. Consistent with the aim of reducing the need for animal testing, hazard data necessary for meeting the recommendations would be limited to in vitro and in silico testing. Exposure would be the key defining criterion. The inclusion of a wide variety of potential sources of hazard data should minimise the potential for introducing false-negative bias during in vitro or in silico testing. To further safeguard against type-II errors, regulators should consider adopting a method to use epidemiological data from existing occupational exposures to establish ‘bounded effect levels’ [484]. This approach sets regulatory thresholds based on confidence limits against the probability of an observed effect (e.g., 95% probability of no observed effect at a given exposure). Essentially, the data sets for meeting the recommendations follow the general requirements of hazard data necessary for registration but are defined according to specific exposure scenarios. A company would need to notify the European Chemicals Agency if it believed that animal testing was necessary to establish that an exposure is sufficiently below a DNEL (or that a DNEL needs to be reviewed, revised and lowered). As under the current legislative REACH proposal, a company generating any data prior to a Registration deadline could obtain mandatory reimbursement from other companies needing the data at the time of Registration. Because chemical use for a given application may result in different exposure scenarios, and because the corresponding risk management options can vary significantly, a company may adopt alternative risk management measures to a given recommendation. In such cases, the company using the chemical would be required to hold evidence to justify the measures as recommendation equivalents. For instance, the supplier of an adhesive for flooring could supply an information leaflet detailing work rotations to control professional exposures instead of reformulating a product. The recommendation equivalent could therefore take the form of sector or product specific best
185
Framework for Chemical Risk Management under REACH practice guidance available from a trade association, trade union, regulatory authority or other source. An indicative set of recommendations is shown in Table 5.4 based on recent EU restrictions, technical guidance and official regulatory reviews. The recommendations would be described according to hazard criteria and potential exposure levels without including the use of personal protective equipment (PPE). Exposure assessments should assume that basic safety instructions are being followed based on typical scenarios of airflow rates, and size of room [201]. A company would therefore be able to modify exposure calculations by decreasing the concentration of toxic substance or reducing the volatility of a mixture. Unless labelled as ‘professional use only’ a product would have to comply with recommendations intended for consumer uses17. Existing restrictions would continue to operate, such as the ban of CMR category 1 and 2 substances in consumer substances and preparations. The recommendations would essentially extend such bans to include substances that are potentially carcinogenic, mutagenic or reprotoxic (CMR) (e.g., category 3) or endocrine disruptors but lack sufficient data for a full hazard assessment. Similarly, any evidence of non-threshold toxicological effects (e.g., based on results of the AMES test described in Section 2.1.2) would warrant a company needing to follow certain recommendations. When a DNEL cannot be established by a company, a dangerous substance should not be used at any concentration for a specified application under the recommendations. This zero-threshold rule would therefore apply if data indicate: (a) a substance is a suspected CMR or endocrine disrupter with a ÛiÀÞÊÜÊ Æ L®Ê >ÊÌ
ÀiÃ
`ÊÃÕLÃÌ>ViÊÀÊ>Ê* /ÊÀÊ6*6 ÊÃÕLÃÌ>ViÆ (c) a substance that may be toxic, a CMR or an endocrine disrupter but there are insufficient data to establish a DNEL. 17
Unless a product is clearly marked, it is sometimes ‘impossible’ for regulatory inspectors to distinguish between professional and consumer use [477].
186
A Systems Framework to Implement REACH The safety measures necessary for following the recommendations or recommendation equivalents could be detailed or referenced in SDS. The process would need to be presented to companies in such a way to ensure that threshold calculations for preparations follow Directive 1999/45 on the classification and labelling of preparations [310]. In other words, additive rules for combined toxicity must be followed unless evidence suggests that synergistic or antagonistic mechanisms are at work. Regulators (Member State, Chemicals Agency or the European Commission) would be responsible for checking if restrictions must support or further define the recommendations or recommendation equivalents. Recommendations would therefore act as ‘templates’ for future regulation. Because they would apply before registration deadlines, recommendations and recommendation equivalents would provide a general format for companies submitting substance registration dossiers for professional and consumer uses. For workability, the recommendations would be limited to products intended for consumer or professional use, organised into a simpleto-follow matrix of hazard and use criteria. Only in cases when regulators require information during evaluation or enforcement would detailed exposure calculations need to be submitted by a company. The recommendations would serve as a regulatory reference source for formulators or manufacturers. Therefore, a downstream user would only need to communicate a chemical use upstream for registration purposes when it believed the chemical application in a manufacturing process fell outside the exposure scenarios for which a recommendation or recommendation equivalent applies. Recommendations would provide a method for checking products marketed in the EU – whether produced in the EU or imported into the EU – including articles. This reinforces the principle (and World Trade Organisation (WTO) requirements) that imports should face the same regulatory requirements as products produced in Europe. A company placing a product on the EU market for
187
Toxic, potential CMR or potential endocrine disruptor
Hazard SubCriteria
Exposure (without PPE)
Exposure > DNEL or zerothreshold N/A
Highly volatile or easily absorbed via the skin
Previous Example
Recommendations for professional and/or consumer uses
Toulene [302]
Not allowed in adhesives for consumer use
ÆÊ
Not allowed in spray applications for consumer or professional use
â À>ÌÃÆÊ*ÆÊ **ÆÊ ViÆÊ* ÆÊÃ>«iÃÆÊLiâiiÊQÎäÓRÆÊ *]Ê *]Ê *]Ê *]Ê *]Ê "*ÆÊvÜÃÊ requirements under the Toys Safety Directive
Not allowed in outer matrix of articles intended to come into direct and prolonged contact with the skin or the mouth Not allowed in outer matrix of articles intended to be in direct contact with children
Exposure > DNEL if accidentally ingested or inhaled or zerothreshold
6>ÌiÊiÃÌiÀÃÊvÊ ÆÊ>nium sulfides [302]
Not allowed as a liquid or gaseous substance or preparation in ornamental devices or in toys
Exposure > DNEL or zerothreshold
ÊQ{ÇÓRÆÊvÜÃÊVÕÀÀiÌÊ RRS, TGD
Supply of correct PPE when selling product at volumes above 500 ml unless selling to a certified or licensed professional user
Framework for Chemical Risk Management under REACH
188
Hazard Criteria
Any exposure
Potential nonthreshold toxicity
N/A
Any exposure
VÌ> ÊQÎäÈRÆÊvÜÃÊVÕÀrent EU risk reduction TGD guidelines [485]
Extends current ban on CMR substances (i.e., categories 1 and 2) in consumer products [302] to include substances with potentially low DNEL or non-threshold effects (e.g., «iÌ>Ê>`ÊVÌ> ÊQ{nÈR®ÆÊ follows restrictions on diffusive applications of chloroform and tetracholoroethane [302]
Supply of correct PPE or provide training when selling product unless selling to a certified or licensed professional user The provision does not include sale of the product in pellet form
Not allowed as consumer substances or preparations
189
Table 5.4 Recommendations for professional and consumer uses according to hazard criteria and exposure levels (without personal protective equipment (PPE)) (chemical acronyms listed in Abbreviations)
A Systems Framework to Implement REACH
Any exposure
Follows current EU risk reduc- Supplier must organise take-back of tion TGD that specifies signifi- unused or remainingconsumer prodcant release to the environment ucts sold in quantities greater than may occur during consumer use 5 litres or 5 kg containers (Example only - size of contaciner reviewed) [202]
Framework for Chemical Risk Management under REACH professional or consumer use can be required to sign conformity checks stating that the product meets the relevant recommendations or a recommendation equivalent. Because the recommendations can be immediately applied, the process avoids potential delays in the application of risk management controls resulting from the 11-year implementation timeline for REACH. Restricted or Authorised Uses For decision-making on EU restriction and authorisation, a method for communicating certain national permitting or licensing schemes appears to be necessary18. Otherwise, variations in existing national legislation could delay reaching consensus on the need for EU action and even act against the subsidiarity principle. Authorisation, approval or certification schemes operated at the Member State level must already be notified to, and approved by, the European Commission under Directive 98/34 on technical standards. Coordinating this process with decision-making under REACH is therefore not expected to represent an additional administrative burden to Member States or the European Commission. REACH must also clearly establish that controlling occupational exposures follows Member State legislation except for restrictions on professional uses. Although measures aimed at occupational health usually follow Article 138 of the EC Treaty, and not Article 95 on which the REACH regulation is based, many professional users do not work in an industry environment where existing occupational health and safety legislation may be monitored and enforced [192]. Furthermore, professional users are likely to lack the knowledge and resource necessary for implementing chemical safety measures [192]. Control of chemical risks resulting from professional uses may therefore depend on restrictions on marketing and use, which must be controlled at a EU level to ensure efficient monitoring and promote the functioning of the single market. It is therefore evident 18
This would apply to product restrictions or licensing schemes that do not significantly affect the functioning of the single market, for example, processes and products authorised for water treatment processes (e.g., [487]).
190
A Systems Framework to Implement REACH that REACH must establish a clear definition of ‘professional’ versus ‘industrial’ use. The research further concludes that chemical uses should be subject to a permitting mechanism at Member State rather than EU level when a political response to control occupational exposure in industrial settings is necessary at a national level. This follows the evidence from the interviews that indicate that serious political pressure at national levels may arise if certain chemicals are not included in an authorisation process (Section 5.3 - Box 5.1). Making these uses subject to Member State permitting-based schemes therefore enables national authorities to set the appropriate level of occupational health protection. While implementing measures relating to Article 138 of the EC Treaty is outside the immediate scope of REACH, it will ultimately affect decision-making. To promote chemical safety, the framework proposes that companyspecific exceptions to following the use-specific REACH Authorisation process be issued by Member States within their national territories. Otherwise the authorisation process could be particularly burdensome for companies involved in activities such as accident response, environmental remediation or chemical management services for handling dangerous substances. Company-specific exceptions would only follow the submission of dossiers during the REACH Authorisation process in which companies demonstrate an in-depth knowledge of chemical safety, personnel monitoring schemes and accreditation for handling dangerous substances. Any exception could be time-limited to grant the company time to develop alternative substances or processes. Tolerable Uses Tolerable uses would apply to substances when overall risk reduction is necessary, but setting EU-wide restrictions or authorisations is identified as ineffective, inefficient or unnecessary. For instance, it may be that risk reduction is necessary in only a few Member States due to particular industrial activities (e.g., reduction of the probability 191
Framework for Chemical Risk Management under REACH that workers are exposed above a DNEL). In other cases it may be that there is a need to establish best practice across the EU through a framework such as IPPC (e.g., reduction in overall emissions). Although implemented using legislation outside REACH at a national level, measures to achieve targets would affect decision-making on restriction and authorisation. Tolerable uses would be linked to target-setting schemes developed under REACH risk reduction decision-making. The concept of integrating target-setting schemes with decision-making under REACH stems from interviewee responses regarding the need for better co-ordination of risk management controls under other legislative frameworks, national legislation and on the lack of flexibility to achieve risk reduction through measures EU-wide restriction and authorisation under REACH. By establishing a Monitoring Network to devise targets and report on their achievement, the target-setting mechanisms under the systems framework would also respond to the need expressed by the interviewed regulators for timelines for implementing measures under other legislative frameworks. If targets were not achieved, then specific uses could be made to restrictions or authorisations. However, during the first process of review, for consistency with other decision-making, tolerable uses would not apply to safe, permissible, authorised or restricted uses. Targets for tolerable uses could be set and based on a number of criteria such as: UÊ ÕÃiëiVwVÊÛÕiÃÊ>ÃÊ>Ê«ÀÝÞÊvÀÊ
Õ>Ê>`ÊiÛÀiÌ>Ê exposures (e.g., as a percentage decrease in the use of a substance for a particular process), UÊ VVÕ«>Ì>Ê
i>Ì
ÊÃÌ>ÌÃÌVÃÊi°}°]Ê>ÃÊ>Ê«iÀViÌ>}iÊÀi`ÕVÌÊÊ health incident reports resulting from the use of the substance within a sector), UÊ `ÕÃÌÀÞÊ «iÀvÀ>ViÊ `V>ÌÀÃÊ i°}°]Ê >ÃÊ >Ê Ài`ÕVÌÊ Ê Ì
iÊ number of accidents within a sector), 192
A Systems Framework to Implement REACH UÊ «Ài`VÌi`ÊÀÊi>ÃÕÀi`ÊiÝ«ÃÕÀiÊiÛiÃÊi°}°]Ê`i}ÀiiÊÊÀi`ÕVÌÊ of exposure levels relative to a DNEL or predicted no-effect concentration (PNEC)), or UÊ LiV
>À}ÊÃV
iiÃÊi°}°]Ê`iVÀi>Ãi`ÊÛÕiÊvÊÌ
iÊÃÕLÃÌ>ViÊ present in certain products or decreased volume of substance use in a sector). Achieving targets remains the responsibility of Member States through a variety of measures, including incentive-based schemes, voluntary measures and local permitting schemes. In some cases, decision-making on the targets can include co-ordination at the EU level such as establishing or communicating best practice. The use of targets would differ from environmental or occupational standards for specific chemicals or chemical groups through existing EU legislation such as the Water Framework Directive, Air Quality Framework Directive [488] or Chemical Agents Directive, but could supplement any target-based schemes found under these pieces of legislation. Publishing tolerable uses on the REACH-IT website could then promote the communication and co-ordination of a wide set of risk management and policy measures within industry subgroups at either EU or national levels19, including eco-labels and industry voluntary agreements [489]. A particular benefit of these activities is their ability to improve industrial relations with consumers, banks and insurance companies [490]. A method for target-setting could promote a level playing field20 for companies operating in the EU while avoiding the need to establish harmonised controls [213]. Member States would retain the ability to enact measures beyond EU targets as under legislation based on Article 175, 153 or 138 of the EC Treaty, but use different 19
For instance, the German Textile Trade Association (TEGEWA) has developed a method to rank and label chemicals according to three levels of priority for reducing emissions to water [491]
20
A specific objective would be to avoid potential for the ‘bidding down of standards’ that can result in the European internal market from competition between rules (see [493]).
193
Framework for Chemical Risk Management under REACH regulatory tools to achieve comparable (minimal levels) of risk control. Therefore, target-setting could prove particularly relevant for managing different national approaches at the EU level, as described as ‘national dimension’ in Box 5.3: Box 5.3 National dimensions – based on interview data Target-setting could be a relevant process for risk reduction under the following scenarios: UÊÊ7
iÊÀi«ÀÌÃÊvÊVVÕ«>Ì>ÊÕÀÞÊÀÊiÃÃÊVVÕÀÊÜÌ
Ê a particular Member State resulting in a need for stringent >Ì>ÊÀi}Õ>ÌÀÞÊVÌÀÃÆ UÊÊ7
iÊ ÃiÊ iLiÀÊ -Ì>ÌiÃÊ
>ÛiÊ `iV`i`Ê ÌÊ «
>ÃiÕÌÊ certain chemicals through international conventions, such as the Marine Convention, but not all Member States are VÌÀ>VÌ}Ê«>ÀÌiÃÊÃiiÊ>ÃÊQ{ÓR®Æ UÊÊ7
iÊiÛiÃÊvÊÃ>viÌÞÊvÀÊÀÃÊ>>}iiÌÊi>ÃÕÀiÃÊ`ÊÌÊ adequately cover variations in the use of assessment factors or differences in anticipated exposure levels in risk assessment data (see also [477]) A further argument for target-setting is that the practicality and acceptability of implementing a measure depends on the regulatory administrative structure and industrial landscape of a country. For instance, broadly comparable reductions in the use of trichloroethylene have been achieved using technical regulations in Germany, company-specific authorisations in Sweden and taxes on chlorinated solvents in Norway [246]. The use of targets also provides greater sensitivity to varying environmental conditions than command and control legislation. The exploitation of environmental conditions present within a Member State that enable a chemical to be naturally absorbed and degraded more rapidly than in another country constitutes an ‘entirely legitimate source of comparative advantage’ [494]. 194
A Systems Framework to Implement REACH To summarise, target-setting and subsequent reporting must reflect national differences in production, use, and environmental conditions, as well as alternative regulatory measures and policy instruments that can be implemented to achieve risk reduction. Just as with any benchmarking type scheme, target-setting must be carefully constructed so as to avoid ‘league tables’, otherwise the complexity of information may be misconstrued and the fundamental objectives of achieving risk reduction through chemical safety may be neglected [382]. Setting targets must also accommodate the recent past activities of industrial sectors so as not to unfairly benefit previously poor performers [382].
5.3.2 Technical Guidance for Decision-Making During REACH implementation, regulatory decision-making will need to identify chemical uses that need the application of further risk reduction. According to the legislative text, this process will occur via an Annex XV dossier21, which must present a justification for the need for risk reduction and the selection of the appropriate risk reduction strategy. The Annex XV dossier would resemble the current RRS procedure but, in addition to the possibility for authorisations, the dossier includes a format for proposals for the harmonisation of classification and labelling of a substance. A 247-page draft TGD for producing an Annex XV dossier was published in May 2006 as part of RIP 4.4 [495], which formed the basis for other guidance in 2007. It presents an excellent, highly technical explanation of how to consider alternative regulatory options. For instance, after explaining the interaction of REACH procedures, how to assess information for the harmonisation of classification and labelling 21
Recent amendments have changed the number of the legal recitals, articles and annexes vÊÌ
iÊi}Ã>ÌÛiÊÌiÝÌÆÊÊ>Ê«ÀiÛÕÃʼÛiÀýÊvÊÌ
iÊ`À>vÌÊi}Ã>ÌÛiÊÌiÝÌ]ÊÜ
>ÌÊÃÊÜÊ referred to as an Annex XV dossier was an ‘Annex XIV dossier’.
195
Framework for Chemical Risk Management under REACH and how to manage potentially confidential data, it begins by stating that an Annex XV dossier for restriction must be able to substantiate an initial concern over a risk to human health or the environment [495]. The document then goes on to explain how to consider alternative risk management options in terms of whether: (1) increased enforcement would alter the risk level, (2) further risk management measures would be needed to supplement any restriction, (3) information on alternatives is available, as well as any socio-economic data, and (4) other legislative frameworks could be used to reduce exposure. A detailed procedure of how to check a risk assessment is explained, such as comparing exposure to DNEL or PNEC. Examples provided in RIP guidance are self-evident. Overall, the RIP project adds little to the current EU RRS TGD when used in conjunction with the supplementing the Risk Assessment TGD. Arguably, the RIP creates a reference document for regulators rather than that a set of simple-to-follow structures to decision-making or a template to organise the results of a risk assessment. The framework presented in this thesis therefore provides a potential structure to inter-link various decision-making processes of REACH and has been checked to ensure that it does not clash with guidance developed in the RIPs, including RIP 4.3 on Inclusion of Substances for Authorisation. The general guidance produced in RIP 4.4 serves as an important source of information on how to identify and assess the multiple variables that input to risk management decision-making. Given that the implementation of REACH will require many additional staff from Member State regulatory authorities, including the new Member States that have little or no experience with the previous RRS process, and the hiring of some 400 personnel for the European Chemicals Agency, outcomes of RIP 4 will prove valuable to many individuals. Fundamentally, this is where the research of this PhD thesis differs to RIP 4: the system framework creates a
196
A Systems Framework to Implement REACH tool for decision-making. The proposed systems framework also co-ordinates many different decision-making processes, including the interaction between RIP 4.3 and 4.4, that are not subject to investigation by the RIPs such as when and how to communicate risk assessment data to stakeholders22. The systems framework proposes to sort chemical uses according to restriction, authorisation (including national permitting schemes) or target-setting (i.e., restricted, authorised. or tolerable uses). To manage this process, a decision-making matrix has been developed that would serve as a: 1. Format to present risk assessment results for the purpose of risk reduction decision-making. 2. Method to recommend appropriate regulatory options, including target-setting. 3. Tool to compare proposals for authorisation, restriction and other regulatory risk reduction options. An overview of the proposed risk reduction decision-making process is shown in Figure 5.10. The primary aim is to input risk information into the decision-making matrix that will then generate a set of ‘most suitable’ regulatory options23. The final selection of the regulatory option is left to decision-making between EU regulators during implementation and must follow the broad legislative requirements summarised in Appendix 5.1.
22
The REACH legal text includes several mechanisms to ensure that confidential business information and other proprietary data are not disclosed.
23
By comparison, according to the REACH legal text, proposals for substances subject to restrictions or authorisations would undergo separate processes and not include any specific mechanism to co-ordinate decision-making with other legislative frameworks.
197
Framework for Chemical Risk Management under REACH
EU Authorisation Member State Permitting Step 4 Selection of regulatory options (Figure 5.11 and Annex 5.1)
Restrictions Targets Further risk assessment
Step 3 Input risk descriptors into the decision making matrix (Figure 5.11)
Step 2 Exclude any risks that should be regulated under other legislative frameworks (eg. construction products, packaging materials) (see Box. 5.5)
Combination of the above options
Exclusions A decision to exclude a risk should be communicated to relevent decision makes under the legislative frameworks
Step 1 Categorise risk assessment results according to the seven Risk Descriptors - apply conclusion “(i), (ii), (iii)” (see Box 5.4)
Risk assessment (i.e., registration or evaluation dossier)
Figure 5.10 Risk reduction decision-making under the systems framework
198
A Systems Framework to Implement REACH As a first step to the risk reduction process, the results of a risk assessment report (i.e., a registration or evaluation dossier) must be organised according to seven risk descriptors shown in Box 5.4. Box 5.4 Risk descriptors for the decision-making matrix (step 1) I
General risks to human health and ecosystems as a result of emissions that arise from disperse sources of pollution or contamination, including releases during use and disposal of consumer or professional products.
II
Specific emissions that result in risks to human health and ecosystems that result from point industrial sources or specific uses.
III
Immediate (high-level) risks to human health, wildlife or ecosystems from exposures via the environment.
IV
Future risks arising from increasing concentrations of chemicals in environmental media or biological systems, ecosystems and human health via the environment.
V
Professional user risks resulting from direct exposure during product use and including occupational exposures not occurring in industrial facilities (e.g., outdoor or underground applications).
VI
Consumer risks resulting from direct exposure during product use a. and.
VII National dimensions: where risks are limited to specific occurrences within certain Member States, interpretation of risk levels and regulatory controls vary significantly between Member States, and control measures do not mandate EU harmonised regulation for achieving risk reduction (see Box 5.3,).
199
Framework for Chemical Risk Management under REACH As part of this first step, one of three possible conclusions used in current EU risk assessment reports would be applied for risk descriptors I to VI: (i) there is a need for further information and/or testing, i.e., significant uncertainties in risk assessments have been identified and disagreement arises on the need for further risk reduction measures before or after completion of the iÛ>Õ>ÌÊ«ÀViÃÃÆ (ii) there is at present no need for further information and/or testing and no need for risk reduction measures beyond those which are currently being applied, Ê
°i°]Ê Ý«ÃÕÀiÊ ÊÀÊ«Ài`VÌi`Ê* Æ
(iii) there is a need to limit the risks, i.e., Exposure>>DNEL or PNEC or a DMEL24 or exposure levels are significant and a PNEC or DNEL cannot be established. Risk descriptor VII, ‘national dimensions’, would be defined as whether or not target-setting or national permitting schemes should be applied for the scenarios that were described in Box 5.3 on. Examples of four types of uncertainty that can lead to conclusion (i) are provided in Table 5.5. As a second step, prior to entering the risk descriptors into the decisionmaking matrix, a potential risk that requires particular regulatory attention needs to be excluded from further decision-making (Figure 5.10). Reasons for excluding potential risks include: UÊ ÝÃÌ}ÊÀiÃÌÀVÌÃÆ 24
A DMEL applies to non-threshold effects or substances where a DNEL or PNEC cannot be established (Section 2.1.2).
200
A Systems Framework to Implement REACH
Type I Hazardous properties
- Epidemiological evidence indicates inherent hazardous properties for a substance but existing laboratory test results lead to the conclusion that the substance is ‘non-dangerous’. - Evidence indicates endocrine disruption but Member States do not agree on the reliability of the findings. - Evidence indicates that metabolites or biodegradation products possess similar hazardous properties as the parent compound.
Type II Dose–response Concentration– effect
Type III Exposure
- Inconsistencies arise in the extrapolation of a DNEL from lower to higher test-level organisms when using different data sets. - Existing data prove insufficient to establish the shape of a dose–response curve at dosages corresponding to anticipated exposure levels. - Adequate levels of safety for consumer spray applications depend on a fixed particle size distribution that varies significantly according to the product manufacturing process. - Environmental emission data are available only as average values (not minima and maxima). - Exposure levels primarily depend on the volume of a substance imported or used on the EU market according to relevant product categories, but it is not possible to generate an approximate values for these volumes.
Type IV Risk characterisation
- Lack of data exists on the level of compliance, compliance monitoring and enforcement for a high level risk that requires specific safety measures to ensure adequate protection. - Insufficient monitoring or modelling data exists to determine the extent of trans-boundary pollution caused by emissions to environment. - Exposure data of ecosystems and humans via the environment primarily depend on emissions from incorrectly disposed products.
Table 5.5 Examples of uncertainty that could lead to conclusion (i) under the risk descriptors
201
Framework for Chemical Risk Management under REACH UÊ «>ÞÊëiVwVÊiÝi«ÌÃÆ UÊ Permissible uses; UÊ ÌÀÊÕ`iÀÊÌ
iÀÊi}Ã>ÌÛiÊvÀ>iÜÀð A decision to exclude a risk must be communicated to the relevant decision-makers and stakeholders. For instance, if it is decided that a specific chemical use should be controlled under another legislative framework, the relevant Directorate-General of the European Commission must be notified as well as the national Member Statecompetent authorities. That means that REACH will need to establish the communication structures between the relevant institutions and organisations (see Section 5.3.3). Once the data are inputted into the decision-making matrix shown in Figure 5.11, the corresponding possible regulatory options can be identified. As a final step, the selection of a regulatory option should be justified according to effectiveness, practicality and monitorability as detailed in Appendix 5.1. Several combinations of regulatory options can be considered, as shown in rows C1–C4 of Figure 5.11. Conducting socio-economic analyses following the collection of information via stakeholder consultation is part of the restriction and authorisation procedures foreseen under REACH. Additional data collection and analysis could remain optional at the national level, especially for target-setting. The general type of information that can be collected is specified in Annex XVI of the REACH regulation (on carrying out socio-economic analyses) and follows the same points discussed in Section 2.3.3 of the literature review25.
25
Annex XVI provides a very brief two-page summary of information that can be considered when conducting a socio-economic analyses such as: impact of bans on «À`ÕVÌÊ «iÀvÀ>ViÊ ÀÊ >Û>>LÌÞÆÊ vi>ÃLÌÞÊ vÊ >ÌiÀ>ÌÛiÊ «ÀViÃÃiÃÊ ÀÊ «À`ÕVÌÃÆÊ VÃÌÃÊvÊ«iiÌ}ÊÀÃÊ>>}iiÌÊi>ÃÕÀiÃÆÊÃV>Ê«V>ÌÃÊÊLÊÃiVÕÀÌÞÆÊ «>VÌÊvÊÀi}Õ>ÌÊÊiÛÀiÌ>ÊÀÊVÃÕiÀÊ«ÀÌiVÌÆÊ>`ÊÜ`iÀÊ«V>ÌÃÊ on industrial competitiveness.
202
A Systems Framework to Implement REACH
Box 5.5 Exclusion from the decision-making matrix (Step 2) The following parameters would define whether a substance, chemical grouping or use should be excluded from further RRS development and selection decision-making: Exclusion Existing restrictions
Companyspecific exemptions
Substances that provide certain safety or utility functions (permissible uses)
Chemical uses already specifically covered by product legislation
Example:
Exceptions
CMR substances None and preparations for consumer use
Uses for land Àii`>ÌÆÊ accident ÀiëÃiÆÊ chemical management services
Company exemptions for the use and handling of dangerous substances should be granted at a national or EU level for conducting remediation projects, responding to accidents (contamination containment and clean-up), or carrying out other relevant activities. Any exception could be time-limited to grant companies time to develop alternative substances or processes.
Permissible uses should not have been prioritised for evaluations or other Colours in decision-making, and could even be Ü>À}ÊÃ}ÃÆÊ issued with time-limited exceptions processing to restriction or authorisation. A chemicals for company should be able to identify pharmaceuticals whether specific uses meet these demands prior to registration.
ÃiÌVÃÆÊ construction «À`ÕVÌÃÆÊ «>V>}}ÆÊ plant protection products
REACH should provide the mechanism to identify risks from substances not specifically covered by product legislation, e.g., environmental fate of cosmetic or emulsifiers used in plant protection products. However, in many cases other legislative frameworks may be more appropriate from devising and enacting the appropriate risk reduction measures.
203
Regulatory options (guidance) VII
Environment* I
II
III
IV
General
Specific
Immediate
Future
1
(i) or (iii)
(i), (ii) or (iii)
(i) or (iii)
2
(i), (ii) or (iii)
(ii)
3
(i), (ii) or (iii)
4
V
VI
National
Consumer Professional
dimension
(i) or (iii)
(i), (ii) or (iii)
(i), (ii) or (iii)
No
(ii)
(i) or (ii)
(ii)
(i), (ii) or (iii)
Yes
(iii)
(iii)
(iii)
(ii)
(i) or (ii)
No
(i) or (ii)
(i), (ii) or (iii)
(i) or (ii)
(i) or (ii)
(i) or (iii)
(i) or (ii)
Yes/No
5
(i) or (ii)
(i), (ii) or (iii)
(i) or (ii)
(i) or (ii)
(ii)
(iii)
No
6
(i) or (iii)
(i), (ii) or (iii)
(i), (ii) or (iii)
(i) or (iii)
(ii)
(i), (ii) or (iii)
Yes/No
7
(ii)
(i), (ii) or (iii)
(ii)
(ii)
(ii)
(i), (ii) or (iii)
No
8
(i)
(i)
(i)
(ii)
(i)
(i)
No
-see Appendix 5.1 for final selection criteriaEU authorisation MS permitting Restrictions for industrial uses Restrictions for consumer products Restrictions for professional products Targets No further EU action Further information necessary via the Evaluation process or monitoring schemes
*Includes risks to human health via the environment 9\ÊVÃÊrÊVÃÕiÀÊ«À`ÕVÌÃÆÊ«ÀvÊrÊ«ÀviÃÃ>Ê«À`ÕVÌÃÆÊ®ÊÌ
iÀiÊÃÊ>Êii`ÊvÀÊvÕÀÌ
iÀÊvÀ>ÌÊ>`ÉÀÊÌiÃÌ}]Ê®ÊÌ
iÀiÊÃÊ>ÌÊ present no need for further information and/or testing and no need for risk reduction measures beyond those which are currently being applied, (iii) there is a need to limit the risks
Figure 5.11 Decision-making matrix (Step 3)
Framework for Chemical Risk Management under REACH
204
Risk descriptors (according to risk assessment conclusions)
A Systems Framework to Implement REACH Information from socio-economic analyses can serve not only to select between alternative risk management measures in the decision-making matrix but can identify the relevant timelines for implementation (e.g., resulting from process change, product reformulation or instillation of pollution abatement technology). In the case of authorisation, REACH specifies that socio-economic data will form a key variable in the identification of suitable alternatives and the period for granting an authorisation before it needs to be reviewed.
5.3.3 Administrative Structures For efficient decision-making, the European Chemicals Agency will need to establish a network for decision-making across the relevant occupational, consumer and environmental legislative frameworks. Judging from the responses of the regulators interviewed, decisionmaking under REACH will require frequent interactions between Directorate-General Environment, Directorate-General Enterprise and Industry, Directorate-General Employment, Social Affairs and Equal Opportunities and Directorate-General SANCO (Health and Consumer Protection) (see Sections 5.1 and 5.2). Unfortunately, the REACH legislation does not propose a detailed structure for coordinating activities between the various Directorates-General, the corresponding Member State Competent Authorities and the new Chemicals Agency. The systems framework also proposes to extend responsibility for decision-making to include a wider set of stakeholders than anticipated under REACH. The current REACH legislative proposal limits formal stakeholder inputs to restriction and authorisation processes mediated via the REACH-IT website. Under the systems framework, a mechanism would be created to enable stakeholder groups to input data relating to the registration and evaluation processes, such as information on the appropriateness of various risk management measures. A separate network would then be responsible for monitoring and reporting the implementation of risk reduction measures adopted under REACH, as well as reviewing any decision-making rules. 205
Framework for Chemical Risk Management under REACH Stakeholder Inputs Registration dossiers from industry would continue to provide the primary input for risk assessment and subsequent decision-making. Regulators, institutions, NGO and academic research centres in EU Member States would be able to submit data directly to decisionmaking processes through a system of peer-review operated by the European Chemicals Agency. In addition to toxicological or epidemiological studies, information from the Stakeholder Inputs would include references to research on the effectiveness of risk reduction measures for risk management decision-making26 . To limit the potential number of information sources to a manageable level and avoid duplicate submission of relevant information, Stakeholder Inputs should be co-ordinated at the EU level through European organisations or national Member State regulatory authorities. A dedicated unit within the Chemicals Agency would then be responsible for the administration of the information. Monitoring network Because decision-making under REACH will frequently depend on the ability of other legislative frameworks to control chemical risks (Section 2.4.3), the European Chemicals Agency will need a Monitoring Network to co-ordinate regulatory activities across the Directorates-General of the European Commission27. In particular, the Monitoring Network would be responsible for tracking and reporting on the implementation of targets under target-setting and standards under restriction and authorisation. As shown in Figure 5.12, besides staff from the Member States and the European Chemicals Agency, the Monitoring Network could comprise 26
In this way, insurance companies, occupational health institutes, trade unions or other organisations can input monitoring data and inform regulators of the effectiveness of any sector-specific guidance or existing risk management measures are having on achieving risk reduction.
27
While the need to co-ordinate the activities of the European Commission DirectorateGenerals was raised by many interviewees, the European Policy Network has also identified this as being necessary for the implementation of REACH [496]
206
A Systems Framework to Implement REACH
CONSUMER
SANCO DG
CHEMICALS AGENCY MEMBER STATE AUTHORITIES CEN ENTR DG ENV DG EEA ENVIRONMENT
OSHA EMPL DG OCCUPATIONAL
Figure 5.12 Monitoring network: box locations indicate the responsibilities of the organisations for consumer, occupational and environmental protection
representatives from two European Agencies, the European Centre for Standardisation (CEN) and four Directorates-General of the European Commission. Each of these European institutions is placed in Figure 5.12 to indicate its relative responsibilities in regulation aimed at consumer, occupational and environmental protection, as detailed in Box 5.6. The triangle in the background of Figure 5.12 represents the continua between these three inter-related areas vÊV
iV>ÊÀÃÊ>>}iiÌÆÊvÀÊÃÌ>Vi]ÊL>}ÊÌ
iÊÃ>iÊvÊ>Ê substance to reduce direct consumer exposure has implications for releases to the environment28. Directorates-General Enterprise and Industry and CEN are centrally positioned in Figure 5.12 because the enforcement of restriction and authorisation usually requires harmonised product test standards. 28
Similarly, banning a product for occupational use will affect releases to the environment during manufacturing that may ultimately result in consumer exposures.
207
Framework for Chemical Risk Management under REACH Box 5.6
Monitoring network representative organisations and Commission Directorates-Generals -the existing chemical activities of each organisation are described below-
OSHA - European Agency for Safety and Health at Work UÊ
Ê À`ÕViÃÊ ÃiVÌÀëiVwVÊ LiÃÌÊ «À>VÌViÊ `VÕiÌÃÊ VÛiÀ}Ê « occupational protection
UÊ
ViVÌÃÊ>`ÊLiV
>ÀÃÊVVÕ«>Ì>Ê
i>Ì
ÊÃÌ>ÌÃÌVÃ
EEA - European Environment Agency UÊ
Ài«ÀÌÃÊÊÌ
iÊ«iiÌ>ÌÊvÊ 1Êi}Ã>ÌÊ
UÊ
ÌÀÃÊÌ
iÊÃÌ>ÌiÊvÊÌ
iÊiÛÀiÌÊ
CEN - European Centre for Standardisation UÊ
>ÀÃiÃÊ«À`ÕVÌÊÃÌ>`>À`ÃÊ>`ÊÌiÃÌÊiÌ
`Ã
UÊ
>ÀÃiÃÊiÛÀiÌ>ÊÌÀ}ÊÃÌ>`>À`Ã
ENTR - Enterprise and Industry DG UÊ
ÊÃiÌÃÊ>ÀiÌ}Ê>`ÊÕÃiÊÀiÃÌÀVÌÃÊÕ`iÀÊ ÀiVÌÛiÊÇÈÉÇÈ
UÊ
Ê À}>ÃiÃÊ «À`ÕVÌÊ ÃÌ>`>À`ÃÊ vÀÊ iÜÊ >««À>V
Ê `ÀiVÌÛiÃ]Ê e.g., toys, construction Products
ENV - Environment DG UÊ
ÊVÀ`>ÌiÃÊiÛÀiÌ>Êi}Ã>Ì]ÊÃÕV
Ê>ÃÊiÃÃÊÌÊ values and waste
UÊ
ÃÊ iÌÃÊÀiÃÌÀVÌÃÊÕ`iÀÊ ÝÌi`i`Ê*À`ÕViÀÊ,iëÃLÌÞÊÀÊ other product regulation
EMPL - Employment, Social Affairs and Equal Opportunities DG UÊ
>>}iÃÊVVÕ«>Ì>Ê
i>Ì
Ê>`ÊÃ>viÌÞÊi}Ã>Ì
UÊ
VÀ`>ÌiÃÊ>`ÊÃiÌÃÊVVÕ«>Ì>ÊiÝ«ÃÕÀiÊÃÌ>`>À`Ã
SANCO - Health and Consumer Protection DG UÊ
ÊÌÀÃÊ>`ÊÃiÌÃÊVÃÕiÀÊ«ÀÌiVÌÊi}Ã>Ì]ÊVÕ`}Ê General Product Safety
UÊ
ÊÀiÛiÜÃÊ>`ÊV>ÌiÃÊ`>Ì>ÊÀi>Ì}ÊÌÊLV`iÃ]Ê«>ÌÊ«ÀÌiVÌÊ products, medicinal products
208
A Systems Framework to Implement REACH Although the power relations and remits for the institutions involved in the Monitoring Network would remain the same as under current EU legislation, a member of staff from each organisation would be made permanently available for REACH decision-making. In some cases, specialist risk assessment knowledge may be needed for decision-making under REACH, such as exposure data maintained by the Directorate-General for Consumer Health and Protection, DG SANCO29. In other cases, temporary chemical bans may need to be rapidly implemented under other legislative frameworks30. As discussed in Section 5.2, although controlling occupational exposures rarely requires EU market restrictions (see also [497]), decisionmaking on the appropriateness of risk reduction measures may depend on the development of best practice documentation by OSHA. If regulators identify new risk data on substances that have already undergone registration and evaluation, perhaps via Stakeholder Inputs, the Monitoring Network should be responsible for assessing if it is necessary to review previous risk management decisions. A further role of the Monitoring Network would be to promote the harmonisation of administrative procedures (as opposed to the standards) for companies demonstrating compliance under various EU legislative frameworks (see Section 5.2).
5.4 Prioritisation of Regulatory Decision-Making The need to develop a method of prioritisation under REACH is highlighted by the fact that approximately 3,700 substances are anticipated to be registered under REACH within the first 3.5 years of implementation (Section 2.5.1). In the history of chemical regulation, such a large number of risk assessments have never been generated over such a short time period. Several substances will be 29
Active substances present in plant protection products, biocides or medicinal products are outside the scope of REACH but product formulants such as emulsifiers are included.
30
As evident from previous European Commission Decisions under the General Product Safety Directive [498] that were only followed by restrictions under Directive 76/769 when sufficient evidence indicated the necessity for harmonised EU technical standards.
209
Framework for Chemical Risk Management under REACH simultaneously identified as needing certain uses controlled through Ài}Õ>ÌÀÞÊ >VÌÆÊ Ài}Õ>ÌÀÃÊ ÜÊ Ì
iÊ ii`Ê ÌÊ «ÀÀÌÃiÊ Ü
V
Ê chemicals should be regulated first. EU regulators will also have to balance their resources between the procedures for restriction and authorisation. As seen in Chapter 4, a focus on a few highly hazardous substances through a procedure of authorisation could otherwise result in the neglect of regulatory or corporate activities in managing other hazardous substances. EU decision-making currently prioritises risk assessment and management by combining “expert judgement” with a tool called EURAM (EU Risk Ranking Method) [499]. EURAM sorts chemicals according to existing hazard data, production volumes, potential exposure scenarios and data gaps. As an extremely technical programme, interpreting EURAM results can be challenging, which limits decision-making transparency (see [500]). In practice, most of the interviewees described the prioritisation of EU activities as primarily based on national regulatory activities that, in turn, are typically based on political influences that vary between regulatory administrations (Chapter 4). In addition to a lack of division between ‘political’ and ‘expert’ judgment, the current system for prioritisation will need to be reformed because EURAM was developed to operate with mostly incomplete data sets. Under REACH, regulators will be faced with needing to prioritise registration and evaluation dossiers, as well as any subsequent proposals for restriction and authorisation. Methods of Prioritisation The REACH legislation stipulates prioritisation for evaluation and authorisation based only on hazardous properties, production and use volumes, and whether a chemical has a “wide dispersive use”. Nevertheless, it is possible that the REACH TGDs could incorporate methods for prioritisation for evaluation, restriction or authorisation based on the Dutch SOMS or the method proposed in the UK RCEP’s report Chemicals in Products (refer to Section 5.2). The SOMS method for prioritisation according to hazard and
210
A Systems Framework to Implement REACH use is straightforward: a highly hazardous substance intended for consumer use ranks higher than one for industrial use, whereas a substance with low hazardous properties would generally not be of any concern [316, 317]. SOMS also makes some obvious distinctions within use categories as in the case where ‘open-batch’ processes and direct emissions to the environment hold a greater regulatory concern than industrial ‘closed-systems’ [316, 317]. While monitoring data have recently been incorporated into hazard definitions of SOMS [317], the RCEP’s proposed system for chemical control makes full use of monitoring data. As noted in Section 5.2, the RCEP approach would focus on identifying and regulating synthetic chemicals in humans, marine mammals and top predators, to account for potentially increasing levels of chemical exposure that may otherwise be overlooked by REACH. The RCEP would also prioritise chemicals for further investigation if “found in unexpected environmental compartments or at unexpected concentrations, or associated with unusual biological phenomena” [501]. A further distinguishing feature of the RCEP system is that it proposes that a stakeholder forum should be responsible for refining and reviewing Ì
iÊ«ÀÀÌÃ>ÌÊVÀÌiÀ>ÆÊÃÕV
Ê«ÌiÌ>ÊvÀÊÌ
iÊVÕÃÊvÊÃV>Ê dimensions of risk is not considered in REACH or SOMS. A particular limitation of the above methodologies is their ability to evaluate diverse exposure scenarios. In particular, neither SOMS nor the RCEP create methods to weigh different sources or pathways of exposure. An alternative prioritisation method for regulatory action has therefore been developed within the systems framework presented here. The proposed methodology qualitatively compares the overall risks resulting from the production and use of each chemical. These chemical-specific priority rankings would then serve to distinguish between different levels of regulatory response. Incorporating Social Dimensions into a Prioritisation Method Because risk levels depend on how the risk (i.e., hazard and exposure) is perceived by individuals or social groups (Section 1.2), any method
211
Framework for Chemical Risk Management under REACH
Exposure
Social mobilisation
Need to take regulatory action
Hazard Figure 5.13 Factors for prioritisation of regulatory action
for comparing chemical risks and the corresponding prioritisation of regulatory action must include social dimensions of risk perception (Figure 5.13). For example, based on previous experiences of risk management set out in Chapter 4, a lack of perceived benefits of a hazardous chemical to the exposed population and the potential availability of a less hazardous substitute can increase the probability that a NGO will trigger media campaigns that heighten or exaggerate the perceived consequence of exposure to that chemical. Actors within society may also group chemicals together based on association rather than actual hazard, as exemplified by ‘victim associations’ in France referring to all glycol ethers as ‘reprotoxic’ (Section 4.5). In turn, such social amplification of risk can easily result in a loss of ‘trust’ in regulatory authorities or worsen the ‘public image’ of the chemical industry (Section 1.3), which is exactly what appears to have happened in Sweden (Sections 4.5 and 4.6). The phenomenon of political campaigning by interest groups, or even Member States, to ban chemicals based on hazard rather than risk is what some researchers describe as ‘social mobilisation’ [12]. Social mobilisation appears to be initiated by (i) the potential hazardous properties of a substance, regardless of exposure or (ii) the exposure of vulnerable groups (e.g., children or elderly) to a chemical, regardless of hazard. Reporting commonly used household products as containing ‘potentially carcinogenic’ or ‘potentially reprotoxic’
212
A Systems Framework to Implement REACH substances can attract media attention and generate regulatory scrutiny, even if there is a very high DNEL for the substance and very low exposure31. Public perception of a risk often becomes heightened when there is a perceived lack of voluntary control on exposure scenarios and exposure levels. It follows that human exposure to chemicals via the environment as a result of pollution or contamination tends to be a particular concern and a useful lobbying focus for many environmental NGO. Social mobilisation can result in risk level attenuation rather than amplification [505]. For instance, when a chemical is perceived as being of particular value to society, the level of risk may be more acceptable to many members of the public [506]. Need to take regulatory action
Event C: Social mobilisation
Event A: Hazard
Event B: Exposure
Event B: Exposure
Other possible sequence or events
Event C: Social mobilisation Event A: Hazard
Figure 5.14 Hazard, exposure and social mobilisation – two possible sequences of events
31
Such an incident recently occurred with ‘borates’, where substances involving borate salts were grouped together as potential reprotoxins. As a consequence, borates received VÃ`iÀ>LiÊi}>ÌÛiÊ«ÀiÃÃÊ>ÌÌiÌÊ>`ÊÌ>À}iÌ}ÊLÞÊ "Êi°}°]ÊQxäÓRÆ®ÊiÛiÊÌ
Õ}
Ê exposure levels in consumer products were negligible and other effects such as vomiting would occur before reaching any exposure level of regulatory concern [503, 504].
213
Framework for Chemical Risk Management under REACH The production and use of a chemical must therefore not only be considered as having the potential to cause damage to human health or the environment, but also the potential to create an adverse effect on society due to perceptions of equity violation amongst stakeholder groups and any associated social mobilisation [12]. What is clear is that the need to take regulatory action depends on a series of events which can be observed to involve hazard, exposure and social mobilisation32 (Figure 5.14). In some cases, regulatory action may involve conducting a risk assessment. In other cases, regulatory action may require improving risk communication to stakeholders or the general public. Prioritisation of regulatory action under the systems framework must account for social mobilisation. To date, the potential for social mobilisation and the actual socio-political consequences of social mobilisation are rarely included in risk assessments or analyses of alternative risk management options. Methods to measure the extent of damage caused by the production and use of a chemical tend to rely on assessing the hazardous properties of the substance and the number of exposed humans, organisms or ecosystems (see Section 2.1.4). Evaluating the consequences of alternative regulatory options is usually limited by the complexity of quantifying and comparing levels of damage to human health and the environment. These assessments tend to be based on scientific or economic data and, based on the comments from the regulators interviewed, EU risk management decision-making then struggles with agreeing on value judgments
32
For instance, following Figure 5.14, the identification of a substance as hazardous [Event A] can trigger a risk assessment that identifies exposure to a large number of the public that live near the manufacturing facility [Event B]. In turn this can result in social mobilisation in that local community [Event C] demanding increased regulatory control of emissions. Another example of a series of events illustrated in Figure 5.14 would be a general concern by certain NGO over the presence of synthetic chemicals in human blood of some members of the public [Event C]. The social mobilisation generated by the NGO leads to a biomonitoring survey that detects a number of synthetic substances present in blood of the general public [Event B]. Industry then carries out hazard assessments of the substances and conclude that the exposure levels reported are usually far lower than any effect level [Event A]. However, regulators must still take regulatory action in the form of restricting some uses of the substances that result in high exposure levels.
214
A Systems Framework to Implement REACH that are needed to establish risk levels and the corresponding level of regulatory attention needed for controlling the risk33. Compared with analysing the socio-economic consequences of chemical production and use in terms of the impact on variables such as employment and industrial competitiveness, EU risk management decision-making clearly lacks a method to evaluate the consequence that perceived chemical risks have on perceptions of equity. Moreover, there appears to be little consideration of how decisions can affect trust in regulatory institutions. Yet, as described in Chapter 2 and explored in Chapter 4, these social dimensions of risk prove fundamental to regulation. To include the potential and the consequence for social mobilisation into a method for the prioritisation of regulatory action to control the manufacture and use of substances under REACH, this thesis proposes an approach adapted from a method for evaluating risks developed by [12]. Here the extent of damage that can be caused by a hazard is described according to damage to human health, the environment and society. Following a review of the risk literature, [12] identified nine risk criteria for comparing levels of risk. These criteria are then grouped to give a risk level in terms of overall extent of damage that can occur and the probability for that extent of damage to occur. For the purpose of prioritising regulatory action under REACH, this thesis proposes to reduce the nine risk criteria identified by Klinke and Renn to four categories that can be used to characterise risk levels: hazard, exposure, social mobilisation and probability. The grouping of the nine specific risk criteria into the four fundamental prioritisation criteria relevant for regulatory chemical risk management is shown in Table 5.6.
33
As posed in Section 2.2.1, to what extent is the consequence of a child developing cancer more severe than for an elderly person? Attempts to resolve such a question often involve placing monetary values on damage caused to human health or the environment [507509], a contentious task that is not supported by Germany or Sweden (Section 4.2).
215
Description
Grouping according to prioritisation criteria proposed under the systems framework
Extent of damage
Adverse effects in units such as cancers and deaths
Hazard
Ubiquity
Defines the geographic dispersion of potential damages (intragenerational justice)
Exposure
Probability of occurrence
Estimate for the relative frequency of a discrete or continuous loss of function
Probability
Incertitude
Overall indicator for different uncertainty components (e.g., variability, random measurement errors)
Probability
Persistency
Defines the temporal extension of potential damages (intergenerational justice)
Probability
Reversibility
Describes the possibility to restore the situation to the state before the damage occurred (e.g., removing the source of exposure and then remediating the environment or treating an adverse effect)
Hazard exposure
Delay effect
Characterises a long time of latency between the initial event and the impact of damage
Hazard
Violation of equity
Describes the discrepancy between those who enjoy the benefits and those who bear the risks
Social mobilisation
Potential for mobilisation
Violation of individual, social or cultural interests and values generating social conflicts and psychological reactions by individuals or groups who feel themselves impacted by the risk VÃiµÕiViÃÆÊÌ
ÃÊVÕ`ÊÀiÃÕÌÊvÀÊ«iÀViÛi`ÊiµÕÌiÃÊÊÌ
iÊ distribution of risks and benefits
Social mobilisation
Table 5.6 Comparison of the criteria proposed by Klinke and Renn and the grouping of four fundamental categories for risk management under the systems framework
Framework for Chemical Risk Management under REACH
216
Risk level criteria developed by Klinke and Renn
A Systems Framework to Implement REACH Exposure and social mobilisation are stochastic, as are the corresponding risk level criteria used by Klinke and Renn shown in Table 5.6. Several probabilities are also associated with any hazard assessment. This thesis therefore proposes to distinguish between probabilities associated with hazard, exposure and social mobilisation. Rather than evaluate risks and the corresponding perceptions that establish risk levels, the thesis proposes to examine the extent of regulatory action necessary to ensure the correct management of a hazardous material and the likelihood that there is the need to take such a regulatory response. ‘Extent of regulatory action’ therefore describes the scope and degree of risk communication, control and enforcement that regulators could need to take to respond to risk and social perceptions of risk. First, the thesis proposes that “extent of regulatory action based on hazard” be given a rating according to a chemical’s toxicological profile. Another rating can then be assigned to the “probability of the need for a regulatory response based on hazard” according to a separate evaluation of probabilities and uncertainties in the hazard assessment. For instance, if uncertainties are high, there could be a particular regulatory concern even if an assessment indicates that the substance is not hazardous. In such a case, the hazard assessment could be incorrect. However, if uncertainty is low, there is probably less need for a regulatory response, even if a substance is hazardous as other regulatory frameworks would be triggered by the classification. For evaluating the extent of regulatory action based on exposure and social mobilisation, this thesis has devised a set of ‘most-probable’ events that can be evaluated in terms of their potential for being of general regulatory concern, which form indicators for “extent of regulatory action based on exposure” or “extent of regulatory action based on social mobilisation”. As a separate step, it is proposed that those events be evaluated in terms of the probability of exposure
217
Framework for Chemical Risk Management under REACH or social mobilisation being identified as a particular regulatory concern. This probability is therefore taken as representing the likelihood of regulators needing to take some form of regulatory response to reduce exposure levels or reduce the potential for social mobilisation, either at the national or EU level. As discussed above, regulators may need to take regulatory action as a result of: trade ÕÃÊV>}ÊvÀÊÌ
iÊÃÕLÃÌÌÕÌÊvÊ>ÊÃÕLÃÌ>ViÊÊÌ
iÊÜÀ«>ViÆÊ VÀi>Ã}Ê VViÌÀ>ÌÃÊ vÊ >Ê ÃÕLÃÌ>ViÊ Ê iÛÀiÌ>Ê i`>ÆÊ ÌÀ>ÃLÕ`>ÀÞÊ«ÕÌÆÊ>Ê>VÊvÊivÀViiÌÊvÊiÝÃÌ}Êi}Ã>ÌÊ in certain Member States34. As specified in the REACH legislative text, an event relating to exposure can be characterised as: a chemical having “wide dispersive use” (i.e., the chemical is commonly used in manufacturing processes and frequently contained in consumer products across EU Member States). While there is a probability associated with that specific event occurring, based on the best available information it is the possibility itself that determines the ranking in the category “extent of regulatory action based on exposure”. If a chemical is commonly used and contained in many consumer goods, there are potentially far more humans and ecosystems exposures to that chemical that need to be controlled than if the chemical use is restricted to a few isolated industrial sites35. Hence, this thesis proposes that the extent of regulatory action based on exposure for regulating wide dispersive use is relatively greater than for a few selected industrial point-sources. A separate indicator could be used to evaluate if the “probability of a regulatory response based on exposure” is of particular high 34
Descriptors used for carrying out qualitative evaluations of the probability associated with such events are presented in the following sub-section of this Chapter.
35
In turn, a larger number of exposed individuals and ecosystems would mean a higher probability that uncertainty factors used in a risk assessment do not sufficiently account for inter-species and intra-species variations. In this respect, there is a greater potential extent of damage that could occur as a result of the manufacture and use of that particular chemical.
218
A Systems Framework to Implement REACH «ÀÀÌÞÆÊÌ
ÃÊÜÕ`ÊVÃ`iÀÊÌ
iÊii`ÊÌÊVÌÀÊ«ÃÃLiÊ`>>}iÊ from exposures in terms of exposure levels, e.g., as a result of using personal protective equipment, applying pollution abatement technology or disposing of a product. If the chemical is being safely managed, one can anticipate low probabilities associated with a need for a regulatory response because exposures would be unlikely to be causing damage to any exposed humans and ecosystems (e.g., exposure would be below any relevant DNEL or PNEC). Under the exposure category, a distinction would thereby be made between (a) the probability of the need for regulatory action occurring in terms of exposure level and (b) the extent of the regulatory action in terms of the potentially exposed populations and ecosystems. It follows that two contrasting ‘most-probable’ scenarios could receive more or less equal ratings in terms of regulatory prioritisation based on the exposure category: (1) due to a large population being exposed to a chemical, the extent of regulatory action could be anticipated to be significant and therefore of high priority but, so long as safety measures are implemented, exposure levels are not anticipated to be of «>ÀÌVÕ>ÀÊÀi}Õ>ÌÀÞÊVViÀÆÊÀ (2) as a result of a small population being exposed to a chemical, the possible extent of regulatory action needed to control iÝ«ÃÕÀiÃÊÃÊÀ>Ìi`Ê>ÃÊLi}ÊÀ>Ì
iÀÊÜÆÊ
ÜiÛiÀ]ÊÌ
iÊ«ÀL>LÌÞÊ of exposure levels being high is anticipated to be of particular regulatory concern. Differentiating between the extent of regulatory action and the probability of a regulatory response being necessary can facilitate the selection of the most-appropriate type of regulatory instrument. For instance, in case (1) certain chemical uses could be banned to reduce the number of exposed individuals, whereas in case (2) product standards could be set to reduce exposure levels.
219
Framework for Chemical Risk Management under REACH For social mobilisation, it is proposed that the events used to assign relative rankings for the “extent of regulatory action based on the potential for social mobilisation” should be based on evidence that (i) the chemical is commonly detected in environmental media, organisms or humans and (ii) there are technically and economically viable substances that could replace the primary uses of the chemical. Together, these two scenarios are considered an indication that a large extent of damage could be caused to society by giving a strong basis for NGO campaigns that draw attention to a specific chemical or set of chemicals to gain media attention and support. As with exposure, the possibility of the above events occurring determines the ranking in the category ‘extent of regulatory action based on the potential for social mobilisation’. The “probability of a regulatory response based on social mobilisation occurring” would be evaluated separately, following an evaluation of factors such as a divide between populations benefiting from the use of the chemical and those exposed to the chemical. Detailed Proposed Method for Prioritisation To provide a methodology that can generate priority ratings for regulatory action based on hazard, exposure and social mobilisation for each chemical under regulatory review, this thesis has developed a set of qualitative ‘control indicators’. Each regulatory action indicator is considered to contribute to the overall extent of regulatory action needed to further assess and limit damage to human health, environment and society that can be caused by the manufacturing and use of a given chemical. In turn, the probabilities associated with each regulatory action indicator must then be assessed – i.e., the likelihood that a regulatory response will be necessary to reduce risk levels. Based on this combined rating, it is proposed that EU regulators prioritise decision-making under REACH.
220
A Systems Framework to Implement REACH 1. Regulatory action indicators (a) Hazard (toxicity) indicator Ranking
Description
High
Non-threshold or genotoxic
Medium
Threshold toxicity with some evidence of genotoxicity or reprotoxicity
Low
No evidence or very limited evidence of genotoxicity or reprotoxicity
(b) Exposure indicator Ranking
Description (scenario)
High
Dispersive use across the EU: used by many small companies, contained in professional and consumer products
Medium
Manufacturing generally limited to large sites with limited consumer product use
Low
Limited manufacturing use: used only as an intermediate or incorporated into a product matrix
(c) Social mobilisation indicator Ranking
Description (scenario)
High
Evidence of widespread environmental contamination or presence of the substance in human populations and evidence of available substitutes
Medium
Limited evidence of environmental contamination and presence of the substance in humans or limited availability of substitutes
Low
No evidence of significant environmental contamination
Table 5.7 Indicators of potential extent of regulatory action 221
Framework for Chemical Risk Management under REACH The three proposed indicators for extent of regulatory action are introduced in Table 5.7. Data from risk assessments, socio-economic analyses and other sources of information are used to ascribe a ‘high’, ‘medium’ or ‘low’ rating for each indicator to chemicals under regulatory review (Table 5.7). A ‘high’ hazard indicator rating corresponds to the need to control an adverse effect that is irreversible with a delay effect due to a genotoxic mechanism that affects future generations. A toxicological property that causes irreversible effects to human health or ecosystems and affects several generations is thereby considered as needing a greater extent of regulatory scrutiny and systematic control than a toxic effect from which a human, organism or ecosystem can easily recover without long-term impairment of its functions36 (refer to Section 2.1.2). When ranking the extent of regulatory action based on exposure, three scenarios have been separately developed to describe possible events (Table 5.7b and c). As with the hazard indicator a ‘high’, ‘medium’ or ‘low’ rating is assigned, but according to the most relevant scenario. For exposure, it is the use during manufacturing and its content in professional and consumer products that determines which scenario is most appropriate for describing the production, import and use of any given chemical. This is taken as a representative indication of the number of populations and ecosystems that are possibly exposed. The larger the number of possible exposed populations and ecosystems, e.g., as a result of wide dispersive and diffuse use, the higher the ranking in terms of potential extent of regulatory action necessary to protect human health and environment. When there are no monitoring data available on environmental contamination, a combination of high production or import volumes (i.e., above 1000 tonnes per year in the EU) and the intrinsic physicochemical properties of persistency and bioaccumulation provide the basis for rating for that part of the indicator for social mobilisation. In this scheme, these two physico-chemical characteristics of 36
Although a highly toxic (with an extremely low DNEL for acute toxicity) can cause the death of a human or an organism, which is an irreversible effect, hazardous properties that can affect multiple generations, such as a reprotoxicity, are given higher hazard rankings (refer to Section 2.1.2).
222
A Systems Framework to Implement REACH a chemical are evaluated separately for hazard37. Persistency, bioaccumulation, and volume are considered to be indicative of the possible extent of environmental pollution, and the presence of the chemical in humans. As will be detailed in a following section entitled ‘Overall Indicator Ratings’, the methodology presented in this thesis proposes to combine the indicator ratings for hazard, exposure and social mobilisation to rank the overall extent of additional regulatory action necessary to control the production and use of each chemical. The probability associated with each selected category of hazard, exposure and social mobilisation is also given a rating, based on different variables. The most relevant scenario for exposure or social mobilisation will respectively correspond to the one associated with the highest probability that exposure or social mobilisation triggering a regulatory response. The variables used for determining the rating of the exposure indicator differ from those used to carry out the subsequent evaluation of probability of that scenario occurring. Moreover, as with the ranking scheme for the regulatory action indicators, probability ratings are relative. Two chemicals that have had the same scenario selected as ‘most relevant’ – i.e., the same exposure indicator rating – may therefore have a different rating according to the probability of the need for a regulatory response occurring. 2. Probability Indicators So far, the method for prioritisation has detailed only how to yield >Ê`V>ÌÛiÊÀ>Ì}ÊvÀÊÌ
iÊiÝÌiÌÊÀi}Õ>ÌÀÞÊ>VÌÊÌ
>ÌÊVÕ`ÊVVÕÀÆÊ the probability that a regulatory response is necessary (or is perhaps already occurring) must be evaluated. The approach proposed here would rate separately the probabilities associated with each regulatory action indicator (hazard, exposure and social mobilisation) as ‘high’, ‘medium’ or ‘low’ and then combine them to give a probability that can be associated with the overall extent of regulatory action occurring. 37
In other words, the hazard indicator only refers to ‘toxicological and ecotoxicological profiles’ that exclude the physico-chemical, biological and environmental fate properties of persistency and bioaccumulation.
223
Framework for Chemical Risk Management under REACH To compensate for potential errors in company or regulatory assessments, uncertainty in the analysis of the probabilities associated with hazard, exposure or social mobilisation that has not been incorporated into the regulatory action indicators. Instead, it is assumed to increase the probability that a need for regulatory response will occur. For example, if an exposure assessment concludes that risk management measures are in place but some contradictory evidence indicates measures are not often applied, the probability of needing a regulator response to further control exposure will be given a ‘high’ or ‘medium’ rather than a ‘low’ rating. An overview of the probabilities associated with each regulatory action indicator is provided below. Probabilities associated with Hazard Uncertainty relating to hazard mostly arises from limited toxicological data and is therefore dependent on the number of tests conducted, and sample sizes. A high uncertainty that the substance is hazardous could arise if there is a high level of hazard data for a substance or if there are significant gaps in available hazard data. In turn, this is considered to influence the probability that there is a need for regulatory action under REACH. Probabilities associated with Exposure Analyses of the probability that exposure to a substance will occur largely depend on the existence and implementation of management controls. For instance, there is a greater likelihood that a direct exposure will occur for untrained or unsupervised use of a chemical, for instance because incorrect personal protective measures are being used. High probabilities of exposure can also be considered for chemicals that are not regulated under existing legislation, for instance because they are not classified as dangerous or because the concentration of the chemical in products is not known. Probabilities associated with Social Mobilisation Based on the analysis of risk management in France, Germany, Sweden and the UK, the probability that regulators need to take
224
A Systems Framework to Implement REACH action as a result of societal responses to risk primarily depends on if (i) the chemical provides a direct benefit to the exposed population and (ii) pregnant workers are exposed to the chemical38. In terms of assessing the potential benefits associated with a chemical, this could include revenue for the exposed workers from manufacturing a product or the product performance that a substance delivers to a consumer (e.g., a hair dye or car wax). The probability that a regulatory response is needed to control trans-boundary pollution would be a clear example of a potential division between sub-populations that benefit from the manufacture of a chemical and those that are exposed to the chemical. 3. Overall indicator ratings Ratings action indicator or probability indicator
Overall rating
Hazard
Exposure
Social mobilisation
(Combination)
Low
Low
Low
Low
Low
Low
Medium
Low
Low
Low
High
Medium
Low
Medium
Medium
Medium
Low
Medium
High
Medium
Low
High
High
High
Medium
Medium
Medium
Medium
Medium
Medium
High
High
Medium
High
High
High
High
High
High
High
Table 5.8 Example of rating combination rules 38
This is considered to indicate that manufacture and use of a substance has a fairly high probability of triggering regulatory and public attention to potential chronic effects on human health, in particular CMR and endocrine properties. Exposure to pregnant women and children in the general public is considered in the regulatory control indicator for social mobilisation.
225
Framework for Chemical Risk Management under REACH Qualitative rules are used for determining an overall rating for extent of regulatory action and an overall rating for the probability associated with that extent of action being necessary. The qualitative ratings for the three action indicators or the indicators for the associated probabilities are separately combined using the rules exemplified in Table 5.8. Because each regulatory action indicator describes an event (e.g., a degree of regulatory action based on a hazard, an exposure at a given level, and social mobilisation) that is defined according to independent and unique variables, then the probability of the overall extent of regulatory action occurring can be seen as dependent on the probabilities associated with each regulatory action indicator. In statistical analysis, this is referred to as conditional probability which forms the basis of Bayes’ theorem [94]. The probability of a regulatory response occurring at the level established by the summation of the regulatory action indicator ratings is therefore be considered to be ‘high’ if there is a high probability associated with each of the indicators for hazard, exposure and social mobility. Conversely, the probability of a regulatory response occurring at the level established by the combined rating of the regulatory action indicators can be considered as ‘low’ if there are low probabilities associated with each indicator. If the rules in Table 5.7 were to be expressed in numerical terms, >ÊÃVÀiÊÃÊ>ÃÃ}i`ÊÌÊi>V
Ê`V>ÌÀÊÜÊrÊ£ÆÊi`ÕÊrÊÓÆÊ
}
ÊrÊ 3) and the overall rating is taken as a linear combination (i.e., add them). The nominal overall rating is then given by the total score ÜÊrÊÎÊÀÊ{ÆÊi`ÕÊxÊÀÊÈÆÊ
}
ÊrÊÇÊÀÊÀi®°Ê/
ÃÊÃÊVÃÃÌiÌÊ with Bayes’ theorem. Different weighting schemes and rules for combining ratings could be applied by decision-makers using methods such as multi-criterion analysis39ÆÊvÀÊÃÌ>Vi]ÊÌ
iÊhazard indicator could be considered as being more significant in determining the extent of regulatory action 39
Multi-criterion analysis is a decision-making tool for qualitative weighting of aggregated impact assessment indicators. Further information is available in the OECD Technical Guidance Document on the use of Socio-economic Analysis in Chemical Risk Management Decision-Making [510].
226
A Systems Framework to Implement REACH needed to control the manufacture or use of a chemical than the exposure indicator, or vice versa. For the purposes of this thesis, which does not focus on prioritisation methods, alternative weighting schemes are not examined. Presentation and Interpretation of Results The method of qualitative rankings developed in this thesis is rather similar to the system for prioritisation developed by [12]. Klinke and Renn propose that regulatory action should be prioritised according to the following six questions: 1. Does the risk exceed pre-specified thresholds of one or more of the criteria? 2. Is the damage potential known and can it be identified? 3. Does the damage potential exceed pre-defined thresholds for catastrophic potential? 4. Does the risk show no significantly high values on any physical criteria but even so rate highly on social criteria? 5. Are the criteria for persistency, ubiquity and irreversibility ‘high’? 6. Is probability ‘high’ or ‘uncertain’? Answering these six questions depends on an evaluation of a risk according to the nine criteria identified by Klinke and Renn and shown in Table 5.5. In this respect, the method developed in this thesis is simpler because it only depends on an evaluation of the three fundamental criteria of hazard, exposure, social mobilisation, as well as the probabilities asscociated with each criterion. The ranking system proposed in this thesis essentially corresponds to establishing the thresholds, including the relevant ‘high’ ratings, referred to in the above questions and assessing whether a risk exceeds any of the thresholds. Although the method proposed by Klinke and Renn requires such an evaluative process, it does not detail a structure for setting thresholds or assessing whether a risk exceeds a given threshold. 227
Probability of the need for a regulatory response
Framework for Chemical Risk Management under REACH
Low action indicator High probability indicator
Medium action High probability
High action High probability
Low action indicator Medium probability indicator
Medium action Medium probability
High action Medium probability
Low action indicator Low probability indicator
Medium action Low probability
High action Low probability
Extent of regulatory action
Figure 5.15 Nine regions on a plot of probability of response versus extent of control This thesis proposes to present the results of risk assessments and related analyses on a plot of extent of regulatory action versus probability of the need for a regulatory response. Using the possible combinations of the ratings for “extent of action” and “probability of response” from Table 5.8, this thesis proposes to differentiate between chemicals by assigning each chemical to one of nine regions shown in Figure 5.16. The method proposed in this thesis produces a comparative indication of the extent of action that could be needed to regulate a chemical and the probability of this regulatory response being necessary based Ê>Û>>LiÊÀÃÊ>ÃÃiÃÃiÌÊ`>Ì>ÆÊÌÊÃÊÌÊÌi`i`ÊvÀÊV>ÀÀÞ}ÊÕÌÊ absolute risk scaling. If a substance does not have a hazard classification, then a review of available risk assessment data must be used to establish which rating is most appropriate. Any uncertainty associated with the data used to rate each indicator would be incorporated as a ‘worst-case’ scenario when evaluating the extent of regulatory control or factored into a subsequent analysis of probability, as discussed above. The method for prioritisation has been devised for EU decision>}ÆÊÊÌiÀÃÊvÊiÝiVÕÌ}Ê`iVÃÃ]Ê>ÊÜ`iÊÕLiÀÊvÊ>VÌÀÃÊV>Ê often be expected to be involved. Regulatory action covers a broad
228
A Systems Framework to Implement REACH range of possible regulatory responses, including risk communication. Therefore the ‘extent’ of regulatory action could in fact be substantial – e.g., a ban – but only require minimum resources for local regulatory implementation in terms of communication and enforcement. The ban of a substance could prove costly to some companies requiring technical assistance that is best provided at a regional level. At the same time, a high-priority substance could warrant EU regulators to seek agreement on a ban at the international level. Prioritisation of regulatory action has implications to resource allocation at local, regional and EU levels. As with the REACH Regulations, enforcement and resource allocation remains the responsibility of Member States. In some cases, a lower priority substance under the systems framework could present a high priority at a local level. The decision-making matrix includes consideration of resource when selecting the most appropriate regulatory option and the Monitoring Network provides a mechanism to input this information to decision-making under REACH. Hypothetical Example A hypothetical example of the application of the methodology follows in Table 5.9. The relevant data from chemical risk assessments and socio-economic analyses are entered into three columns corresponding to hazard, exposure and social mobilisation. For each of the two substances, the first row characterises the relevant scenarios according to the regulatory action indicators from Table 5.7. The second row, which is shaded in grey, then describes the key parameters that influence the probabilities associated with each action indicator. The corresponding ratings for the action indicators and the probability of occurrence indicators are presented in bold italics. The purpose of the example is to illustrate the level of information necessary to perform the ratings. Compared with a risk assessment of several hundred pages that describes and interprets numerous scientific studies and other data, the level of information presented in Table 5.8 is very concise. Nevertheless, a complete or near-complete ÀÃÊ>ÃÃiÃÃiÌÊÃÊiViÃÃ>ÀÞÊvÀÊÌ
iÊiÝiÀVÃiÆÊÌ
ÃÊÃÊ>ÌV«>Ìi`ÊÌÊ be available under REACH following registration and evaluation.
229
Framework for Chemical Risk Management under REACH Using the rules for combining ratings shown in Table 5.7, the levels of prioritisation for regulating the manufacture and use of substance A and substance B can be described as: Substance A Indicator rating for extent of regulatory action = medium (hazard indicatorÊ rÊ i`ÕÆÊ exposure indicatorÊ rÊ i`ÕÆÊ social mobilisation indicator = medium) Probability indicator for the need of the regulatory response = medium (probability rating associated with hazardÊrÊ
}
ÆÊprobability rating associated with exposureÊ rÊ ÜÆÊ probability rating associated with social mobilisation = low) Substance B Indicator rating for extent of regulatory action = high (hazard indicatorÊrÊ
}
ÆÊexposure indicatorÊrÊÜÆÊsocial mobilisation indicator = high) Probability indicator for the need of the regulatory response = medium (probability rating associated with hazardÊrÊ
}
ÆÊprobability rating associated with exposureÊ rÊ ÜÆÊ probability rating associated with social mobilisation = medium) Substance B would be prioritised over substance A for regulatory action, even though both substances are likely to be hazardous and current exposures to substance B are more likely to be of regulatory concern than exposures to substance A. When comparing the two substances, it is the potentially highly hazardous properties of substance B and the potential for social mobilisation that make it a higher regulatory priority than A. Under REACH, it is anticipated that the level of data generated through registration and evaluation will be sufficient to perform such an analysis. The operability of this prioritisation scheme is tested in Chapter 6 for the same 33 chemicals used to test the decision-making matrix.
230
Hazard Regulatory action indicator
Exposure Medium
Medium
threshold carcinogen and ÀÀÌ>ÌÆ
commonly used as a solvent for cleaning industrial equipment in >Ê>Õv>VÌÕÀ}ÊÃiVÌÀÃÆ
evidence that the chemical can be commonly detected in environmental media, organisms ÀÊ
Õ>ÃÆÊÌÊ«iÀÃÃÌiÌÊÀÊ L>VVÕÕ>ÌÛiÆÊÃÕLÃÌ>ViÊ
>ÃÊ low substitute availability
Substance A
not contained in consumer products or professional products
High
Medium
Low
considerable uncertainty relating to the hazard assessment data and limited hazard data available – therefore, under a worstcase scenario, there is a high probability that the substance is highly hazardous
high probability that workers handling the substance will be frequently exposed to the ÃÕLÃÌ>ViÆ
due to the location of production sites trans-boundary pollution is ÕiÞÊÌÊVVÕÀÆ
measures applied to control iÃÃÃÊ`ÕÀ}ÊÕÃiÆ
231
there are a few unknown exposure scenarios, especially because the substance could be contained in professional products
pregnant workers are not allowed to handle the substance in the workplace and restrictions apply to professional and consumer «À`ÕVÌÃÆÊ the substance provides a particular direct function to the user or members of the public
A Systems Framework to Implement REACH
Medium
some evidence genotoxicity
Probability of regulatory response indicator
Social mobilisation
High
Low
High
ÊÌ
ÀiÃ
`ÊV>ÀV}iÆ
ÃÌÞÊÕÃi`Ê>ÃÊ>ÊÌiÀi`>ÌiÆ
mutagen
open-batch use limited to a few ÃÌiÃÆ
persistent and bioaccumulative properties but there are no relevant ÌÀ}Ê`>Ì>Æ
Substance B
not contained in professional or consumer products Probability of regulatory response indicator
}
Ê«À`ÕVÌÊ>`Ê«ÀÌÊÛÕiÃÆ technical reports indicate that several substitutes exist
High
Low
Medium
in-depth risk assessment data available resulting in few uncertainties in the hazard assessment – therefore there is a high probability that the substance is highly hazardous as indicated above
personal protective equipment is likely to be used to limit the «ÌiÌ>ÊvÀÊÀÀÌ>ÌÊivviVÌÃÆ
evidence of trans-boundary pollution VVÕÀÀ}Æ pregnant workers are unlikely to be iÝ«Ãi`ÊÌÊÌ
iÊÃÕLÃÌ>ViÆ
exposure levels and scenarios are known and emissions are likely to workers benefit from use via wealth be controlled – therefore there is not creation a high level of uncertainty involved in the exposure assessments and the probability indicator relating to exposure is low
Table 5.9 Hypothetical example of the priority-rating scheme used to compare two substances - clear cells correspond to regulatory action indicators and shaded cells to probability indicators - ratings are shown in bold italics
Framework for Chemical Risk Management under REACH
232
Regulatory action indicator
A Systems Framework to Implement REACH
5.5 Conclusions Previous risk management decisions can serve as a basis for future regulatory procedures. The current slow and resource-intensive substance-by-substance approach may thereby be avoided. By presenting risk information and decision-making structures in a clear and consistent format, the systems framework aims to create an organised structure for exchanging data and identifying safety concerns between relevant stakeholders and regulatory administrations. The proposed set of regulatory recommendations should promote stable and predictable decision-making processes. Establishing safe and permissible uses seeks to reduce the administrative burdens that businesses and regulators anticipate under REACH. Information requirements for registration and evaluation would be focussed on all aspects of consumer risk assessment for listed uses, and regulatory monitoring and enforcement of products on the EU market would be prioritised accordingly. Listing chemicals used in certain consumer products would also extend responsibility for conducting product monitoring and carrying out relevant risk communication activities onto retailers, and consumer and environmental NGO. Finally, any chemical use requiring regulatory risk reduction would be categorised as tolerable, restricted, or authorised. This would avoid decisions under REACH being dependent on legislative frameworks outside its immediate scope. Political aspects have been factored into the criteria proposed to evaluate chemical risks and subsequent regulatory risk management action. The research findings indicate that Member State-permitting schemes should serve as a mechanism to control occupational exposures in industrial settings that are subject to political scrutiny. Similarly, existing national approval, certification or licensing schemes should be factored into decision-making under REACH, even though their implementation depends on regulatory competences and procedures outside the scope of REACH. The procedure for the target-setting of tolerable uses would provide a much needed mechanism that provides flexibility for Member States to achieve
233
Framework for Chemical Risk Management under REACH independent policy objectives in areas that influence decision-making on restriction and authorisations under REACH. In terms of administration under REACH, the research has identified a lack of formal structure for incorporating stakeholder inputs into Registration and Evaluation. Risk assessment and management data from Stakeholder Inputs would be subject to regulatory peerreview and incorporated into decision-making through Member State regulators and relevant EU stakeholder organisations. A Monitoring Network comprising of Member State regulators and officially created EU agencies, networks and foundations could also be responsible for co-ordinating activities across the DirectoratesGeneral of the European Commission. Two further roles of the network would be to: (1) review the decision-making rulesÆÊ >`]Ê (2) promote the harmonisation of administrative procedures for companies demonstrating compliance across REACH and other EU legislative frameworks. The European Chemicals Agency could learn from the structures and experiences of the Member State regulatory administrations reported in Chapter 4. First, careful consideration must be given to the availability and prioritising of Member State resources necessary for regulatory monitoring and enforcement of registrations, authorisations, restriction, emission limit values, product standards, environmental quality standards, and SDS. For a balanced approach to risk management, consideration of risks outside the scope of the European Chemicals Agency must also be considered. For instance, substituting small quantities of hazardous substance by larger quantities of a less hazardous substance for equivalent process performance can result in higher ergonomic risks for workers manually transporting the chemical (a problem recently encountered in Sweden). While the systems framework contains structural elements of the chemical strategies that have been proposed by the Dutch VROM and the UK RCEP, it focusses on decision-making under REACH. In comparison, the RCEP specifies action relating only to phase-out of
234
A Systems Framework to Implement REACH certain substances and does not examine decision-making processes at the EU level. The Dutch process establishes a detailed matrix that combines hazard criteria with industrial, professional and consumer uses, but does not detail specific guidelines for decision-making on risk management controls or wider policy instruments. Future development of the systems framework could therefore focus on further integrating the relevant elements of the Dutch and RCEP proposals into REACH. This could be achieved by refining the criteria for the categorisation of chemical uses and the corresponding decision-making rules. In conclusion, the research stresses that regulatory decision-making criteria should be established prior to implementing REACH. Although the proposed systems framework for decision-making has been constructed so that it can be integrated into REACH without changes to the current legislative text, experience with EU chemical risk management indicates that changes to decision-making structures will prove difficult to achieve without political attention. The development of the TGD and the internal functioning of the European Chemicals Agency should therefore be carefully monitored by all Member State regulators and stakeholder groups. An approach has been developed to screen chemicals for prioritisation for risk management, based on their likely damage to human health, the environment and society. This approach is developed further, with illustrative examples, in the next chapter.
235
6
Evaluating the Systems Framework
- Outcome …the wisdom of an act… judged by the light that the doer had when he performed it (Ambrose Bierce, The Cynic’s Word Book, 1906)
Introduction Risk management under REACH must be transparent, practical and predictable. Regulators, companies and stakeholder organisations need a set of rules and structures to implement a highly complex piece of regulation. Decision-making will need sufficient flexibility to incorporate Member State approaches and allow companies to search for innovative solutions to risk management. A framework for decision-making under REACH must therefore grant regulators and the relevant actors the opportunity to use various control instruments to achieve European Union (EU) standards and objectives. At the same time, as defined under the European Commission Treaty, regulators must retain the ability to go beyond certain EU standards. The challenge consists of achieving this without creating an overbureaucratic or unsystematic process. ‘Bottom-up’ regulatory approaches need to be integrated into REACH. Not only must this involve collecting and reviewing data necessary to carry out risk assessments, it will require identifying, communicating and promoting safety management best practices. Enhancing the competitiveness of companies with the highest
237
Framework for Chemical Risk Management under REACH standards of safety management should catalyse a move towards knowledge-based sales in the EU industry. In this respect, REACH presents a once-in-a-lifetime regulatory opportunity to reform chemicals policy. Achieving such a business environment depends on creating a level playing field among companies across the EU and ensuring compliance of imports. This Chapter demonstrates how the key elements of the systems framework could meet these objectives. To begin, Section 6.1 describes the ability of the proposed framework to achieve the two most fundamental aspects of chemical control: identifying risks and promoting chemical safety. The operability of the systems framework is then tested using 33 chemicals (Section 6.2). The result of this assessment is presented in Section 6.3 together with a discussion on the outcome of testing the scheme for prioritising regulatory activities. The Chapter concludes by reflecting on how decisions may be reached among the 25 European Member States (Section 6.4 and 6.5).
6.1 From Risk to Safety: Review of the Systems Framework Input: Risk
Actors: Roles, responsibilities, relations, resources
Process: Risk management decision-making & implementation
System constraints: Legal, scientific economic social
Outpur: Safety
Figure 6.1 Overview of the systems framework
238
Evaluating the Systems Framework Input and Actors The ‘soft systems’ approach described in the Methodology provides a general overview of how the proposed framework would operate (Figure 6.1). As a first stage in implementing REACH, the systems framework would ensure a regulatory focus on the input of existing chemical risk data from a wide set of stakeholder organisations. As described in Section 5.4, compared with the current REACH regulation, the Stakeholders Inputs would provide a mechanism for incorporating data to supplement the registration dossiers submitted by industry. In parallel, the systems framework proposes the launch of several initiatives to collect data on the chemical contents of certain categories of consumer products (listed uses). This should support the generation of information on the ‘use phase’ of chemicals, thereby helping regulators and companies to overcome a major hurdle in the risk assessment process. Involving stakeholder associations at these early stages of implementing REACH is also expected to facilitate the identification of chemicals and exposure scenarios of potential public concern. Output Communicating risk information through supply chains forms the basis of promoting chemical safety. While REACH places requirements on producers and importers to conduct and communicate hazard-based risk assessments, it does not contain decision-making rules detailing risk management duties for chemical users. Some form of regulatory recommendations will need to be incorporated into future technical guidance documents / ®ÆÊ Ì
iÀÜÃi]Ê Ì
iÊ Ài}Õ>ÌÀÞÊ ÕÌViÃÊ vÊ , Ê ÜÊ ÌÊ be predictable for companies supplying or using chemicals. This research project has identified that generating information to supplement registration data on listed uses and making it publicly available should assist companies in meeting any recommendations. The mechanism for listed uses would also enable companies and stakeholder organisations to deal directly with a central European Chemicals Agency, thereby minimising the need to exchange
239
Framework for Chemical Risk Management under REACH information up-and-down international supply chains simply for the purposes of submitting upstream registration dossiers. Together with the listed uses, publishing substances subject to target-setting on the REACH-IT website is anticipated to assist regulators, companies and stakeholder organisations to prioritise and coordinate risk management activities. Evidence presented in Chapter 5 indicates that identifying and tracking regulatory activities will otherwise continue to be difficult for chemicals that are not subject to restriction or authorisation. Several large EU retailers and manufacturing organisations have expressed their interest in gathering information on consumer products [78, 511]. One representative of the UK Retailers Consortium has even gone so far as to state that the chemical content of certain goods must be made available to maintain consumer trust in regulators and industry [512]. While such data available on a publicly accessible database can support corporate risk management and ensure safe products, one would expect it to catalyse the development of voluntary initiatives already adopted by a number of actors, such as the phase-out of very persistent and very bioaccumulative (VPVB) substances. This would then create market incentives for the upstream substitution of hazardous substances in products [78, 511]. Data would also serve to facilitate the activities of non-governmental organisations (NGO) involved in product testing such as the Öko-test in Germany or the Swedish Society for Nature Conservation’s eco-labelling in Sweden (Section 4.4.1). The combined effort of these relative ‘newcomers’ – i.e., consumer NGO and retailers – to chemical risk management could potentially provide a rigorous support network to regulatory product compliance and enforcement schemes. Two recent case-examples shown in Box 6.1 illustrate the potential effectiveness of chemical risk data gathering and safety communication programmes:
240
Evaluating the Systems Framework Box 6.1 Successful risk data generation and safety communication schemes Industrial clothing (see [513]) A group of members of the Association of the Netherlands Textile Industry investigated the potential risks posed by the use of several chemicals used in the production of overalls. Monitoring employee exposures identified areas where additional risk management measures were required. Consultation with upstream producers and a large downstream customer also enabled the companies to improve final product quality. Detergents (see [172]) A collaborative study on the Human and Environmental Risk Assessment (HERA) of cleaning products between Cefic and the International Soap, Detergent and Maintenance Products Association (AISE) resulted in the risk assessment of several substances through the relevant supply chains. The project enabled downstream users to develop a better understanding of the final use and environmental endpoints of these products. A major result of the initiative has been the creation of a website where customers and consumers can be informed on the risks associated with specific substances.
Based on these experiences, it is anticipated that several similar initiatives could be successfully launched at national and international levels. The first example in Box 6.1 originates from schemes implemented under the Dutch Strategy on Management of Substances (SOMS). Rather than developing the decision-making rules originally proposed in SOMS (Section 5.2), recent development of the Dutch chemicals policy has emphasised generating, sharing and communicating existing risk and safety data [317]. Categorising chemicals according to risk levels, which was originally part of the Dutch decision-making rules, provided an important basis for organising and prioritising the relevant stakeholder activities
241
Framework for Chemical Risk Management under REACH during the implementation [514]. The second example in Box 6.1 is the result of a voluntary industry initiative triggered by a public demand for transparency in risk assessment processes. Recent case trials of registration further support the need for greater involvement of downstream users in REACH implementation than mandated in recent legislative drafts [315, 462] . These studies confirm that such measures would distribute responsibility for generating and communicating risk information. Decision-making Process A clear message from the interviewees was the need to streamline EU regulatory risk management and render it predictable. The identification of safe and permissible uses would limit the number of substances subject to various stages of REACH. Surprisingly, no such concept appears in the REACH regulation or preliminary TGD1 even though many interviewees expressed concern that the regulatory system could become overloaded. The decision-making matrix under the systems framework proposes to create the first EU mechanism that regulators and stakeholders could use to predict regulatory outcomes. It follows that the matrix would also provide a method to review previous decisions, a fundamental aspect of regulation that escapes current EU decision-making. The Monitoring Network would have a crucial role in developing risk management measures and communicating the outcome of regulatory decisions. Fundamentally, a co-ordinated approach to regulation which provides information on implementation timelines, monitors the outcomes and harmonises compliance reporting would facilitate corporate and regulatory risk management activities across the EU. Recent discussions on the REACH legislative proposal have highlighted the need to prioritise regulatory activities [168]. Following the systems framework, regulatory decision-making and 1
As of April 2006, the preliminary REACH technical guidance documents (TGD) available to date relate to the risk assessment processes of REACH, such as the use of toxicological QSARs and the development of exposure scenarios [126].
242
Evaluating the Systems Framework risk management activities should focus on VPVB substances, an agreed political objective from the original chemicals policy. The system framework’s prioritisation method includes such chemicals under restriction and target-setting. By contrast, the legislative proposal does not provide any specific method to prioritise these substances other than during evaluation or their possible inclusion under authorisation. When determining risk levels, the framework would also include ‘future’ risks based on rising concentrations in environmental media and would factor in the potential for societal concerns2, aspects not considered in the legislation. System Constraints A major constraint of risk management arises from the changing nature of scientific knowledge. For this reason, companies should be granted considerable flexibility for deciding on how to meet regulatory demands, as offered under the recommendations. Regulators should retain the ability to review and adjust decision-making rules and targets according to scientific developments or new information on risks. Under the decision-making matrix measures to control occupational exposure at national levels would continue to operate under legislation based on Article 138 of the European Commission Treaty and most environmental emissions would be controlled through legislative frameworks based on Article 175 (see Sections 2.4.3 and 5.3.1). When effective enforcement mandates the control of specific products or processes at the EU level, restriction and authorisation would apply following Article 95 on the internal market. The criteria for the recommendations proposed under the systems framework are based on scientific evidence and previous regulatory decisions that have been subject to rigorous EU legislative processes. It is therefore anticipated that these regulatory guidelines should also conform to international trade laws. Because the decision2
For example, the systems framework prioritisation accounts for potential exposure to susceptible populations and the ability of an exposed population to control the risk.
243
Framework for Chemical Risk Management under REACH making rules and recommendations would apply equally to EU and imported products, the systems framework should be compatible with World Trade organisation (WTO) agreements on technical barriers to trade. To maintain the international competitiveness of the EU chemical industry, some form of recommendations must be applied immediately to consumer and professional products, otherwise regulators will have to wait for phase-in periods over the 11-year REACH implementation before beginning to regulate products already on the EU market. In turn, the appropriate recommendations could create a level playing field for companies producing registered substances and those whose products have yet to undergo regulatory review.
6.2 Testing the Systems Framework 6.2.1 Regulatory Outcomes To test the systems framework, official EU risk-reduction strategies (RRSs) for 28 substances and five chemicals currently under regulatory review have been surveyed (see Section 3.7). Following the format of the technical guidance proposed in Section 5.3.2, risks have been categorised according to if exposures to the given substance or group of substances pose risks to human health or the environment (step 1a). Any significant uncertainties in EU risk assessments have been identified (step 1b), as well as chemicals involving particular political dimensions at the national level (refer to Box 5.3 in Chapter 5). Automatic bans for carcinogenic, mutagenic or reprotoxic (CMR) substances following the restrictions procedure under Directive 76/769 have been excluded (step 2) because these will continue to apply under REACH. The result of the risk criteria evaluation for each substance according to the decision-making matrix is shown in Table 6.1 (step 3). Based on the selection procedure detailed in Section 5.3.2, the most suitable regulatory outcome
244
Evaluating the Systems Framework has been selected. An overview of the results of this process is explained in Box 6.3 Testing the systems framework has been limited to chemicals that have undergone risk assessments under Regulation 793/93. The data sets for these substances can be anticipated as being comparable to what will be available at the outcome of the evaluation process under REACH. Because these are high production volume chemicals that have previously been identified as posing high-level risks by regulators, restriction has been considered to be necessary to supplement the systems framework recommendations. This may not be the case for many chemicals presenting lower risk levels during the actual implementation of REACH. Furthermore, whether supplementary bans under restriction would be necessary would depend on: UÊ prioritisation of regulatory decision-making (see next Section), and/or UÊ iÛ`iViÊvÀÊStakeholder Inputs or the Monitoring Network indicating that the recommendations are not being met. A key point of the framework is that the recommendations would apply before any decisions on restriction may be reached, thereby minimising ongoing risks resulting from delays in decision-making. The relevance of recommendation equivalents is illustrated by three substances (MAA, MMA and MTBE) for which the official EU RRS requires the development and dissemination of risk management best practice guidelines. Under the systems framework, companies would be given the option of adopting such practices to meet the regulatory recommendations. During implementation, regulators could check whether companies are carrying out this task. In cases of non-compliance, restrictions would need to be enacted. In this way, the systems framework removes one step from current EU decision-making.
3
A more detailed description of the application of the methodology is available in the original version of this thesis, as published in 2007.
245
Framework for Chemical Risk Management under REACH
Risk criteria
Immediate
Future
Cons.
Profess.
National dimensions
Regulatory outcome
Specific
Environment* General
Chemical
1
penta-BDE
i
iii
i
i
ii
ii
No
EU authorisation
2,3
octaanddeca-BDE
i
i
i
i
ii
ii
No
EU authorisation
4
TCB
iii iii iii
iii
iii
iii
No
EU authorisation
5
TCE
iii
i
ii
i
Yes
MS permitting
ii
ii
6,7
ÆÊ
ii
ii
ii
ii
iii
iii
Yes
MS permitÌ}ÆÊÀiÃÌÀVtions for cons and prof
8,9
ÆÊ2O2
ii
iii
ii
ii
iii
iii
No
Restrictions for cons and prof
iii iii iii
iii
ii
ii
No
Restrictions for industrial uses
10,11
*ÆÊ *
12,13
ÆÊ>VÀÞ>`i
ii
iii
ii
ii
ii
iii
No
Restrictions for prof
14,15
-ÆÊ£]{`Ýane
ii
ii
ii
ii
ii
iii
No
Restrictions for prof
16
Phthalate subgroup: based on BBP, DBP, DEHP, DIDP, DINP, DNOP
i
i
i
i
i
ii
Yes
Restrictions for cons
17
toluene
iii
ii
ii
ii
iii
iii
No
Restrictions vÀÊVÃÆÊ targets
246
Evaluating the Systems Framework
18
SCCP
i
iii i
iii
ii
i
Yes
Restrictions vÀÊ`ÕÃÌÀ>ÆÊ targets
19
o-anisidine
i
ii
ii
i
iii
ii
No
Restrictions vÀÊVÃÆÊ«Ãsible) targets
20
MCCP
i
i
i
i
ii
i
Yes
Targets
21
butadiene
i
ii
ii
i
ii
ii
No
Targets
22
MTBE
iii iii iii iii
ii
ii
No
Targets
23,24
ÆÊ
ii
iii ii
ii
ii
iii
No
No EU action
25,26, >VÀÞ>`i
Þ`iÆÊ 27,28, ÆÊ ÆÊ{
ÆÊ 29,30 ÆÊVÕii
ii
ii
ii
ii
ii
ii
No
No EU action
31
acetonitrile
ii
iii ii
ii
ii
ii
No
No EU action
32
acrylonitrile
i
iii ii
ii
ii
ii
No
No EU action
33
bisphenol A
i
i
ii
i
i
No
Further information
i
9\ÊVÃÊrÊVÃÕiÀÊ«À`ÕVÌÃÆÊ«ÀvÊrÊ«ÀviÃÃ>Ê«À`ÕVÌà (i) = there is a need for need for further information and/or testing, (ii) = there is at present no need for further information and/or testing and no need for risk reduction measures beyond those already being applied, (iii) = there is a need to limit the risks.
Table 6.1 Result of regulatory outcomes following the systems framework - chemical acronyms listed in Abbreviations
Box 6.3 Systems framework regulatory outcomes for 33 chemicals EU authorisations (four substances) The risks from penta, octa-BDE and TCB mandate EU action because these chemicals have been detected in ecosystems, their concentration in environmental media across Member States may be increasing, and scientific evidence indicates that their hazardous properties could
247
Framework for Chemical Risk Management under REACH affect human health or ecosystems at low levels. Due to deca-BDE’s similarity to penta and octa-BDE, and the political controversy across the EU countries concerning the risks of brominated flame-retardants, it would also be subject to the EU authorisation process even though the substance risk assessment requires further hazard data on the environmental break-down products. Evidence indicates that these some of these products may be bioaccumulative and toxic. Following the systems framework, EU authorisations have been identified as the most suitable control mechanism because they would enable immediate action to control all uses through one risk management mechanism. Due to high uncertainty sometimes involved with evaluating risks (e.g., deca-BDE), authorisation of a substance or its specific uses could be reviewed following the generation of new risk data. The authorisation system would therefore enable the identification of all intended uses, which would facilitate detailed exposure assessments and enable the rapid implementation of any risk management measures subsequently deemed necessary. For TCB, this is important because of the large number of facilities using the product, especially as an intermediate, and its potential for causing trans-boundary pollution. Making these substances subject to the EU authorisation process may provide a general disincentive for use. Carrying out socio-economic analyses for these substances should be fairly straightforward because substitutes are available for brominated flame-retardants and most high-level risks from TCB result from specific industrial processes rather than products. Member state permitting (one substance) with restrictions (two substances) Interview responses indicated that DEGBE, DEGME (two glycol ethers) and TCE necessitate stringent regulatory responses in certain, but not all, Member States. These three chemicals would therefore be subject to Member State permitting procedures for controls on occupational settings, supplemented by restrictions for certain professional and consumer uses of DEGBE and DEGME. Although implementing the Member State permitting procedures would fall
248
Evaluating the Systems Framework outside the immediate scope of REACH, the research reveals that it is important to incorporate these mechanisms into decision-making. Restrictions (nine chemicals) Following evaluations under REACH, the nine chemicals listed below would fall directly under the REACH restriction system. Authorisation would not be necessary because many (non-restricted) uses do not result in risks to local environmental compartments or contribute significantly to overall environmental concentrations. - The genotoxic substances acrylamide, DMS and 1,4-dioxane and one substance with a very low derived no effect level (DNEL), AA, require specific restrictions to control direct exposures for professional uses and potential diffuse environmental emissions from high total volumes of use in professional products. - One substance, AA, would require restrictions on professional uses of adhesives. - The uncertain toxicological profile of the six major phthalates (BBP, DBP, DEHP, DIDP, DINP, DNOP) and their potential effect on human health at low doses would mandate immediate consideration for enacting restrictions of all articles intended to be in contact with children. An appropriate category of phthalates (a phthalate subgroup) would need to be treated as a ‘group of substances’ based on similarity in chemical structure. Although phthalates have been identified as a subject of political contention, controversy surrounding their chronic toxicity and potential endocrine disruption appears to differ between Member States. Moreover, some phthalates do not appear to be very persistent very bioaccumulative (VPVB) compounds and may not exhibit certain toxicological effects which would make decision-making on many uses difficult to agree upon between Member States unless on a substance-by-substance basis. By limiting the scope of restrictions and not making phthalates subject to EU authorisation, Member States will retain flexibility for applying a wider set of national restrictions or other regulatory measures (e.g., taxes) to achieve further
249
Framework for Chemical Risk Management under REACH risk reduction. Continuing collection of further information on the risks of substances with categorisation according to appropriate subgroups, especially biomonitoring studies examining potential VPVB properties and endocrine disruption, would be expected. - Restrictions would be relevant to control environmental exposures of NP and NPE resulting from certain industrial applications rather than specific point sources. Consumer products with a potential for causing significant direct exposures would also be subject to the listed uses under the systems framework, thereby facilitating consumer and environmental non-governmental organization (NGO) awareness campaigns or product boycotts aimed at reducing environmental emissions from products. Member States would could apply current or future policy measures, such as taxes, to reduce risk levels within their national territories. - The corrosive and toxic properties of HF and corrosive and oxidising properties of H 2 O 2 would mandate controls for consumer and professional uses to support the systems framework recommendations. Targets (three substances) with restrictions (three substances) - MCCP and SCCP are subject to international conventions to which not all Member States are signatories. MCCP would proceed directly to target-setting because uncertainty in the risk assessment could result in differences between Member State views on the immediacy and severity of regulatory action. By comparison, certain specific uses of SCCP have been identified as requiring immediate risk controls through restrictions. All other emissions could then be controlled through target-setting because some form of overall EU risk reduction would be necessary to prevent environmental concentrations increasing to unsafe levels. - Uncertainty in the risks posed by butadiene would make it subject for target-setting to ensure adequacy of protective measures and reduce overall exposures by controlling use volumes.
250
Evaluating the Systems Framework - Sufficient reduction of risks to professional users from handling MTBE would be considered achievable through the development and review of recommendation equivalents. Targets would however be necessary to reduce overall environmental burdens particularly from accidental spills. - Toluene and o-anisidine would also require target-setting to reduce general environmental risks, but restrictions would need to be applied for controlling direct risks from consumer products. No EU risk reduction action (10 substances) According to the outcomes of EU risk assessments, action for controlling risks at the EU level was not deemed necessary for eight substances (acetonitrile, acrylonitrile, acrylaldehyde, BA, 4-CC, EEA, DDAC, cumene). Under REACH, this would correspond to a substance not requiring any action following evaluation. In other words, companies producing, using and marketing the substances must follow any risk reduction measures detailed in the registration dossiers as well as comply with any existing legislation triggered by any hazard classification and labelling. Under the systems framework recommendations and recommendation equivalents would apply as further mandatory controls for any risks, including environmental risks, resulting from professional uses (acetonitrile, acrylonitrile, MAA, MMA). Reporting of monitoring data may be required for specific emissions that result from potential sources of transboundary pollution resulting from industrial uses of acetonitrile and acrylonitrile. Further risk information to be collected (one substance) Significant uncertainty exists in the current risk assessment of bisphenol A, which has led to the EU regulatory risk assessment conclusion that further risk data need to be collected, especially because it is a potential endocrine disruptor. Under the systems framework, recommendations would apply during the time necessary to collect further information on bisphenol A, thereby providing a
251
Framework for Chemical Risk Management under REACH safety mechanism for consumers and professionals. Because it is not used in consumer products that come into direct contact with children other than present in some materials at trace levels, there is no current need to set any restrictions on articles. Following the review of further data, bisphenol A may be a candidate for any combination of measures.
Safe use exceptions could apply to any use that does not result in significant contributions to environmental risks resulting from ‘specific’ (i.e., point) sources. Equally, safe use exceptions to risk reduction decision-making would limit the number of consumer or professional uses subject to any given risk reduction strategy. This would apply to most industrial uses of the substances reviewed under the official EU risk assessment reports. The systems framework would therefore have avoided the need first for industry to report and then for regulators to review this information. The concept of permissible uses appears particularly relevant for the use of NP and NPE in spermicides and the use of penta-BDE in aircraft emergency evacuation systems [515]. Several chemicals require controls for point sources to environmental media (acetonitrile, acrylonitrile, HF, H2O2) that emanate only from a few industrial sites. From the review of previous RRS reports and existing EU legislation, there does not appear to be a method that requires Member States to demonstrate effective risk reduction compliance monitoring and enforcement at the national level for point source emissions during EU decision-making. Regulators appear to have few options to require reporting of compliance and enforcement data unless it is already mandated under existing legislative frameworks or the European Court of Justice. Moreover, several of the regulators interviewed stated that in addition to differences in implementing EU legislation, such as deviating from EU values to compensate for local environmental or economic conditions [208, 469], Member States report data in a variety of different formats: average values, aggregate values, and maxima (see also [481]). Because most European Community legislation is
252
Evaluating the Systems Framework enforceable at the national rather than the EU level, regulators also possess few mechanisms to respond to any breach of regulation in neighbouring national territories other than notifying the European Commission or resorting to the European Court of Justice [516]. The regulators interviewed explained how setting environmental standards usually requires elaborate co-ordinated schemes dependent on the dedication of significant monitoring resources at national levels. Moreover, standard setting requires agreement at the EU level even if there are only a few manufacturing facilities located across the entire EU. Therefore future development of reporting should focus on individual sources of emissions with a high potential for causing transboundary pollution rather than on environmental quality objectives or more general discharge limits from which regulators may deviate. To facilitate this procedure, and help make the relevant data available for decision-making under REACH, the systems framework proposes that such point source emissions be regulated under restrictions or authorisations, rather than other legislative frameworks, unless the data are made immediately available. Target-setting could avoid delays in decision-making resulting from (i) gathering further data to resolve different interpretations of risk assessment data, or (ii) the need to conduct socio-economic analyses of risk reduction measures. Such detailed analyses would need to apply only for restriction and authorisation procedures for reducing overall risks to human health or the environment. The enactment of restrictions for controlling direct (high-level) exposures to professional and consumer users could be differentiated from more general risk scenarios and proceed immediately at the EU level. Socio-economic analysis would otherwise apply only at the national level for selecting cost-effective strategies to meet targets, allowing Member States to decide on the level of assessment detail necessary to politically justify any decision. Finally, publishing the substances subject to target-setting could facilitate national regulators when issuing permits for manufacturing, such as under Integrated Pollution Prevention and Control (IPPC). The substances can also be linked to eco-labelling schemes at the national or EU level.
253
Framework for Chemical Risk Management under REACH Just as the criteria used to describe and establish targets can vary significantly, ranging from use volumes to health indices, regulators can involve a variety of measures to attain agreed targets. If existing legislation is already in place but appears ineffective, an increase in penalties could be used in conjunction with enforcement campaigns to improve compliance. In other cases, subsidies to support certain companies substitute the use of a substance under target-setting could be appropriate at the local level. Similarly, technology transfer schemes that improve existing processes can be used to achieve targets. In some cases, achieving targets under the systems framework could depend on communicating the need to use gloves in the workplace or to correctly dispose of waste material. Overall, the systems framework appears to be able to contend with the multitude of factors to be considered in regulatory decisionmaking. The decision-making matrix shown in Table 5.11, with seven categories for risk criteria, should facilitate the identification and selection of the appropriate regulatory outcome for any given chemical risk. Of the 22 chemicals for which co-ordinated EU risk reduction regulatory action is mandated4, 14 chemicals would proceed directly to specific risk reduction mechanisms (one Member State permitting, 4 EU authorisations, 2 targets, 7 restrictions). Only eight substances would need to undergo further EU decisionmaking to determine the appropriate risk reduction mechanism (i.e., a possible combination of targets and restriction or Member State permitting and restriction). Finally, a single substance would require the collection of further risk assessment data, either through industry data following evaluation or inputted via the Stakeholder Inputs. Although enacted under national legislation, the identification of the three substances subject to Member State permitting would
4
The other 10 substances would need to conform with general registration requirements under REACH and comply with national or EU legislation triggered by their classification and labelling.
254
Evaluating the Systems Framework facilitate EU decision-making (DEGBE, DEGME, TCE). Similarly, risk reduction measures would also be implemented at a national level for substances subject to target-setting, but minimum targets would be set at the EU level to achieve overall risk reduction.
6.2.2 Prioritisation of Regulatory Decision-Making – Results Following the methodology for the prioritisation of regulatory decision-making proposed in Section 5.4, the 33 chemicals have been rated. As part of the process, the chemicals have been categorised according to the nine possible quadrants of the plot shown in Figure 6.2. Recall that the priority ratings are qualitative, based on an indication of the overall probabilities that there is a need for a regulatory response to communicate, limit or otherwise control potential adverse events resulting from the production and use of each chemical and the extent of the corresponding action needed from regulators. While testing the systems framework, it was found that the prioritisation method provided a general format to summarise the results of each risk assessment and supplement the presentation of risk assessment conclusions used in the decision-making matrix5. As noted in Section 5.4, the rules for combining the relevant prioritisation rankings for the regulatory action indicators and the related probability indicators could be revised through the use of other decision-making tools, such as multi-criterion analysis (see Section 5.4). The 33 chemicals can be broadly divided between ‘high’, ‘medium’ and ‘low’ priority ratings, as indicated by the shading in Figure 6.2. 5
Information from EU risk assessment reports used to carry out the priority ratings is detailed in the original publication of this thesis in 2007.
255
Framework for Chemical Risk Management under REACH Assignment of regulatory measures to control risks from Section 6.2.1 is also shown. Chemicals that would be subject to Member State permitting, restrictions or targets are found in several different regions of the plot. This reinforces a crucial point of the research findings: although the REACH legislation intends for the authorisation process to control risks from ‘substances of very high concern’, it is not necessarily the best regulatory instrument to achieve risk reduction. If Member States disagree on the uses that should be authorised, delays to enacting regulation will occur. Furthermore, a country will have limited ability to enact more stringent regulatory requirements at the national level once a use has been authorised. Other regulatory options under REACH may sometimes be more appropriate for controlling ‘substances of very high concern’. It is apparent from Figure 6.2 that there are no instances of a risk with a high overall priority rating for the extent of action and low rating for the probability of a regulatory response needing to occur, or vice versa. This can be explained by: UÊ /
iÊ
>â>À`ÊV>ÃÃwV>ÌÊvÊ>ÊÃÕLÃÌ>ViÊÌÀ}}iÀÃÊi`>ÌiÊVÌÀÃÊ under existing regulation to control exposures. Because a hazard is identified and the rating for extent of action increases, the controls on exposure under existing hazard-based legislation schemes reduce the probability of the regulatory response being necessary. This is why no chemicals fall in the bottom right-hand region of Figure 6.2. Under REACH, the mixture hazard assessment process is specifically designed to identify substances with potential hazardous properties through in vitro and in silico techniques. In this respect, REACH may result in over-protective measures to control exposures if it does not generate sufficient data on the probability of an adverse event occurring to enable resource-efficient or cost-effective application of controls under existing legislative frameworks.
256
Probablility of the need for regulatory action
Evaluating the Systems Framework
Targets and Restrictions: toluene Restrictions: 1,4-dioxane
EU authorisation: penta, octa, deca-BDE Targets and restrictions: o-anisidine phthalate sub-group SCCP Restrictions: acrylamide Targets: MCCP
No EU Action: acetonitrile acrylonitrile MAA MMA
EU Authorisation: TCB MS Permitting: DEGME Restrictions: AA DMS No EU Action: BA Further Information: bisphenol-A
Restrictions: NP/NPE Targets: MTBE
Restrictions: H2O2 HF No EU Action: cumene 4CC DDAC EAA
MS Permitting: DEGBE TCE Targets: butadiene No EU Action: acryaldehyde
Extent of regulatory action Figure 6.2 Relative positioning of 33 chemicals on a qualitative plot of probability of the need for regulatory response versus extent of Ài}Õ>ÌÀÞÊ>VÌÆÊ`>ÀiÀÊÃ
>`}ÊÀi«ÀiÃiÌÃÊ
}
iÀÊ«ÀÀÌÞÊV>Ìi}ÀÞ°
257
Framework for Chemical Risk Management under REACH UÊ /ÊÃiÊiÝÌiÌ]ÊiÝÃÌ}Êi}Ã>ÌÊVÌÀÃÊÌ
iÊ«ÀL>LÌÞÊvÊ damage to human health and ecosystems resulting from largevolume emissions to the environment regardless of intrinsic chemical hazard. For instance, biological oxygen demand must be considered when discharging high volumes of a ‘non-toxic’ substance into the aquatic environment and the use of personal protective equipment (PPE) must be worn in the workplace when handling intermediates. A substance may also be part of a where other components result in the mixture being classified and subsequently managed as a ‘dangerous preparation’ or ‘hazardous waste’. This aspect of existing legislation reduces the potential occurrence of a high exposure level to non-hazardous substances6 (i.e., top left-hand region in Figure 6.2) There are several instances where the combined rating for the regulatory action indicators is the same for one or more chemical, but differences in the probability rating results in different priority scores. For instance, DEGME and toluene receive ‘medium’ ratings for the overall extent of regulatory action: with similar ratings for the exposure indicator, DEGME rates higher than toluene for the hazard indicator and toluene rates higher than DEGME for the social mobilisation indicator. There is considerable uncertainty associated with the hazard assessment of toluene. Taken as indication that toluene is likely to have a hazardous property that has yet to be identified, this results in a higher priority rating for toluene than DEGME. In comparison, the risks of acetonitrile and bisphenol A vary considerably in terms of the overall action indicator rating. Acetonitrile has a limited potential to cause chronic toxicity, whereas evidence indicates that bisphenol A is a reprotoxin and an endocrine disruptor7ÆÊ Lë
iÊ Ê >ÞÊ LiÊ «ÀiÃiÌÊ Ê >ÞÊ 6
Physical observations of negative effects on human health and the environment are often more likely to be observed and identified following high-level exposures.
7
As noted in Box 6.3, although bisphenol A is a potential endocrine disruptor, exposure levels are not anticipated to exceed any DNEL when used as a substance or preparation when correct safety measures are used. Further information is needed to confirm this risk assessment conclusion and examine the likelihood that exposure exceeds DNEL when safety measures are not applied.
258
Evaluating the Systems Framework products, albeit at trace levels, yielding a slightly higher rating than acetonitrile for the exposure action indicator. Probabilities associated with the need for regulatory action to manage potential risks associated with the manufacture and use of these two substances are relatively equal in terms of the qualitative ratings because controlling exposures to these substances largely depend on risk management measures – e.g., emission control of during >Õv>VÌÕÀ}ÊvÊ>ViÌÌÀiÊ>`ÊLë
iÊÆÊ** Ê`ÕÀ}ÊÌ
iÊÕÃiÊvÊ iÌ
iÀÊÃÕLÃÌ>ViÆÊ«>ÀÌVÕ>ÀÊV>ÊiÝ
>ÕÃÌÊÛiÌ>ÌÊVÃ`iÀ>ÌÃÊ for certain uses of acetonitrile. Although the probability rating for the need of a regulatory response based on hazard is higher for bisphenol A than acetonitrile, this rating is not high enough to increase the overall probability indicator rating for bisphenol A from medium to high. Because both substances receive low action and probability indicator ratings associated with social mobilisation, it is the action indicator ratings for hazard and exposure rather than the probability indicator ratings that result in bisphenol A being placed in a higher priority category than acetonitrile. The proposed systems framework’s prioritisation method can help decide on measures and monitoring activities that target the control of the probability or the extent of further regulatory action being necessary, for example: UÊ ÃÕÀ}Ê Ì
>ÌÊ ÀiiÛ>ÌÊ Ã>viÌÞÊ i>ÃÕÀiÃÊ >ÀiÊ Li}Ê VÀÀiVÌÞÊ implemented can reduce the probability for needing additional Ài}Õ>ÌÀÞÊ>VÌÆ UÊ *ÀiÛiÌ}ÊÌ
iÊÀii>ÃiÊvÊ>Ê6*6 ÊÃÕLÃÌ>ViÊÌÊÌ
iÊiÛÀiÌÊ can limit the extent of regulatory action by reducing the size of the potentially exposed population(s) or ecosystem(s). Chemical Priority-Ratings The priority-rating scheme identifies that acrylamide, o-anisidine, MCCP, SCCP, phthalates and the three BDE substances demand the most immediate attention. Appearing in the highest region on the plot in Figure 6.2, EU regulators should consider dedicating resources
259
Framework for Chemical Risk Management under REACH to co-ordinating and monitoring further action at an international level to control these chemicals. By comparison, 14 chemicals that do not require regulatory action at the EU level generally fall in the lower left-hand region of Figure 6.2. Existing legislative frameworks and other measures implemented at the national or local level should continue to be used to manage the risks from these substances without the need for co-ordinated action or decision-making under REACH. Three exceptions are present: BA, bisphenol A and DEGME are identified as posing medium-risk levels but do not mandate EU regulatory action according to the system framework’s decision-making matrix. Given the risk potential of these substances, some form of market restrictions should be placed on the use of these substances, at least until further hazard or exposure data are generated. Judging by previous EU legislation, temporary bans may be best issued in the form of a commission decision (see Section 5.3.3). In parallel, Member State regulators should prioritise compliance monitoring and enforcement activities on these three substances. Toluene, NP/NP and MTBE need particularly prompt attention. To ensure an appropriately high degree of protection, regulators need to enact rapid regulatory decisions while enforcement activities target the implementation of existing safety measures by chemical users. In terms of preventive measures, national risk management activities should seek to create incentives for companies to review existing products and processes involving these substances. Controlling risks from 1-4 dioxane appear to be of greater concern than risks from AA and DMS, even though all three substances require restrictions on their concentration in professional products. Out of the five chemicals identified for target-setting, establishing and implementing targets for butadiene would be of lowest priority. Regulating MTBE would also be of relatively low priority, partially because existing national and European Commission legislation already cover many sources of exposure. Rather than setting specific use or exposure-reduction targets, the Monitoring Network could
260
Evaluating the Systems Framework decide to co-ordinate environmental monitoring programmes, enforcement activities or attempt to improve certain practices through the directive on IPPC. Two chemicals subject to Member State permitting (DEGBE, TCE) and two chemicals that should be subject to restrictions (H2O2, HF) are located in the low-priority group. These findings coincide with the complaint by several of the interviewed regulators that ‘too much attention’ is being directed at these substances. Evidently, this could have been avoided using the proposed system framework prioritisation method. Other substances with draft or completed risk assessments would have been made subject to risk reduction decision-making while recommendations would provide a general protective coverage for consumer and professional products containing the substances. None of the three chemicals for Member State permitting (DEGME, DEGBE and TCE) fall in the high-priority region of Figure 6.2. Because the details of Member State permitting schemes are left to decision-making and implementation at the national level, the prioritisation method would serve to help communicate relative degrees of priority to regulators and stakeholders. However, when a chemical identified for Member State permitting were to receive a medium to high priority-rating, action to restrict use at the European Community level could always be considered as an option, as in the case of DEGME. In practice, the method for prioritisation proposed as part of the systems framework is able to present priorities for EU decisionmaking in a clear and concise format. A two-dimensional plot of the probability of a regulatory response being necessary versus extent of the regulatory action shows a variety of information, thereby offering a method to help regulators focus monitoring and enforcement activities. The application of the prioritisation method has yielded some interesting results, such as the identification of the need for regulatory action to be taken to manage the risks from acrylonitrile and bisphenol A even though the conclusions of the
261
Framework for Chemical Risk Management under REACH EU risk assessment and the application of the systems framework decision-making matrix both indicate otherwise. The prioritisation method could also isolate several substances that should be given particularly low priorities at the EU level (e.g., BA, butadiene, TCE) which demonstrates some of the inefficiencies of the current regulatory system that could continue under REACH.
6.3 National Approaches under the Systems Framework The systems framework described here would incorporate aspects of the regulatory approaches of France, Germany, Sweden and the UK. First, it proposes a hazard-based approach to maximise the use of existing data during registration and generate data on the chemical contents of certain consumer products (listed uses). Second, it seeks to combine hazard criteria with technical-based approaches to controlling exposures by isolating specific uses that would be subject to the recommendations. Third, the identification of permissible and safe uses would follow a risk–benefit rationale, where potential risks from chemical uses are likely to be outweighed by their benefits. Fourth, the systems framework would rely on technical-based data for monitoring emissions from industrial facilities with a high potential for causing trans-boundary pollution. Fifth, the target-setting mechanism follows a risk–benefit approach for Member States to achieve overall risk reduction for tolerable uses. Following the discussion in Section 6.1, it is not anticipated that adopting the listed uses would create a major hurdle to regulators, especially because listed uses would apply only to products with a potential for resulting in high direct consumer exposure levels. Furthermore, the listed uses follows the general principles of transparency of regulation and public ‘right to know’ enshrined under the Aarhus Convention. Although currently an anomoly, such an exposure-based rationale to regulation would be scientifically justifiable. As detailed in the literature review, sufficiently high exposure to any substance can present a risk.
262
Evaluating the Systems Framework The proposed mechanism to verify the control of manufacturing point-sources of trans-boundary pollution is also expected to be broadly accepted by most Member States. It offers the possibility to identify, evaluate and compensate for variations in the implementation of existing EC environmental legislation. With the continual increase in chemical production and a potentially large number of VPVB substances being identified under REACH, such a mechanism may prove particularly valuable in assisting future EU decision-making and chemical enforcement activities across the Directorates-General of the European Commission. Similarly, because controls in the workplace fall under Article 138 of the EC Treaty (see Section 5.2), the proposed scheme should not cause any (justifiable) rejection by EU regulators. Moreover, not making all substances of ‘very high concern’ subject to authorisations under REACH would enable Member States such as Sweden to include, or continue to include, such substances subject to existing national occupational authorisation schemes. Equally, existing product or process permitting schemes could continue to operate for substances undergoing target-setting, which may be particularly relevant for legislation based on either Article 138 or 175 of the EC Treaty. Risk–Benefit Approaches Countries that rely on statistics for occupational safety, industrial accidents, environmental emissions, or other measured data, such as France and the UK, could maximise their existing use of statistics for devising how to achieve EU targets. Equally, these countries may be able to provide important data when devising targets at the EU level and establishing the corresponding benchmarking schemes. By increasing regulatory attention on mutagenic and reprotoxic substances, as well as VPVB substances, the systems framework could expand existing national occupational monitoring schemes that focus on carcinogenic substances or substances with acute effects. A Member State that takes a risk–benefit approach, such as the UK, can conduct such analyses when deciding on how to meet target-setting requirements.
263
Framework for Chemical Risk Management under REACH Each national regulatory authority could meet targets by engaging with different actors. Based on Chapter 4, it is probable that this would involve trade unions and environmental NGO in Sweden, the manager unions and occupational health officials in France, retailers and environmental NGO in the UK, and the consumer NGO in Germany. This may improve communication between certain national regulatory authorities, for example the Chemicals Inspectorate (KemI) and the Swedish Environmental Protection Agency (SEPA) in Sweden. It may even reduce the potential for worsening stakeholder relations by focusing the attention of EU decision-makers and increasing the influence power of EU NGO. As with most EU decision-making, agreement on regulatory options under REACH can be expected to follow qualified majority voting rules. Presenting the regulatory options using the decision-making matrix as shown in Table 6.1 could facilitate consensus formation. First, as the decision-making matrix can result in a set of possible regulatory outcomes, each potential outcome could be voted on. Therefore, if option ‘1’ is not agreed, there is still the chance that option ‘2’ may be. Second, the decision-making matrix would provide a clear and simple format to compare previous RRS decisions. It may thereby provide some basis for Member States to develop voting strategies, which currently does not appear to exist due to the lack of experience by regulators with EU decision-making (Section 5.1). Technical Approaches Strong opposition to adopting the systems framework could result from the regulatory recommendations for consumers and professional users. Based on previous experience with risk assessment and risk management, industry and regulators are often reluctant to change (Section 2.3). Even though any restrictions necessary for supplementing the regulatory recommendations would maintain a case-by-case approach, the concept of recommendations deviates significantly from the current substance-by-substance approach to EU chemical regulation.
264
Evaluating the Systems Framework Due to the technical basis of the recommendations, it is conceivable that France, Germany and Sweden may support the proposed procedure. In particular, the inclusion of recommendation equivalents closely resembles existing German regulation. Because the Netherlands initially proposed establishing a similar scheme for decision-making rules, which it appears to have abandoned due to the forthcoming REACH regulation, this country could also be willing to support the approach. Opposition to this mix of hazard- and technicalbased approach to regulation could arise from countries that take a risk-benefit approach, such as the UK or Southern Member States that are generally reluctant to support new EU legislation or policy. However, the recommendations are based on previous restrictions. It can therefore be argued that they do not radically deviate from previous regulation or policy. Probably in favour
Possibly in favour
Uncertain
Unlikely to support
Denmark - 3
Austria - 4
Belgium - 5
Cyprus - 2
Finland - 3
Estonia - 3
Italy - 10
Greece - 5
France - 10
Czech Republic -5
Luxembourg - 2 Ireland - 3 Portugal - 5
Germany - 10 Hungary - 5
Malta - 2
Netherlands - 5 Lithuania - 3
Spain - 8
Sweden - 4 Latvia - 3
UK - 10 Poland - 8 Slovakia - 3 Slovenia - 3
Table 6.2 Predicted voting for adopting the system framework regulatory recommendations
265
Framework for Chemical Risk Management under REACH If Germany holds political sway on Eastern Member States due to trade8 and Sweden is representative of Nordic states9, this would give 68 votes in favour of the regulatory recommendations (refer to Table 6.2). Because the Netherlands initially proposed a method of ‘decision-making rules’ broadly similar to the recommendations (i.e., a matrix of hazard and use criteria) it is foreseeable that this country would also vote in favour, bringing a total Council of European Ministers vote count to 73. To be approved by a qualified majority, a proposal need only secure 62 of the total 87 votes – based on 25 Member States before the latest enlargement. The recommendations could therefore be adopted even if the following countries withdraw or abstain from voting: UÊ ÊÃ}iÊ>À}iÊVÕÌÀÞ]Êi°}°]ÊÀ>ViÊÀÊ*>`Æ UÊ /ÜÊi`ÕÃâi`ÊVÕÌÀiÃ]Êi°}°]Ê âiV
Ê,i«ÕLVÊÀÊÕ}>ÀÞÆ UÊ /
ÀiiÊÃ>ÊVÕÌÀiÃ]Êi°}°]Ê ÃÌ>]Ê>ÌÛ>]ÊÌ
Õ>>]Ê-Û>>Ê or Slovenia. Out of the remaining 11 Member States, Austria may also contribute a favourable position to the recommendations. Austria has typically been categorised as a ‘leader’ in environmental policy [44] and its shared history with Germany could mean that its regulatory system is technically oriented. Finally, it is possible that Belgium, Italy and Luxembourg which have been grouped as ‘fence-sitters’ together with France [44] ‘jump on the bandwagon’ of supporting the recommendations.
6.4 Consensus on Banning Chemical Production or Use A major obstacle in reaching consensus during EU risk management decision-making arises when a large variation in the economic impact
8
There is a growing chemical trade between Germany and several East European Member States [318, 330].
9
This is referenced in Section 3.8 and supported by several interviewee responses.
266
Evaluating the Systems Framework of a ban occurs across Member States [517]. According to several of the regulators interviewed, a Member State will often be reluctant to ban chemical manufacturing or use when significant production of that chemical occurs within its territory. Several examples provided by the interviewees were used to support this assertion, such as the description of the UK opposition to regulating DCM while being the major EU producer (see [518]) and a French objection to regulating para-DCB (see [519]). The converse of the above argument may also hold true. As noted by some interviewees, Member States often appear willing to set stringent regulation when significant levels of production do not occur within their territory. In the past year, the UK and Sweden proposed national bans on substances that are not significantly produced or processed in their countries – PFOS and HCBD, respectively [512, 518, 520]. Similarly, interviewees stated that Germany is not a major producer or user of DCM [521, 522] and that France does not produce large quantities of DEGBE or DEGME (see Box 5.1 in Section 5.1). There is no evidence from the research interviews or the literature review to indicate that carrying out a socio-economic analysis of the impact of alternative regulatory measures facilitates EU decisionmaking. At the EU level, conducting in-depth analyses of socioeconomic data can be highly complex, taking around 50 persondays for an individual chemical single-use evaluation and costing around 50,000 [523]. Even when the process has been completed, interviewees described the results of a socio-economic analysis as highly subject to independent interpretation by each Member State, as one would expect from the findings presented in Chapter 4. Obviously, the costs and benefits of any regulatory measure will also depend on the national health care systems, and local employment. Socio-economic analyses of regulatory options appear far more relevant at national and local levels. In particular, this decisionmaking tool has the potential to identify businesses that may require particular financial or technical support to help them to conform with new regulatory demands (Section 2.4.5). Unfortunately, according
267
Framework for Chemical Risk Management under REACH to the REACH legislative proposal, socio-economic analysis of alternative regulatory measures will need to be done at the EU level during the restriction and authorisation processes. Consultation necessary for conducting this socio-economic analysis under REACH will be more structured than under the current RRS process because substances facing a ban will be published on the REACH-IT website and all businesses producing or selling these substances will be notified. Deadlines for submitting consultation contributions will also be set for businesses, stakeholder groups and regulators. Nevertheless regulators, business and NGO will need to contend with the fact that a producer of a substance already on the market may have more resources available to dedicate to costly regulatory consultation procedures than the producer of a new alternative substance10. The outcome of consultation and any subsequent socio-economic analysis may therefore be biased towards an existing producer. Similarly, following experiences with the current ‘RRS’ process, downstream users of chemicals also need an opportunity to voice their concern over potential chemical regulation. For instance, upstream compliance costs will depend on whether downstream users are willing to pay a higher price for a substitute. By creating the listed uses, tolerable uses and the recommendations, the systems framework attempts to circumvent the regulatory complexities of substitution. Interviewees agreed that the substitution principle is as much, if not more, a ‘hands-on’ risk management tool rather than a ‘top-down’ regulatory instrument. Efficient and effective application of the substitution principle depends on companies having the necessary management structures and information available to control occupational exposures and create ‘safer’ products rather than communicating potential risks and benefits of legislative action to regulators. As several regulators explained during the interviews, decision-making on regulatory measures ultimately comes down to a vote between the 10
As exemplified by the case of alternatives to DEHP presented in Section 2.2.2.
268
Evaluating the Systems Framework Member States. The systems framework proposes a structured method to facilitate this voting procedure: each Member State can vote on whether it agrees with the outcome of the decision-making matrix, the method for prioritisation, or whether restrictions are necessary to support the recommendations. One of the regulators interviewed who had worked in the European Commission described Member States as tending to agree on EU risk management following discussions on alternative regulatory measures that avoid reference to specific detailed data entries. According to this regulator, a more general overview of the advantages and drawbacks of alternative regulatory outcomes can suffice and even promote the process of negotiation. In this respect, the systems framework could provide an important methodology to facilitate the transition from quantitative to qualitative data that is often necessary for political decision-making.
6.5 Discussion and Conclusions Compared with the REACH proposal, the systems framework proposed in this dissertation would focus more on gathering and communicating risk information. This serves two purporses. First, the systems framework would involve a wider number of stakeholders than mandated in the REACH legislative text. Creating a simple mechanism for these actors to follow without needing to understand all the technicalities of registration should distribute responsibility for achieving chemical safety. In turn, these activities and the publication of information on the REACH-IT website would facilitate the registration process for chemical producers and importers. Second, the systems framework would avoid the potential for companies to wait for registration deadlines before engaging in necessary risk management activities. In this way, the proposed scheme would limit aspects of ‘safety’ management that may otherwise be overlooked during the initial stages of implementing REACH. If adopted, the decision-making rules and recommendations would represent a fundamental transition away from the current case-bycase approach to regulatory risk management. The decision-making
269
Framework for Chemical Risk Management under REACH matrix would also enable regulatory outcomes to be presented in a clear, concise and predictable way. Testing the systems framework has demonstrated its potential to streamline and structure the current RRS process. In total, only 10 out of 33 chemicals had more than one possible regulatory option. Subsequent selection of the regulatory options for these chemicals followed criteria from the existing RRS TGD and current requirements of the REACH legislative text but condensed into 11 pages. Avoiding delays in decision-making can be very important in safeguarding human health and the environment. The research findings demonstrate why a resource-intensive use-by-use authorisation procedure is not necessarily the most effective EU regulatory option for controlling high-level chemical risks. Other regulatory options are anticipated to result in more rapid and efficient agreement between Member States, particularly because authorisations will limit the extent to which a Member State can deviate from centrally defined standards. The two-dimensional plot of probability versus extent of regulatory action enabled regulatory prioritisation to be presented in a format that should be easily interpreted by regulators, stakeholders and members of the public. Based on the experiences of risk management reported in Chapter 4, drawing attention to specific chemicals of high regulatory concern is important for many stakeholder groups activities. For instance, the information could be used by German consumer NGO to monitor the chemical contents of consumer products, Swedish trade unions to monitor occupational exposures, and by British NGO to monitor exposures in the general public. At a national level, different regulators may also use this information for regulatory activities that fall outside of REACH restriction and authorisation, such as national permitting procedures for the control of occupational exposures or water treatment. Altogether, the systems framework demonstrates just how important the role of future TGD can be for the implementation of REACH. For instance, setting regulatory recommendations that apply
270
Evaluating the Systems Framework immediately would affect imports as well as chemicals produced in the EU. In turn this can impact international competitiveness of EU chemical producers. While the systems framework contains many technical elements, it is important to recall that many aspects of chemical risk management remain political in nature. By excluding the TGD from the official co-decision process involving the European Parliament, the development of the TGD could remain in the hands of technical regulators. Judging by the difficulties that these same regulators have already had to face with the current RRS process, it is foreseeable that many current risk management dilemmas could persist under REACH. Evaluating the framework’s ability to accommodate different regulatory approaches suggests that many aspects of the systems framework could be politically adopted at the EU level, especially if France, Germany and Sweden first agreed between themselves. Nevertheless, even if a framework resembling the one proposed in this dissertation is adopted, many technical aspects relating to monitoring and enforcement under REACH still need to be resolved before decision-making can begin. For instance, reference test materials need to be available to regulators analysing the chemical contents of products. Similarly, evidence from the research indicates that Customs and Excise inspectors in many Member States will need to have significant financial resources made available to them for implementing REACH.
271
7
Conclusion
- Accord Harmony (Ambrose Bierce, The Cynic’s Word Book, 1906)
Introduction The PhD project that forms the basis of this book appears regulatory approaches to chemical risk management and the first to focus on REACH procedures. The research shows how differences in national approaches manifest themselves and how they become even more pronounced during European Union (EU) decision-making. The regulatory strengths and weaknesses of each country were determined by investigating the underlying social, political and economic causes for the variations. While a set of recommendations has been formulated for improving regulation in each country, unless deficiencies in current risk management practices are addressed at a national level, then future EU legislation will need to compensate for the limitations. Given the variation between national approaches, this could prove particularly cumbersome to EU decision-making and could further complicate the implementation of REACH. No matter to what extent the surveyed countries address the specific recommendations proposed in this thesis, certain EU regulatory practices need to change. Fundamentally, EU decision-
273
Framework for Chemical Risk Management under REACH making must account for different national approaches. The first section of this Chapter reviews exactly how the systems framework aims to harmonise EU decision-making. The overall implications of the research findings are then set against different contexts: (1) sustaining the competitiveness of the EU chemical industry (2) «ÀÛ}Ê >Ì>Ê >`Ê 1Ê `iVÃ>}ÆÊ Î®Ê «iiÌ}Ê , ÆÊ {®Ê >>ÞÃ}Ê ÀÃÆÊ >`Ê x®Ê >>}}Ê }L>Ê V
iV>Ê risks. Finally, after a brief retrospective analysis of how the entire conceptualisation of REACH could have been devised differently, consideration is given to the research contribution of this thesis and future avenues of research that could be explored.
7.1 Harmonisation of EU Risk Management The new EU chemicals policy has not focussed on the risk >>}iiÌÊ>ëiVÌÃÊvÊ 1ÊV
iV>ÊÀi}Õ>ÌÆÊÃÌi>`]Ê, Ê has concentrated on generating and collecting laboratory-derived hazard assessment data. A number of strategies to compensate for the potential inefficiencies of EU risk management can therefore be drawn from the research findings (Table 7.1). The proposed EU strategies encompass most of the national recommendations of the four countries (Section 4.7), the exception being that many small and medium-sized enterprises (SME) need ‘hands-on’ chemical safety technical advice and support, which is expected to become even more pronounced when implementing REACH. Many actors, from trade unions to insurance institutions, can aid SME with regulatory compliance, but this aspect of risk management may be best left to national rather than EU level activities.
274
Conclusion
Prevention
Protection
Develop extended producer responsibility as a regulatory tool
Develop methods to standardise and promote best practice
Product-based legislation
Develop methods to improve ‘bottom-up’ substitution
Develop rapid methods to set EU product standards
Socio-economic analysis (SEA)
Streamline SEA by targeting specific sectors or limiting its application to the national level
Incorporate mutagens and reprotoxins into regulatory priorities, statistical studies and SEA
Supply chain
Table 7.1 Strategies to improve chemical safety that should be adopted at the EU level
The proposed systems framework for EU decision-making under REACH seeks to counterbalance the weaknesses and to draw on the strengths of the national approaches while addressing the points shown in Table 7.1. Specifically, the framework would fuse hazard, technical and risk–benefit approaches to risk management (Table 7.2). Compared with the current process of chemical legislation and the recent REACH proposal, the systems framework would avoid a linear substance-by-substance approach by applying a set of decision-making rules based on hazard and use to all chemicals (Section 5.3.1).
275
Framework for Chemical Risk Management under REACH
Increased input of epidemiological and environmental monitoring data to REACH Hazard
Regulatory recommendations for professional and consumer products Listed uses (published on REACH-IT website) for substances in certain consumer products Safe and permissible uses for certain applications Regulatory recommendations and optional use of recommendation equivalents
Technical
Development of monitoring and benchmarking schemes for target-setting Regulatory decision-making dependent on timelines for implementation and availability of testing standards for compliance monitoring and enforcement Target-setting schemes for countries to establish levels of safety in national territories that go beyond minimum EU standards
Risk–Benefit
Member State permitting option for controlling occupational exposures Prioritisation method for regulatory decision-making
Table 7.2 Hazard, technical and risk-benefit aspects of the Systems Framework
Harmonised EU decision-making does not imply setting uniform chemical regulation across Member States but refers to countries reaching an accord on when action should be taken at the EU level and how national actions can be co-ordinated. The systems framework therefore proposes a structured process for stakeholder 276
Conclusion participation, a communication network between the DirectoratesGeneral of the European Commission (Section 5.4) and a set of harmonised objectives for EU risk management in the forms of: (i) decision-making rules, (ii) recommendations and (iii) prioritisation (Sections 5.3 and 5.5). Achieving consensus between Member States during EU decisionmaking depends on generating an awareness of the contexts behind national approaches among the regulators and stakeholders involved. The systems framework attempts to create opportunities for Member States to communicate particular situations that have evolved in their country. This inclusion of ‘national dimensions’ into decision-making would respect the subsidiarity principle, where a Member State can set the level of protection to health or ecosystems that it deems necessary within its national realm. By incorporating a wider set of control options into REACH (e.g., target-setting) and co-ordinating actions under various legislative frameworks, the proposed framework aims to maximise a Member State’s ability to achieve independent political objectives, which in turn should facilitate reaching agreement on EU risk reduction strategies (RRS). A tremendous amount of information is available on chemical risks and safety measures, distributed among many regulatory administrations, companies and other organisations. EU regulation currently offers few opportunities or resources to collect, collate or exchange these data. Incorporating a ‘target-based’ control option into REACH decision-making could promote the identification, dissemination and adoption of best practice and support incentivebased risk management schemes. Co-ordinating data processing aspects of EU chemical risk management would respond to a demand from companies to harmonise the administrative processes for demonstrating regulatory compliance across EU countries. The systems framework proposes that the newly created European Chemicals Agency provide a necessary and long-overdue platform for exchanging activities and sharing experience relating to chemical risk management between Member State regulators and the European Commission.
277
Framework for Chemical Risk Management under REACH
7.2 Implications of the Research 7.2.1 Sustainability in the EU Chemical Industry The chemical industry is facing a crisis in terms of its sustainability. Public trust in the industry needs to be restored in many EU countries and economic growth in the sector needs to be uncoupled from increased production volumes. REACH only begins to address these two issues. While REACH will generate information on chemical risks and will reset the frame of reference on what constitutes ‘sufficient’ toxicological data for carrying out hazard assessments, it does not address how to control exposure levels. Apart from the authorisation process, REACH does not even propose specific mechanisms to control the increasing concentrations of synthetic chemicals in human blood and environmental media. There are winners and losers under any regulation. Banning a substance can negatively affect the profits of a company producing the substance but prove necessary to protect human health and ecosystems. Regulating the use of a substance can also provide a market opportunity for the manufacturer of a suitable alternative product. Given the economic contribution of chemical production and supply to manufacturing across Europe, it is not surprising that countries are often reluctant to regulate when significant production occurs within their territory (i.e., they are the potential ‘losers’). Evidently, the voice of producers facing a ban often resonates louder than that of downstream users or producers of alternative products. A substance-by-substance approach to regulation can result in one dangerous chemical being banned, replaced by another which in turn is banned and replaced by a third, and so on. Unpredictable business environments are thereby created through manufacturing supply chains. Regulators should establish a clear set of guidelines for industry to follow. Rather than hitting chemical producers and users with separate demands on a case-by-case basis, applying a set of regulatory requirements according to chemical use categories
278
Conclusion should create a more equitable approach to risk management. As any given country will contain a mix of producers and users, it is possible that such a regulatory approach may achieve a balance between the ‘winners’ and ‘losers’. It is even foreseeable that many companies could simultaneously experience market obstacles and opportunities under such an approach. Chemical companies can provide downstream users, suppliers and retailers with a range of products and services, from the screening of hazardous substances to the supply of personal protective equipment (PPE). Manufacturers and retailers need to make use of science and engineering to control the release of chemicals during product use and disposal, which are central tenets of product stewardship and extended producer responsibility. Developing these avenues of EU policy can ensure continued growth in the chemicals sector while controlling the increasingly numerous sources and levels of chemical exposure. Due to international competition, creating a ‘level playing field’ for EU businesses largely depends on controlling imports. Based on the four surveyed countries, this will require a significant increase in regulatory resources in most Member States. As future regulatory activities will need to be carefully prioritised, potential risks need to be clearly communicated to regulators and businesses according to product types. At the same time, reducing administrative burdens for demonstrating regulatory compliance can support innovation and promote industrial competitiveness. Altogether, the three primary mechanisms – decision-making rules, recommendations, and prioritisation – proposed under the systems framework should improve the definition of regulatory–business interfaces and provide clear guidance to companies. Using regulatory measures to create a responsible chemical culture can be limited if the political aspects of risk management are not distinguished from the scientific. Industry or regulators can be biased against taking regulatory action and then use science as a basis for delaying the enactment of control measures. Arguing against evidence of a risk on purely scientific grounds can also undermine stakeholder 279
Framework for Chemical Risk Management under REACH trust when it neglects the ‘scientifically’ recognised yet ‘socially’ classified political dimensions of risk. Publishing listed uses and assigning liability for following the recommendations are two mechanisms that could support industry and regulators in adopting more proactive and responsible methods of evaluating and managing risk data. The prioritisation method proposed under the systems framework should serve as a ‘wake-up’ call to many companies and regulators by demonstrating how risk levels increase if exposures to persistent and bioaccumulative substances are not controlled. To counter potential sources of bias, EU chemical regulation also needs to develop and strengthen methods that incorporate wider sources of data, such as epidemiological surveys, occupational health statistics and poison centre reports. Without this type of ‘surveillance’ monitoring, risks may not be identified or correctly managed, even if stakeholders or regulators are observing damaging effects on human health or the environment. In the area of occupational health protection, developing such aspects of risk management is likely to be particularly important for countries with substantial national insurance coverage for work-related health and safety. A key conclusion of the research project is that Germany may play an even larger part in EU decision-making and consequently exert a strong influence on the future of the EU chemical industry. This is the result of a complex mix of the following factors: UÊ Ì
iÊÃâiÊ>`ÊÃÌÀÕVÌÕÀiÊvÊÌ
iÊiÀ>ÊV
iV>Ê`ÕÃÌÀÞÆ UÊ iÀ>Þ½ÃÊ«ÌiÌ>ÊyÕiViÊÊ>ÞÊ 1ÊiLiÀÊ-Ì>ÌiÃÆ UÊ iÀ>Þ½ÃÊVÕÀÀiÌÊiVVÊVÀÃÃÆ UÊ vÕÌÕÀiÊLÕÃiÃÃÊ«ÀiÃÃÕÀiÃÊ>ÌV«>Ìi`ÊÊ«iÌÀV
iV>ÃÆÊ>`] UÊ Ì
iÊ«ViÃÊ>`«Ìi`ÊLÞÊiÀ>ÊÃÌ>i
`iÀÊÀ}>Ã>ÌÃÊÌ
>ÌÊ avoid interrupting chemical production.
280
Conclusion Unfortunately, Germany’s influence on EU policy and industry may not be positive. It is foreseeable that German industry and regulators may seek to delay effective bans for regulating substances produced within its territory even though the risks arise in other countries. In such cases, it may be particularly necessary to empower regulators and stakeholder organisations in other Member States to control their own level of human health or environmental protection using instruments outside the scope of REACH. Otherwise, public confidence in the chemical industry across the EU may continue to decline.
7.2.2 National Approaches and EU Decision-Making Political dimensions of risk management contribute to the quality, type and level of information needed during assessments of hazards, exposures and regulatory impacts. This book illustrates just how Member States may sometimes need to establish different levels of protection within their national territories. The analysis of the national approaches demonstrates why Member States give preference to certain risk management control instruments. The research also shows how a single country can influence EU decision-making. For instance, the UK appears responsible for delaying restrictions on dichloromethane (DCM) and France has prioritised glycol ethers (Section 5.1). In some cases, clarifying the inter-relation between different legislative frameworks could facilitate decision-making (Section 5.1). In other instances, co-ordinating the activities of the Directorate-Generals of the European Commission could avoid delays in decision-making. Although much attention has been paid to hazard assessments, regulators and stakeholders must recognise that risk assessment procedures will not be harmonised in the near future. A compromise could be to adopt more exposure-based approaches to regulation. By providing a balancing measure to the necessity for performing in-depth hazard assessments, exposure-driven regulation could
281
Framework for Chemical Risk Management under REACH facilitate decision-making because this approach recognises and uses: (i) the ability of risk management controls to limit a potential ÀÃÆÊ >`]Ê ®Ê Ì
iÊ >Û>>LÌÞÊ vÊ `>Ì>Ê ÌÊ >Ê ÕLiÀÊ vÊ ÃÌ>i
`iÀÊ organisations involved in a variety of risk management programmes. Ultimately, regulators can use a wide variety of risk management measures to reduce exposure levels without the need to agree on all parts of a risk assessment.
7.2.3 Implementing REACH A particular drawback with REACH is the necessity for carrying out regulatory impact assessments before enacting restriction and authorisation. Incorporating mechanisms for the identification of safe, permissible, and tolerable uses into REACH could reduce the need and the scope of conducting socio-economic analyses of regulatory proposals at the EU level. Member States would also retain the option to conduct socio-economic analyses when setting or achieving any EU targets that would be implemented under legislation based on Article 138 or 175 of the EC Treaty. While countries such as France and the UK may wish to focus on the use of statistics on industrial accidents, environmental emissions, occupational health or other data for achieving any targets set under REACH, Germany may opt for setting industrial standards and Sweden may continue with developing more company-specific engagement practices. All EU countries would be responsible for meeting EU targets, thereby promoting a level playing field that accounts for different national approaches. Target-setting would further support the need to harmonise methods for companies to demonstrate regulatory compliance across legislative frameworks, as well as for Member States to report compliance monitoring data during actual decision-making. Apart from the incentive provided by the authorisation process for companies to avoid a small number of substances, EU chemicals policy does not contain many aspects of chemical risk prevention.
282
Conclusion Judging by the interviewee responses, many companies require clear guidance on how to adopt preventive measures, such as coherent and predictable regulation. Companies need a consistent set of regulatory demands that creates a constant and level playing field while promoting best practice. In response, the decision-making rules under the proposed systems framework could provide a sound basis for addressing regulatory inefficiencies created by the EU substance-by-substance regulatory approach. Instead of waiting 11 or more years for the complete implementation of REACH, the rules would apply immediately. The more demanding regulatory recommendations that support the rules would be limited to potential high-level risks (i.e., certain professional and consumer uses). Companies would be granted considerable flexibility for using various risk assessment, management and communication tools to meet the regulatory recommendations. In this way, the proposed scheme avoids limitations of prescriptive ‘top-down’ regulation and seeks to achieve a more integrated approach to chemical risk management. To further balance the ‘top-down’ elements of the REACH regulation, the following ‘bottom-up’ approaches to risk management have been incorporated into the systems framework: 1. Communicating national and EU activities through the Monitoring Network; 2. Incorporating Stakeholder InputsÊÌÊ`iVÃ>}Æ 3. Devising benchmarking schemes under target-setting; 4. Publishing listed uses of chemicals in certain consumer «À`ÕVÌÃÆ 5. Providing regulatory guidance for corporate risk management and regulatory decision-making in the form of recommendations L>Ãi`ÊÊ«ÀiÛÕÃÊi}Ã>ÌÆ
283
Framework for Chemical Risk Management under REACH 6. Constructing a decision-making matrix that compiles previous decisions and therefore facilitates comparisons of regulatory ÕÌViÃÆ 7. Creating a simple and transparent mechanism to prioritise decision-making. These potential ‘bottom-up’ aspects of REACH offered by the systems framework can be important for future relations between different actors at a national level. This could avoid potential strains caused by increasingly centralised European decision-making. For instance, relations could otherwise deteriorate between German regulatory officials and the German Consumer NGO or between the Swedish regulators and the chemical industry. Incorporating ‘bottom-up’ procedures into REACH could compensate for the lack of political attention to the practicalities of compliance monitoring and enforcement. A further finding of the research is that harmonising the methods and formats could support improving compliance with legislation across Europe used for demonstrating and reporting compliance rather than the actual environmental or product standards.
7.2.4 Risk Analysis The findings from the PhD hold important implications for EU regulators and stakeholder organisations. In terms of evaluating their roles and responsibilities in risk management, four major questions have emerged: 1. Do existing management practices promote a culture of responsible chemical use? 2. Are decision-making structures transparent? 3. Are decision-makers accountable? 4. Should a regulator or stakeholder organisation raise public concerns over a potential chemical risk if it cannot provide a potential risk management solution? 284
Conclusion All evidence gathered during the research project indicates that Sweden has created a chemical safety culture in companies and the general public. While some regulators perceive Sweden’s regulatory approach as extreme, it must be understood that policy objectives serve to guide decision-making and corporate activities. It does not necessarily mean that Swedish politicians, regulators or the general population believe that a target is absolutely achievable1 nor does it mean that the regulatory approach is efficient or effective in the long-term2. Achieving a culture of responsibility in German and French industry appears hampered by a lack of incentives to reduce risks beyond regulatory compliance standards, such as through the use of market instruments or workplace insurance schemes. Technical regulation does help to ensure that minimal levels of safety are met. In this respect, the lack of technical regulation means that the UK must contend with exactly how to ensure that basic safety standards are being achieved while examining if its mixed and flexible objective targets can promote best practice within industry3. Decision-making lacks transparency in Sweden and the UK. In Sweden, this results from allowing political objectives to set technical regulatory demands. In the UK, transparency takes the form of political influences entering scientific decision-making. A simple divide between political and scientific regulatory activities therefore makes decision-making comparatively transparent in France and Germany. It follows that personal accountability within a ministry should also be easier to establish in France and Germany. Exactly 1
For instance, Sweden recently adopted an approach to achieve ‘zero’ traffic accident fatalities or serious injuries by 2020 even though politicians and regulators recognise that some road accidents cannot be avoided [577].
2
A ‘zero-risk’ approach such as phasing out the use of a substance can backfire if the regulatory institutions or the individuals working within the institution cannot differentiate between prescriptive and objective elements of policy. In particular, a ‘zero-risk’ approach can cause inefficient and ineffective prioritisation of regulatory and corporate resources (Section 4.4.1).
3
Incidentally, a similar concern was recently voiced by a presenter at a conference of the UK Treasury in the broader context of corporate risk management [578].
285
Framework for Chemical Risk Management under REACH how final political positions are reached remains rather unclear and undefined in all four countries. One may also expect that attempting to assign political accountability for a decision would prove challenging in any country that does not evaluate previous decisions. Finally, should a regulator or stakeholder organisation raise public concerns over a potential chemical risk if it cannot provide a potential risk management solution? This third question raises moral issues that were outside the scope of the research. However, the research has provided a number of examples when regulators or stakeholders may have acted irresponsibly. It is hoped that by highlighting such instances the research will contribute to a wider debate on chemical risk management.
7.3 Global Dimensions of REACH and the Research REACH seeks to address universal challenges on the lack of risk data and the poor quality of many safety data sheets (SDS). With its dominant role in the global chemicals market, the EU could be ideally situated for propagating change through international supply chains. Even the prospect of REACH has already catalysed change within the chemical industry. For example, the Franco-American company Rhodia notes that REACH has reinforced its efforts in developing product stewardship programmes and reviewing SDS [526]. Similarly, the Swedish Chemical and Plastic Industry Federation has responded to REACH by launching a major initiative to improve SDS information supplied by its companies [527]. From the other end of the risk communication chain, the Dutch-American company Rohm and Haas has responded with a scheme for collecting data on the chemical contents of its upstream raw materials [528]. REACH has also prompted countries outside the EU to review their own systems of regulation [529]. Recent proposals in the USA
286
Conclusion have even been described as ‘REACH-like’ by the press [182, 530] and some aspects of the UNEP Strategic Approach to International Chemicals Management (SAICM) resemble REACH [531]. While such developments may be worthwhile, an inefficiently implemented REACH could deter other countries from reforming their chemical regulation or promoting similar schemes at the international level. The success or failure of REACH has important implications for future regulation across the world. Fundamentally, REACH is spurring the international acceptance of new risk assessment procedures, such as the use of in-silico methodologies. From a risk management perspective, regulatory agreement at the EU level can also provide a strong political force for action at an international level. As the only internationally endorsed RRS guidance is based on the current EU document [532], REACH technical guidance documents (TGD) may prove instrumental for the future development of international guidance. Perhaps more importantly, substances subject to authorisation due to their persistent, bioaccumulative and toxic (PBT) characteristics will also be proposed for inclusion on the International Persistent Organic Pollutant (POP) register or other international conventions [533]. The extent to which the administrative resources necessary for REACH will compromise the ability for Member States to engage and support wider risk management activities, such as in developing countries, has not been considered because it lies outside the scope of this dissertation. Nevertheless, it reinforces the importance of creating an efficient management system under REACH.
7.4 Milestones REACH came into force on 1 June 2007. As of early 2009, several TGD for implementing REACH have yet to be published. There does not appear to have been any public discussion on the precise structure of the Agency, the inter-relation of REACH with other legislative
287
Framework for Chemical Risk Management under REACH frameworks or any evaluation of Member State technical capabilities and resource availability to enforce the regulation. While some aspects of the registration procedure have been ‘tested’ through two collaborative initiatives between the European Commission, several Member States and industry (Sections 3.8 and 5.1), the outcomes highlighted the need for a number of regulatory clarifications and technical tools. The potential contribution of these two projects are however rather limited because many of these issues were already anticipated as needing clarification in TGD. As the procedures for restrictions and authorisation have not been tested at the EU level, this thesis concludes that many elements of the implementation of REACH will depend on the functioning of the Agency and the individual Member State Competent Authorities. Drawing out a timeline of how REACH has developed helps to reflect on what could have been done differently. Compared with the sequential key stages of the REACH policy and legislative process summarised in Figure 7.1(a), the results of the thesis would suggest that establishing preliminary TGD on how the REACH process could operate in practice before the White Paper would have facilitated the testing of the registration process and the drafting of the legislation (Figure 7.1(b)). The preliminary TGD could have been based on previous experiences of EU chemical risk management and then finalised in accordance with the Commission proposal and the final legislation. The contrast between Figure 7.1(a) and (b) supports the view expressed by many interviewees that REACH is another example of EU ‘top-down’ legislation. One cannot therefore help but ask whether EU ministers and parliamentarians know what they are voting on. There must be a suspicion that the scientific and technical aspects of the regulation, and hence its practical viability, do not matter to EU politicians.
288
(b)
Conceptualisation of REACH
Conceptualisation of REACH
White Paper
Preliminary Technical Guidance
Commission Draft Proposal (2002)
White Paper
Commission Consultation (2002)
Registration Testing
Commission Proposal (2003)
Commission Draft Proposal
Registration Testing (2005)
Commission Consultation
Preliminary Technical Guidance
Commission Proposal
REACH Regulation (2006)
REACH Regulation
Final Technical Guidance Documents (2007)
Final Technical Guidance Documents
289
Figure 7.1Ê, \Ê>®ÊÃÊÌÊÃÆÊL®ÊÜÊÌÊVÕ`Ê
>ÛiÊLii
Conclusion
(a)
Framework for Chemical Risk Management under REACH
7.5 Research Contribution This research lays a foundation for future work on national approaches to chemical risk management. Three contrasting national approaches emerge from the analysis of the four countries: hazard as the main criterion for risk management in Sweden, technical considerations in France and Germany, and risk-benefit in the UK. A set of sub-criteria was identified as necessity to further define and differentiate possible permutations of each regulatory approach: UÊ Ì
iÊÕÃiÊvÊLiVÌÛiÊversusÊ«ÀiÃVÀ«ÌÛiÊÀi}Õ>ÌÆ UÊ Ì
iÊÕÃiÊvÊÀi}Õ>ÌÀÞÊÌÃÊ>`Ê«VÞÊÃÌÀÕiÌÃÆ UÊ Ì
iÊ`i}ÀiiÊvÊViÌÀ>Ãi`Ê>`ÃÌÀ>ÌÆÊ>`Ê UÊ Ì
iÊ iÝÌiÌÊ vÊ Ãi«>À>ÌÊ LiÌÜiiÊ «ÌV>Ê >`Ê ÃViÌvV decision-making. As many countries are currently reviewing their administrative structures in anticipation of REACH, the experiences of regulation in France, Germany, Sweden and the UK may serve as a series of ‘lessons learnt’ for other countries. With respect to EU decision-making, the project has responded to the limited ability for regulators to address fundamental structural issues in the current risk management process. The proposed systems framework for decision-making demonstrates why restriction and authorisation under REACH should be selected according to several risk criteria, not simply according to hazard and volume. Moreover, regardless how the risk is evaluated, authorisation should not be seen as the most appropriate measure to control high-level risks. Reaching agreement between Member States will often depend on the use of other regulatory options that may result in more effective and efficient decision-making. It is therefore important that this aspect of REACH be communicated to regulators and stakeholder organisations.
290
Conclusion The research appears to be the first to propose: (1) a set of decision-making rules that avoid substance-by-substance Ài}Õ>ÌÆ (2) a set of regulatory recommendations to define liability for V
iV>ÊÕÃiÊ>`ÊÌÊ>VÌÊ>ÃÊ>ÊÌi«>ÌiÊvÀÊvÕÌÕÀiÊi}Ã>ÌÆ (3) a decision-making matrix to compare previous decisions and VÃ`iÀÊvÕÌÕÀiÊÀi}Õ>ÌÀÞÊ«ÌÃÆ (4) a prioritisation scheme for EU chemical regulation that includes ÃV>Ê`iÃÃÊvÊÀÃÆÊ>` (5) a mechanism to control point-source emissions with high potential for cross-border pollution. The systems framework may therefore provide a basis for the further development of such tools. Fundamentally, while the REACH Implementation Programmes (RIP) are providing guidance, they do not appear to be proposing specific methods or tools to facilitate regulatory decision-making. Altogether, it is hoped that the research findings will serve to promote a mutual understanding of the different national approaches and provide insight to countries that have even less practical experience with the EU RRS process. By focussing on regulatory decision-making processes that are independent of the final REACH legislative text, the research has essentially ‘worked backwards’ to (i) identify a number of issues that must be addressed Ê Ì
iÊ }Õ`>ViÊ `VÕiÌÃÊ Ì
>ÌÊ
>ÛiÊ ÞiÌÊ ÌÊ LiÊ V«iÌi`ÆÊ ®Ê develop a number of strategies that regulatory administrations and stakeholder organisations can adopt when preparing to implement , ÆÊ >`Ê ®Ê `iÛi«Ê >Ê >`ÃÌÀ>ÌÛiÊ ÃÌÀÕVÌÕÀiÊ ÜÌ
Ê the European Chemicals Agency to facilitate decision-making. To contribute to current REACH debates, various forms of the research findings were circulated to participants of the research study, several European Commission representatives and certain RIP experts. 291
Framework for Chemical Risk Management under REACH
7.6 Future Research Further avenues of research should focus on developing and communicating guidelines for improving chemical safety. Although a tremendous amount of information on chemical risk management best practice exists in the EU, there is no research that examines the effectiveness and efficiency of EU networks for exchanging and disseminating this experience. The development of the decision-making rules, as proposed under the systems framework, deserves further attention. Regulators should review and agree on safe and permissible uses. Stakeholder organisations, including the representatives of retailers, need to group together and decide on listed uses. Specific consideration should also be given as to how the recommendations can be incorporated into existing best practice documents and environmental management systems. As a separate but related activity, Member State regulators should devise strategies that can assist SME in meeting such regulatory obligations and adopt preventive approaches to risk management. The use of benchmarking in chemical regulation has received little attention at the EU level, but offers a potentially very important avenue for facilitating agreement on RRS among Member States. A set of indicators that weigh levels of protection and prevention according to the types of chemical use could provide a valuable tool during risk assessment and risk management at both national and EU levels. Such indicators could then be used for prioritising regulatory activities necessary to achieve targets or ensure compliance. Inevitably, there will be a need to carry out a regulatory impact assessment on the resources needed to implement REACH. Not only does this have implications on ensuring a level playing field for business across the EU, it may also affect regulatory resources available for promoting international action on chemical control.
292
A
cknowledgements
I would like to express my gratitude to all the regulators and stakeholder representatives that participated in my PhD research project. This thesis would simply not have been possible without their willingness to relate their experiences in chemical risk management to me. The warm welcome that I received in France, Germany, Sweden and the UK while conducting the interviews forms a strong and lasting memory, for which I am especially thankful. The multi-disciplinary approach to the study was endorsed by the Centre for Environmental Strategy in the School of Engineering at the University of Surrey and funded by the UK Engineering and Physical Sciences Research Council. From the conception of this PhD research project, I received dedicated support from my two supervisors, Dr Walter Wehrmeyer and Professor Roland Clift. The dynamic between this duo created a stimulating work environment which allowed me to explore and develop many aspects of my professional career and academic research. In particular, I feel indebted to Walter for the practical experiences relating to project management and I am obliged to Roland for sharing his knowledge and professional contacts. I therefore thank both Walter and Roland for all the opportunities to contribute to research projects and activities during four years at the Centre for Environmental Strategy.
293
R
eferences
1.
CSTEE, Scientific Committee on Toxicity, Ecotoxicity and the Environment, Opinion On Risks To Health From Chromium VI In Cement, Opinion expressed at the 32th CSTEE plenary meeting, C2/AST/csteeop/ChromiumVI 27062002/D(02), Brussels, Belgium.
2.
Reducing Risks, Protecting People: HSE’s Decision-making Process, Her Majesty’s Stationery Office, HSE, UK Health and Safety Executive, Norwich, UK, 2001.
3.
Reducing Risks, Protecting People: HSE’s Decision-making Process, Her Majesty’s Stationery Office, HSE, UK Health and Safety Executive, Norwich, UK, 2001, p.3.
4.
R. Nozick, Anarchy, State, and Utopia, Blackwell Publishers, Oxford, UK, 2001, p.57.
5.
White Paper Strategy for a Future Chemicals Policy, COM(2001) 88 final, CEC, Brussels, Belgium, 2001, p.7.
6.
T.S. Kuhn, The Structure of Scientific Revolutions, 3rd Edition, University of Chicago Press, Chicago, IL, USA, 1996.
7.
P. Hoyningen-Huene, E. Oberheim and H. Andersen, On Incommensurability, Studies in History and Philosophy of Science A, 1996, 27, 131.
8.
D. Pels, The Politics of Symmetry, Social Studies of Science, 1996, 26, 277.
295
Framework for Chemical Risk Management under REACH 9.
A. Tukker, Frames in the Toxicity Controversy: Risk Assessment and Policy Analysis Related to the Dutch Chlorine Debate and the Swedish PVC Debate, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1999, p.21.
10. A. Tukker, Frames in the Toxicity Controversy: Risk Assessment and Policy Analysis Related to the Dutch Chlorine Debate and the Swedish PVC Debate, Kluwer Academic Publishers, Dordrecht, The Netherlands, 1999, p.17. 11. K.R. Popper, The Logic of Scientific Study, 4th Edition, Hutchinson and Co., London, UK, 1972. 12. A. Klinke and O. Renn, A New Approach to Risk evaluation and Management: Risk-based, Precaution-based and Discourse-based Strategies, Risk Analysis, 2002, 22, 1071. 13. J. Adams, Risk, 5th Edition, Routlege, London, UK, 2001, p.9. 14. S.M. Macgill and Y.L. Siu, The Nature of Risk, Journal of Risk Research, 2004, 7, 315. 15. W. Poortinga and N.F. Pidgeon, Exploring the Dimensionality of Trust in Risk Regulation, Risk Analysis, 2003, 23, 961. 16. C.W. Scherer and H. Cho, Social Network Contagion Theory of Risk Perception, Risk Analysis, 2003, 23, 261. 17. C. Lahusen and R. Münch, Introduction: Democratic Politics and Ecological Challenges, In Democracy at Work, A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger Publishers, Westport, CT, USA, 2000, p.18. 18. P. Calow, Controlling Environmental Risks from Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997, p.2.
296
References 19. Chemical Abstact Services, CAS Registry, 2005. http://www.cas.org/EO/regsys.html 20. White Paper Strategy for a Future Chemicals Policy, COM(2001) 88 final, CEC, Brussels, Belgium, 2001. 21. Proposal for a Regulation of the European Parliament and of the Council concerning the Registration, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency and amending Directive 1999/45/EC and Regulation (EC) {on Persistent Organic Pollutants}, COM(2003)644, CEC, Brussels, 2003, p.66. 22. Revised Business Impact Assessment for the Consultation Document, Working Paper 4 - prepared for the European Commission Environment Directorate-General, Risk and Policy Analysts, London, 2003, p.92. 23. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, 2001, p.23. 24. P.W. Satchell and M. Dunwoodies, Pan-European Chemicals: Which Direction for Chemicals? ING Financial Markets, 2004. 25. The Report of the Scientific Steering Committee’s Working Group on Harmonisation of Risk Assessment Procedures in the Scientific Committees Advising the European Commission in the Area of Human and Environmental Health, First Report on the Harmonisation of Risk Assessment Procedures Part 1, CEC, Brussels, Belgium, 2000, p.17. http://www.europa.eu.int/comm/food/fs/sc/ssc/out83_en.pdf 26. C.J. Friedriech, Man and His Government, McGraw-Hill, New York, NY, USA, 1963, p.70.
297
Framework for Chemical Risk Management under REACH 27. J.E. Anderson, D.W. Brady, C.S. Bullock III and J.S. Steward, Jr., Public Policy and Politics in America, Brookes-Cole, Montgomery, CA, USA, 1984, p.4. 28. CEC, Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities, Luxembourg, 1998. 29. P. Calow, Controlling Environmental Risks from Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997, p.3. 30. P. Calow, Controlling Environmental Risks from Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997, p.33. 31. The Environment and the European Community, The Analysis of Political Integration, European Integration and Environmental Policy, Eds., J.D. Liefferink, P. Lowe and A.P.J. Mol, Belhaven Press, London, UK 1997, 2, 1. 32. The Environment and the European Community, The Analysis of Political Integration, European Integration and Environmental Policy, Eds., J.D. Liefferink, P. Lowe and A.P.J. Mol, Belhaven Press, London, UK 1997, 2, 6. 33. A. Tukker, Frames in the Toxicity Controversy, Risk Assessment and Policy Analysis Related to the Dutch Chlorine Debate and the Swedish PVC Debate, Kluwer Academic Publishers, Dordrecht, the Netherlands, 1999. 34. Economic Evaluations in Environmental Policies and Legislation, Final Report prepared for the European Commission Directorate-General Enterprise, RPA, Risk and Policy Analysts Ltd, London, UK, 1998.
298
References 35. EP, Recommendations for a Second Reading. Council common position as amended, Council common position as amended, Regulation (EC) No …/2006 of the Euopeant Parliament and of the Council concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC of the European Parliament and of the Council and repealing Council Regulation (EEC) No 793/93 and Commission Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/EEC, 93/67/EEC, 93/105/EC and 2000/21/ EC, A6-0352/191-191, December 2006, 6. www.europarl.eu.int. 36. Proposal for a Regulation of the European Parliament and of the Council concerning the Registration, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency and amending Directive 1999/45/EC and Regulation (EC) {on Persistent Organic Pollutants}, COM(2003)644, CEC, Brussels, Belgium, 2003. 37. J.H. Jans, European Environmental Law, 2nd Edition, Europa Law Publishing, Groningen, The Netherlands, 2000, p.16. 38. First Report on the Harmonisation of Risk Assessment Procedures Part 1: The Report of the Scientific Steering Committee’s Working Group on Harmonisation of Risk Assessment Procedures in the Scientific Committees Advising the European Commission in the Area of Human and Environmental Health, CEC, Brussels, Belgium, 2000. 39. European Chemical Industry Council, Chemicals Management, Cefic, 2001. 40. W. Sura, The Chemical Industry in the European Union, Statistics in Focus, Industry Trade and Services, 2004, p.47.
299
Framework for Chemical Risk Management under REACH 41. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.53. 42. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.70. 43. Swedish Ministry of the Environment, Sweden’s Offensive Policy of Chemicals, Fact Sheet Number 6.1, ME, Stockholm, Sweden, 1997. 44. The Impact of the Pioneers on EU Environmental Policy, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.1. 45. J. Peterson and E. Bomberg, Decision-Making in the European Union, Macmillan, London, UK, 1999. 46. EC, Regulation (EC) No 1907/2006 of the Euopean Parliament and the Council of 18 December 2006 concerning the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency, amending Directive 1999/45/EC and repealing Council Regulation (EC) No 1488/94 as well as Council Directive 76/769/EEC and Commission Directives 91/155/ EEC, 93/67/EEC, 93/105/EC and 2000/21/EC. 30.12.2006. Official Journal of the European Union, 2006, No L396. 47. ECB Newletter, 2006, 3, 25. 48. Commission of the European Communities, Commission Staff Working Paper, Extended Business Impact Assessment, SEC 1171/3, CEC, Brussels, Belgium, 2003, p.6. 49. Horizon 2015: Perspectives for the European Chemical Industry, Executive Summary of ‘Chemical Industry 2015: Roads to the Future’, CEFIC, Brussels, Belgium, 2004.
300
References 50. REACH – A Leap Forward for Industry: Nordic Concerns and Benefits, Nordic Council of Ministers, Copenhagen, Denmark, 2004, p.7. 51. P. Calow, Can Ecosystems be Healthy? Critical Consideration of Concepts, Journal of Aquatic Ecosystem Health, 1992, 1, 1. 52. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, p.120. 53. C.G. Daughtan, ‘Emerging’ Chemicals in the Environment: A 21st Century Perspective, Renewable Resources, 2005, 23, 4, 6. 54. J.S. Latimer and J. Zheng, The Sources, Transport and Fate of PAHs in the Marine Environment, PAHs: An Ecotoxicological Perspective, Ed., P.E.T. Douben, John Wiley and Sons, Chichester, UK, 2003, p.9. 55. Third European Working Conditions Survey on Working Conditions, Data: Physical Environment, European Working Conditions Observatory, Dublin, Ireland, 2000. http://www.eurofound.eu.int/working/3wc/3wcindex_2.htm. 56. M.D. Anway, A.S. Cupp, M. Uzumcu and M.K. Skinner, Epigenetic Transgenerational Actions of Endocrine Disruptors and Male Fertility, Science, 2005, 308, 1466. 57. J. Kaiser, Endocrine Disrupters Trigger Fertility Problems in Multiple Generations, Science, 2005, 308, 1391. 58. 10 Million Chemical Substances Registered, International Register of Potentially Toxic Chemicals, IRPTC Bulletin 1991, 10, 9. 59. A. Woolley, A Guide to Practical Toxicology :Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, p.230.
301
Framework for Chemical Risk Management under REACH 60. USA Environmental Working Group, Boden Burden: The Pollution in Newborns, Executive Summary, EWG, www.ewg. org 2005. 61. T. Musu, A Trade Union Perspective on REACH, European Trade Union Confederation, REACH Made Easy, European Environmental Bureau Seminar, Brussels, Belgium, 2005. 62. B.N. Ames and L.S. Swirsky Gold, Cancer Prevention and the Environmental Chemical Distraction, Environment and Health: Myths and Realities, Eds., K. Okonski and J. Morris, International Policy Press, London, UK, 2004, 24, 21. 63. Chemicals for Beginners, An Investing Guide, Deutsche Bank AG, London, UK, 2003, p.9. 64. Chemicals for Beginners, An Investing Guide, Deutsche Bank AG, London, UK, 2003, p.13. 65. Bioctechnology for Clean Industrial Products and Processes, OECD Publications, Paris, France, 1998, p.32. 66. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, 2001, p.35. 67. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, 2001, p.115. 68. M.J. Chadwick and J. Nilsson, Environmental Quality Objective – Assimilative Capacity and Critical Load Concepts in Environmental Management, Clean Production Strategies, Developing Preventative Environmental Management in the Industrial Ecology, Ed., T. Jackson, Lewis Publishers, London, UK, 1993, p.29. 69. Arctic Pollution, Persistent Organic Pollutants, Heavy Metals, Radioactivity, Human Health, Changing Pathways, Arctic Monitoring and Assessment Programme, 2002. http://amap.no/documents. 302
References 70. C.A. de Wit, An Overview of Polybrominated Flame Retardants in the Envrionment, Chemosphere, 2002, 46, 583. 71. Contamination: The Results of WWF’s Biomonitoring Survey, WWF-UK National Biomonitoring, WWF-UK, Goldaming, Surrey, UK, 2003. 72. T. Luckenbach and D. Epel, Nitromusk and Polycyclic Musk Compounds as Long-Term Inhibitors of Cellular Xenobiotic Defense Systems Mediated by Multidrug Transporters, Environmental Health Perspectives, 2005, 113, 1, 17. 73. Arctic pollution, Persistent Organic Pollutants, Heavy Metals, Radioactivity, Human Health, Changing Pathways, AMAP, Arctic Monitoring and Assessment Programme, 2002, p.12. http://amap.no/documents 74. Impact on Manufacturing: Chemicals, The Single Market Review Subseries, Office for Official Publications of the European Communities, CEC, Luxembourg, 1997, 1, 5. 75. S. Mahdi, P. Nightingale and F. Berkhout, A Review of the Impact of Regulation on the Chemical Industry, Final Report to the Royal Commission on Environmental Pollution, 2002. www.rcep.org.uk/chemicals.htm. 76. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, 2001, p.26. 77. Assessment of the Business Impacts of New Regulations in the Chemicals Sector Final Report prepared for the European Commission Directorate-General Enterprise, RPA and Statistics Sweden, London, UK, 2002. 78. What We Need from REACH: Views on the proposal for a new chemical legislation within the EU, ChemSec, 2005.
303
Framework for Chemical Risk Management under REACH 79. J. Oldham and T. Votta, Chemical Management Services: Greening the Supply Chain, Greener Management International, 2003, 41, 89. 80. B. Andersen and V. Walsh, Co-evolution within Chemical Technology Systems: A Competence Bloc Approach, Industry and Innovation, 2000, 7, 77. 81. B. Andersen, J.S. Metcalfe and B.S. Tether, Distributed Innovation Systems and Instituted Economic Processes, Innovation Systems in the Service Economy: Measurement and Case Study Analysis, Eds., J.S. Metcalfe and I. Miles, Kluwer, Dordrecht, The Netherlands, 2000, p.15. 82. J.A. Christiansen, Building the Innovative Organisation, Palgrave Macmillan, London, UK, 2000, p.12. 83. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, 2003, p.29. 84. Japanese National Institute of Resources and Environment, 2003. Figure from website at http://www.nire.go.jp/eco_tec_e/ hyouka_e.htm. 85. Conclusions and Recommendations Workshop on the Collection, Reporting and Use of Human Data, International programme on Chemical Safety, Cardiff, UK, 2004. 86. U. Müller-Herold, M. Morosini, O. Schucht and M. Scheringer, The Application of the Precautionary Principle in the European Union: Precautionary Sreening of New Organic Chemicals with Respect to Global Threat Potentials, Eidgenössische Technische Hochschule, Zürich, Switzerland, 2001. 87. A. Long, Predicting Metabolic Fate: The Role of Computer Reasoning in the Generation of Metabolites, Silico ADMET:
304
References Design of Bioactive Compounds, Society of Chemical Industry Conference, London, UK, 2005. 88. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.110. 89. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, p.126. 90. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, 2003, p.253. 91. Globally Harmonized System of Classification and Labelling of Chemicals, UNECE, United Nations Economic Commission for Europe, 2005. 92. Globally Harmonized System of Classification and Labelling of Chemicals, UNECE, United Nations Economic Commission for Europe, 2005, 241. 93. J. Adams, Risk, 5th Edition, Routlege, London, UK, 2001, p.8. 94. D.M. Kammen and D.M. Hassenzahl, Should We Risk It? Exploring Environmental, Health and Technological Problem Solving, 2nd Edition, Princeton University Press, Oxford, NJ, USA, 1999, p.131. 95. P. Calow, Controlling Environmental Risks From Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997, p.23. 96. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, p.58. 97. J. Timbrell, Introduction to Toxicology, 3rd Edition, CRC Press, London, UK, 2002, p.180.
305
Framework for Chemical Risk Management under REACH 98. V.C. Gordon, J. Harvell, M. Bason and H. Maibach, In Vitro Methods to Predict Dermal Toxicity, In Vitro Toxicology, Ed., S.C. Gad, Raven Press, New York, NY, USA, 1994, p.47. 99. S.G. Whittaker and E.M. Faustman, In Vitro Assays for Developmental Toxicity, In Vitro Toxicology, Ed., S.C. Gad, Raven Press, New York, NY, USA, 1994, p.97. 100. E. McIvan, Presentation for the British Union for the Abolition of Vivisection and the European Coalition to End Animal Experiments, New EU Chemical Strategy Stakeholder Consultation Conference, Department for Environment, Food and Rural Affairs, International Conference Centre Birmingham, UK, 2004. 101. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, 58, 66. 102. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, 58, 297. 103. First Report on the Harmonisation of Risk Assessment Procedures Part 1: The Report of the Scientific Steering Committee’s Working Group on Harmonisation of Risk Assessment Procedures in the Scientific Committees Advising the European Commission in the Area of Human and Environmental Health, CEC, Brussels, Belgium, 2000, p.29. 104. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, 58, 24. 105. Commission Directive 2003/3/EC of 6 January 2003 relating to restrictions on the marketing and use of “blue colourant” (twelfth adaptation to technical progress of Council Directive 76/769/EEC), Official Journal of the European Union, No. L. 004 EC, 2003.
306
References 106. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, 58, 303. 107. J. Timbrell, Introduction to Toxicology, 3rd Edition, CRC Press, London, UK, 2002, p.168. 108. P. Calow, Controlling Environmental Risks From Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997, p.20. 109. A. Woolley, A Guide to Practical Toxicology: Evaluation, Prediction and Risk, CRC Press, London, UK, 2003, 58, p.100. 110. J. Timbrell, Introduction to Toxicology, 3rd Edition, CRC Press, London, UK, 2002, p.164. 111. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.25. 112. First Report on the Harmonisation of Risk Assessment Procedures Part 1: The Report of the Scientific Steering Committee’s Working Group on Harmonisation of Risk Assessment Procedures in the Scientific Committees Advising the European Commission in the Area of Human and Environmental Health, CEC, Brussels, 2000, p.85. 113. J. Timbrell, Introduction to Toxicology, 3rd Edition, CRC Press, London, UK, 2002, p.11. 114. Technical Guidance Documents in Support of Commission Directive 93/67/EEC on Risk Assessment for New Notified Substances and Commission Regulation (EC) No 1488/94 on Risk Assessment for Existing Substances, Parts I, II, III and IV. ECSC-EC-EAEC, Office for Official Publications of the European Communities, CEC, Luxembourg, 1996.
307
Framework for Chemical Risk Management under REACH 115. C.J. van Leeuwen, F. Bro-Rasmussen, T.C.J. Feijtel, R. Arndt, B.M. Bussian, D. Calamari, P. Glynn, N.J. Grandy, B. Hansen, J.J. van Hemmen, P. Hurst, N. King, R. Koch, M. Müller, J.F. Solbé, G.A.B. Speijers and T. Vermeire, Risk Assessment and Management of New and Existing Chemicals, Environmental Toxicology and Pharmacology, 1996, 2, 243. 116. C.J. van Leeuwen, H.J.M. Verhaar and J.L.M. Hermens, Quality Criteria and Risk Assessment for Mixtures of Chemicals in the Aquatic Environment, Human and Ecological Risk Management, 1996, 2, 419. 117. E.J. Calabrese and L. Baldwin, Toxicity Mechanisms at Low Doses, Nature, 2003, 421, 691. 118. M.F. Cavieres, J. Jaeger and W. Porter, Developmental Toxicity of a Commercial Herbicide Mixture in Mice: I, Effects on embryo implantation and litter size, Environmental Health Perspecitves, 2002, 110, 1081. 119. S. Paterson and D. Mackay, A steady-state fugacity-based pharmacokinetic model with simultaneous multiple exposure routes, Environmental Toxicological Chemistry, 1987, 6, 395. 120. G.G. Gibson and P. Skett, Introduction to Drug Metabolism, 3rd Edition, Nelson Thornes, London, UK, 2001, p.90 121. R.S. Thomas, D.R. Rank, S.G. Penn, G.M. Zastrow, R.K. Hayes, K. Pande, E. Glover, T. Silander, M.W. Craven, J.K. Reddy, S.B. Jovanovich and C.A. Bradfield, Identification of Toxicologically Predictive Gene Sets using cDNA Microarrays, Molecular Pharmacology, 2001, 60, 1189. 122. J.F. Waring, R. Ciurlionis, R.A. Jolly, M. Heindel and R.G. Ulrich, Microarray Analysis of Hepatotoxins in vitro Reveals a Correlation between Gene Expression Profiles and Mechanisms of Toxicity, Toxicology Letters, 2001, 120, 359.
308
References 123. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.233. 124. D.G. Barnes and M. Dourson, Reference dose (RfD): Descriptions and use in health risk assessments, Regulatory Toxicology and Pharmocology, 1988, 8, 471. 125. J.J. Cairns and D.S. Cherry, Freshwater multi-species test systems, Handbook of Ecotoxicology, Ed., P. Calow, Blackwell Science, Oxford, UK, 1998, p.101. 126. Reference Preliminary Technical Guidance Document, Human health hazard assessment, ECB, 2006, Chapter 3. 127. Principles for the Assessment of Risks to Human Health from Exposure to Chemicals, Environmental Health Criteria 210, International Programme on Chemical Safety, World Health Organisation, Geneva, 1999, p.55. 128. M.F. Weegels and M.P. van Veen, Variation of Consumer Contact with Household Products: A Preliminary Investigation, Risk Analysis, 2001, 21, 499. 129. N.E. Klepeis, W.C. Nelson, W.R. Ott, J.P. Robinson, A.M. Tsang, P. Switzer, J.V. Behar, S.C. Hern and W.H. Engelmann, The National Human Activity Pattern Survey (NHAPS): A Resource for Assessing Exposure to Environmental Pollutants, Journal of Exposure Analysis and Environmental Epidemiology, 2001, 11, 231. 130. Total Human Exposure Assessment Study, FP6, WP2005, Action No. 1312, Institute for Health and Consumer Protection, Joint Research Center (JRC), Ispra (Varese), Italy, 131. J. Timbrell, Introduction to Toxicology, 3rd Edition, CRC Press, London, UK, 2002, p.67.
309
Framework for Chemical Risk Management under REACH 132. Risk Reduction Strategy for Acrylamide, Report prepared for the European Commission Directorate-General Environment, RPA, London, UK, 2001. 133. Environment Institute, Joint Research Centre - European Commission, European Chemicals Bureau, EUSES, European Union System for the Evaluation of Substances, EUR 17308, Ispra (Varese), Italy, 1997. 134. H. Marquart, S. Maidment, J.L. McClaflin and M.C. Fehrenbacher, Harmonization of Future Needs for Dermal Exposure Assessment and Modelling: A Workshop Report, Applied Occupational and Environmental Hygiene, 2003, 16, 218. 135. J.W. Cherrie, J. Tickner, J. Friar, K.S. Creely, A.J. Soutar, G. Hughson, R. Rae, N.D. Warren and D.E. Pryde, Evaluation and Further Development of the EASE Model 2.0, Report Prepared by the Institute of Occupational Medicine, UK Health and Safety Executive Research Report 136, HSE Books, Norwich, UK, 2003. 136. P. Frier, Hazard and Exposure Considerations Related to Chemical Risk Assessment, Greener Management International, 2003, 41, 47. 137. S. Bredendiek-Kämper, Do EASE Scenarios Fit Workplace Reality? - A Validation Study of the EASE Model, Applied Occupational and Environmental Hygiene, 2003, 16, 182. 138. H.J. Bremmer and M.P. van Veen, Children’s Toys Fact Sheets to Assess the Risks to Consumers, RIVM Report No. 61281002/2002, National Institute for Public Health and the Environment, Bilthoven, The Netherlands, 2002. 139. H.J. Bremmer and M.P. van Veen, Children’s Toys Fact Sheets to Assess the Risks to Consumers, RIVM Report No. 61281002/2002, National Institute for Public Health and the Environment, Bilthoven, The Netherlands, 2002, p.2. 310
References 140. Poor Design Practices: Gaseous Emissions from Complex Products, Hamburger Umwelt Institute Report, Hamburg, Germany, 1997. 141. D.M. Kammen and D.M. Hassenzahl, Should We Risk It? Exploring Environmental, Health and Technological Problem Solving, 2nd Edition, Princeton University Press, Oxford, NJ, USA, 1999, p.3. 142. C.J. Jardine and S.E. Hrudey, Mixed Messages in Risk Communication, Risk Analysis, 1997, 4, 489. 143. B. Ballantine, Improving the Quality of Risk Management in the European Union: Risk Communication, The European Policy Centre: Brussels, Belgium, 2003. 144. L.J. Frewer, S. Hunt, M. Brennan, S. Kuznesof, M. Ness and C. Ritson, The Views of Scientific Experts on How the Public Conceptualize Uncertainty, Journal of Risk Research, 2003, 6, 75. 145. K.M. Thompson and D.L. Bloom, Communication of Risk Assessment Information to Risk Managers, Journal of Risk Research, 2000, 3, 333. 146. D.M. Kammen and D.M. Hassenzahl, Should We Risk It? Exploring Environmental, Health and Technological Problem Solving, 2nd Edition, Princeton University Press, Oxford, NJ, USA, 1999. 147. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.96. 148. M. Seigrist and G. Cvertkovich, Better Negative than Positive? Evidence of a Bias for Negative Information about Possible Health Dangers, Risk Analysis, 2001, 21, 199. 149. B.C. Martinson, M.S. Anderson and R. de Vries, Scientists Behaving Badly, Nature, 2005, 435, 737.
311
Framework for Chemical Risk Management under REACH 150. Prospect and Association of University Teachers (AUT), Professional Voices, 2005. 151. Assessment of the Business Impacts of New Regulations in the Chemicals Sector, Final Report prepared for the European Commission Directorate-General Enterprise, RPA, RPA and Statistics Sweden, London, UK, 2002, p.29. 152. Managing Chemicals for a Better Environment: The Environment Agency’s Strategy, EA, Bristol, UK, 2003, p.28. 153. P. Eriksson, E. Jakobsson and A. Fredriksson, Brominated Flame Retardants: A Novel Class of Developmental Neurotoxicants in Our Environment? Environmental Health Perspectives, 2001, 109, 903. 154. DEPA, 2005. 155. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.42. 156. B. Hileman, Panel Ranks Risks of Common Phthalate, Chemical and Engineering News, 2005, 83, 46, p.32. 157. Characterising the Potential Risks Posed by Engineered Nanoparticles, A First UK Government Research Report, Defra, London, UK, 2005. 158. D. Ugarte, Curling and Closure of Graphitic Networks under Electron-Beam Irradiation, Nature, 1992, 359, 707. 159. B. Ballantine, Improving the Quality of Risk Management in the European Union, Risk Communication, The European Policy Centre, Brussels, Belgium, 2003, p.6. 160. A. Amendola, Recent Paradigms for Risk Informed Decision Making, Safety Science, 2001, 40, 17.
312
References 161. D.E. Patton, The ABCs of Risk Assessment. Journal of Environmental Protection and Assessment, 1993, 19, 10. 162. F. Yokota, G. Gray, J.K. Hammitt and K.M. Thompson, Tiered Chemical Testing: A Value of Information Approach, Risk Analysis, 2004, 24, 1625. 163. G. Winter, Redesigning Joint Responsibility of Industry and Government, Risk Assessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, p.178. 164. F.M. Christensen, J.H.M. de Bruijn, B.G. Hansen, S.J. Munn, B. Sokull-Klüttgen and F. Pedersen, New Assessment Tools under the New EU Chemicals Policy, Greener Management International, 2003, 41, 5. 165. Manual for Investigation of HPV Chemicals, Description of OECD Work on Investigation of High Production Volume Chemicals, OECD, Paris, France, 2005. http://www.oecd.org/document/21/0,2340,en_2649_34379_1 939669_1_1_1_1,00.html. 166. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003. 167. B. Ballantine, Improving the Quality of Risk Management in the European Union: Risk Communication, The European Policy Centre, Brussels, Belgium, 2003, p.14. 168. Setting Environmental Standards, 21st Report, Her Majesty’s Stationary Office, RCEP, London, UK, 1998, p.62. 169. EC (1999a) Directive 1999/45/EC of the European Parliament and of the Council of 31 May 1999 concerning the approximation of the laws, regulations and
313
Framework for Chemical Risk Management under REACH administrative provisions of the Member States relating to the classification, packaging and labelling of dangerous preparations. Official Journal of the European Union, No. L. 200, later amended, 1967. 170. R. Allanou, B.G. Hansen and Y. van der Bilt, Public Availability of Data on EU High Production Volume Chemicals, European Commission, Joint Research Centre, Institute for Health and Consumer Protection, European Chemicals Bureau, Ispra (Varese), Italy, 1999. 171. Revised Business Impact Assessment for the Consultation Document, Working Paper 4 - prepared for the European Commission Environment Directorate-General, RPA, London, UK, 2003, p.59. 172. J. Solbé, Project HERA (Human and Environmental Risk Assessment): An Industry Initiative Anticipating the New EU Chemicals Policy, Greener Management International, 2003, 41, 21. 173. B.N. Ames and L.S. Swirsky Gold, Cancer Prevention and the Environmental Chemical Distraction, Environment and Health: Myths and Realities, Eds., K. Okonski and J. Morris, International Policy Press, London, UK, 2004, p.21. 174. R.D. Combes, Prospects for using Data from (Q)SARs and Expert Systems in Integrated Testing Strategies for the Risk Assessment of Chemicals, Silico AMET: Design of Bioactive Compounds, SCI Conference, London, UK, 2005. 175. S.O. Hansson and C. Rudén, Improving the Incentives for Toxicity Testing, Journal of Risk Research, 2003, 6, 3.
314
References 176. European Council for Plasticisers and Intermediates, Inform, 2000, issue 4. 177. K. Harries-Rees, Editorial: Mirror on Chemistry, Chemistry World, December 2004. 178. S.H. Swan, K.M. Main, F. Liu, S.L. Stewart, R.L. Kruse, A.M. Calafat, C.S. Mao, J.B. Redmon, C.L. Ternand, S. Sullivan and J.L. Teague, The Study for Future Families Research Team Decrease in Anogenital Distance among Male Infants with Prenatal Phthalate Exposure, Environmental Health Perspectives, 2005, 113, 8. 179. European Council for Plasticisers and Intermediates, Inform, 2000, issue 4, 4. 180. B. Hileman, Chemical Intolerance, Chemical and Engineering News, 2005, 83, 41, 24. 181. L.H. Li, J.M. Donald and M.S. Golub, Review on Testicular Development, Structure, Function, and Regulation in Common Marmoset, Birth Defects Research, Part B, 2005, 74, 450. 182. M. McCoy, Dry Cleaning Dreams, Chemical and Engineering News, 2005, 83, 19. 183. J. Lohse, L. Lißner, S. Lundie, A. Ahrens and M. Wirts, Never Change a Running Process? Substitution of Hazardous Chemicals in Products and Processes: Definition, Key Drivers and Barriers, Greener Management International, 2003, 41, 57. 184. A. King, Organizational Response to Environmental Regulation: Punctuated Change or Autogenesis? Business Strategy and the Environment, 2000, 9, 224. 185. V. MacDonald and N. Seewald, Household and Personal Care: Awash in New Rules, Chemical and Engineering News, 2005, 167, 25.
315
Framework for Chemical Risk Management under REACH 186. M.S. Reisch, Keeping Well-Preserved, Chemical and Engineering News, 2005, 83, 25. 187. Carcinogens in the Context of Council Directive 96/82/EC, Report by the Technical Working Group 8, Joint Research Centre and European Commission, ISIS, Institute for Systems Informatics and Safety, Seville, Spain, 2000. 188. A. Wegener Sleeswijk, General Prevention and Risk Minimization in LCA: A Combined Approach, Environmental Science and Pollution Research, 2003, 10, 69. 189. O. Mont, The Swedish Product Choice Principle, European Environmental Law Review, 2001, 10, 351. 190. The Impact of the New Chemicals Policy on Health and the Environment, Final Report prepared for the European Commission Environment Directorate-General, Risk and Policy Analysts and BRE Environment, RPA, London, UK, 2003. 191. A.H. Tullo, Cutting Out Phthalates, Chemical and Engineering News, 2005, 83, 43, 29. 192. European Classification and Labelling Inspections of Preparations including Safety Data Sheets, Final Report, Chemical Legislation European Enforcemnet Network (CLEEN), June 2004. www.cleen-europe.eu 193. C. Spelt, Solid Enforcement of New Substances, Proceedings of the Fifth International Conference on Environmental Compliance and Enforcement, 1999, 1, 653. http://www.inece.org/5thvol1/5thTOC.htm. 194. P. Cox, B. Fischhoff, S. Gerrard, J. Niewöhner, N. Pidgeon and D. Riley, Developing a Methodology for Designing Messages about Chemical Risks in the Workplace using
316
References the Mental Models Approach, Report for the UK Health and Safety Executive, University of East Anglia, School of Environmental Sciences, RSU, Norwich, UK, 2003, Reference 4048/R64.061. 195. A. Geyer, G. Kittel, L. Vollebregt, J. Westra and H. Wriedt, Assessment of the Usefulness of Material Safety Data Sheets (MSDS) for SMEs, Final Report SAFE Project, Arbeit and Gesundheit, Hamburg, Germany, 1999, No. SOC 97 201817. 196. Workshop on Exchanging Information across a Chemical Product Chain, OECD Series on Risk Management, Workshop held on 15-16 June 2004, Stockholm, Sweden, OECD Publications, Paris, France, 2004, N18, p.31. 197. KO and KemI, Sweden’s Consumer Agency and Chemicals Inspectorate Goods and Children: The Swedish Chemicals Inspectorate and Consumer Agency Inspection Project 2002/2003, Swedish Consumer Agency, Stockholm, Sweden, 2004. 198. Krüger Consult, Techno-Economic Aspects of Measures to Reduce Water Pollution in Small and Medium Sized Enterprises on the Basis of an Industry and Market Driven Approach. European Commission, Directorate-General XI.C5 Environment, Nuclear Safety and Civil Protection, Luxembourg, 1995, Contract No.B4-3040/93/001170. 199. European Pollutant Emission Register, EPER Review Report June, - Final Report, CEC, 2004, p.25. http://eper.eea.europa.eu/eper 200. Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities, Luxembourg, CEC, 1998, p.10.
317
Framework for Chemical Risk Management under REACH 201. D.M. Riley, B. Fischhoff, M.J. Small and P. Fischbeck, Evaluating the Effectiveness of Risk-Reduction Strategies for Consumer Chemical Products, Risk Analysis, 2001, 21, 357. 202. Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities, CEC, Luxembourg, 1998, p.15. 203. D.F. Hitchens, F. Farrell, J. Lindblom and U. Triebswetter, The Impact of Best Available Techniques (BAT) on the Competitiveness of European Industry, Report EUR 20133, Institute for Prospective Technicological Studies, European Commission Joint Research Centre, Seville, Spain, 2002. 204. D.H. Cole, Pollution and Property, Cambridge University Press, Cambridge, UK, 2002. 205. Good Practice for Dangerous Substances, European Agency for Safety and Health at Work (OSHA), 2005. 206. R. Rühl, E. Lechtenberg-Auffarth and G. Hamm, The Development of Process-Specific Risk Assessment and Control in Germany, Annals of Occupational Hygiene, 2002, 46, 119. 207. The Advantages and Drawbacks of Introducing Communitywide Restrictions on the Marketing and Use of 2-(2butoxyethoxy) ethanol (DEGBE), Revised Final Report, RPA, London, UK, 2001. 208. S. Fineman, Enforcing the Environment: Regulatory Realities, Business Strategy and the Environment, 2000, 9, 62. 209. D.H. Cole, Pollution and Property, Cambridge University Press, Cambridge, UK, 2002, p.41. 210. Council Directive 92/85/EEC of 19 October 1992 on the introduction of measures to encourage improvements in the safety and health at work of pregnant workers and workers
318
References who have recently given birth or are breastfeeding (tenth individual Directive within the meaning of Article 16 (1) of Directive 89/391/EEC), Official Journal of the European Communities, EEC, 1992, No.L.348. 211. Council Directive 94/33/EC of 22 June 1994 on the protection of young people at work, Official Journal of the European Union, EC, 1994, No.L.216. 212. Objectives, Instruments, and Institutions, Objectives, Instruments and Implementation, Environmental Policy: Objectives, Instruments, and Implementation, Ed., D. Helm, Oxford University Press Inc., New York, NY, USA, 2000, 18, 1. 213. New Instruments for Environmental Policy in the EU: Introduction and Overview, New Instruments for Environmental Policy in the EU, Ed., J. Golub, Routledge, London, UK, 1998, p.1. 214. New Instruments for Environmental Policy in the EU: Introduction and Overview, New Instruments for Environmental Policy in the EU, Ed., J. Golub, Routledge, London, UK, 1998, p.4. 215. Implementing Domestic Tradeable Permits: Recent Devleopment and Future Challenges, OECD Publications, Paris, France, 2002. 216. Scope for the use of Economic Instruments for Selected Persistent Pollutants, Final Report prepared for the Department for Environment, Food and Rural Affairs, RPA, London, UK, 2002. 217. E. Gawel, Enforcement of Environmental Charges: Some Economic aspects and Evidence from the German Waste Water Charge, European Environment, 2001, 11, 223.
319
Framework for Chemical Risk Management under REACH 218. Proceedings of the OECD Workshop on Non-Regulatory Initiatives for Chemical Risk Management, Environment, Health and Safety Publications, Series on Risk Management, OECD Publications, Paris, France, 1997, No.7. 219. Environmental Agreements: Environmental Effectiveness, Environmental Issues Series, Environmental Issue Report, EEA, Copenhagen, Denmark, 1997, 1, 3. 220. Commission Communication on Environmental Taxes and Charges in the Single Market 97/C 224/04, Official Journal of the European Communities, CEC, 1997, No.C. 224/6. 221. OECDEnvironmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, OECD, 2001, p.89. 222. Commission Directive 2003/3/EC of 6 January 2003 relating to restrictions on the marketing and use of “blue colourant”, (twelfth adaptation to technical progress of Council Directive 76/769/EEC), Official Journal of the European Union, EC, 2003, No.L.004. 223. Commission Decision of 17 July 2000 on the implementation of a European pollutant emission register (EPER) according to Article 15 of Council Directive 96/61/EC concerning integrated pollution prevention and control (IPPC), Official Journal of the European Union, EC, 2000, No.L.192. 224. S. Waygood and A. Morley, How Does Your Business Measure Up? Developing a Corporate Health and Safety Performance Indicator, Responsible Investment, Eds., R. Sullivan and C. Mackenzie, Greenleaf Publishing, London, UK, 2006, p.62. 225. R. Hillary, Evaluation of Study Reports on the Barriers, Opportunities and Drivers for Small and Medium Sized Enterprises in the Adoption of Environmental Management Systems, UK Department of Trade and Industry, Environment Directorate, London, UK, 1999. 320
References 226. Regulation (EC) No 761/2001 of the European Parliament and of the Council of 19 March 2001 allowing voluntary participation by organisations in a Community ecomanagement and audit scheme (EMAS), Official Journal of the European Union, EC, 2001, No.L.114. 227. D. Bradford, A. Gouldson, J. Hemmelskamp, H. Kottmann and A. Marsanich, The Impact of the EU Eco-Audit Regulation on Innovation in Europe, Ed., L. Delgado, Report for the Institute for Prospective Technological Studies EUR 19722 EN, European Commission Joint Research Centre, Seville, Spain, 2000. 228. Communication from the Commission to the European Parliament, the Council, the European Economic and Social Committee and the Committee of Regions on the Practical Implementation of the Provisions of the Health and Safety at Work Directives 89/391 (Framework), 89/654 (Workplaces), 89/655 (Work Equipment), 89/656 (Personal Protective Equipment), 90/269 (Manual Handling of Loads) and 90/270 (Display Screen Equipment), COM (2004) 62 final, CEC, Brussels, Belgium, 2004. 229. Regulation (EC) No 1980/2000 of the European Parliament and of the Council of 17 July 2000 on a revised Community eco-label award scheme, Official Journal of the European Union, EC, 2000, No L. 237. 230. R. Schwarze, Environmental Liability and Accident Prevention: Preliminary Experience in Germany, European Environment, 2001, 11, 314. 231. A.G. Heyes, Making Things Stick: Enforcement and Compliance, Environmental Policy: Objectives, Instruments, and Implementation, Ed., D. Helm, Oxford University Press Inc, New York, NY, USA, 2000, p.91.
321
Framework for Chemical Risk Management under REACH 232. Environmentally Related Taxes in OECD Countries: Issues and Strategies, OECD Publications, Paris, France, 2001. 233. Catalogue of Economic Incentive Systems for the Improvement of the Working Environment, Office for Official Publications of the European Communities, EFILWC, European Foundation for the Improvement of Living and Working Conditions Luxembourg, 1994. 234. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.91. 235. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.94. 236. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.134. 237. A. Gendebien, A. Leavens, A. Blackmore, A. Godley, K. Lewin, B. Franke and A. Franke, Study on Hazardous Household Waste (HHW) with a Main Emphasis on Hazardous Household Chemicals (HHC), Report for the European Commission, Environment Directorate-General, Brussels, Belgium, 2002. 238. W. Wehrmeyer, S. Erler and G. Johnstone, From Processes to Products: Regulation, Competitiveness and the Future of EHS, Final Report June 2004, Centre for Environmental Strategy, University of Surrey, Guildford, UK, 2004. 239. Organisation for Economic Co-operation and Development, The State of the Environment, OECD Publications, Paris, France, 1991, p.192.
322
References 240. Objectives, Instruments, and Institutions. In: Environmental Policy: Objectives, Instruments, and Implementation, Ed., D. Helm, Oxford University Press Inc, New York, NY, USA, 2000, 103, 1. 241. European Pollutant Emission Register, EPER Review Report June CEC, Final Report, 2004. http://eper.eea.europa.eu/eper 242. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, 2001. 243. Directive 2002/95/EC of the European Parliament and of the Council of 27 January 2003 on the restriction of the use of certain hazardous substances in electrical and electronic equipment, Official Journal of the European Communities, EC, 2002, No.L.114. 244. Directive 2000/53/EC of the European Parliament and of the Council of 18 September 2000 on end-of life vehicles, Official Journal of the European Union, EC, 2003, No.L.260, 2000. 245. European Environment Agency, Environmental Taxes: Implementation and Environmental Effectiveness, Environmental Issues Series, Environmental Issue Report, Copenhagen, Denmark, 1996, p.1. 246. D. Slunge and T. Sterner, Implementation of Policy Instruments for Chlorinated Solvents, A Comparison of Design Standards, Bans and Taxes to Phase Out Trichloroethylene, European Environment, 2001, 11, 281. 247. Communication from the Commission to the Council and the European Parliament on Integrated Product Policy: Building on Life-Cycle Thinking, COM(2003) 302 final, CEC, Brussels, Belgium, 2003.
323
Framework for Chemical Risk Management under REACH 248. J.H. Jans, European Environmental Law, 2nd Edition, Europa Law Publishing, Groningen, The Netherlands, 2000, p.14. 249. J. Weirs, Trade and Environment in the EC and the WTO, Europa Law Publishing, Groningen, The Netherlands, 2002, p.14. 250. Guide to the Implementation of Directives Based on the New Approach and the Global Approach, CEC, Brussels, Belgium, 1999. http//europa.eu.int/comm/enterprise/newapproach/legislation/ guide/document/1999_1282_en.pdf 251. L. Krämer, Precaution, the Protection of Health and the Environment, and the Free Circulation of Goods Within the European Union, The Role of Precaution in Chemicals Policy, Eds., E. Freytag, T. Jackl, G. Loibl and M. Wittmann, Diplomatische Akademie, Vienna, Austria, 2001, p.47. 252. European Court of Justice, Case C-473/97, ECR, 2000, 5681. 253. L. Krämer, Introduction into the European chemicals regulation: Basic Structures and Performance, Risk Assessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000. 254. J. Weirs, Trade and Environment in the EC and the WTO, Europa Law Publishing: Groningen, The Netherlands, 2002, p.202. 255. Commission Communication on Precautionary Principle, COM(2000) 1 Final, CEC, Brussels, Belgium, 2000, www. cmsforum.org p.15.
324
References 256. P. Calow, Controlling Environmental Risks from Chemicals: Principles and Practice, John Wiley and Sons, Chichester, UK, 1997, p.5. 257. A. Tukker, Risk Analysis, Life Cycle Assessment - The Common Challenge of Dealing with the Precautionary Frame, Risk Analysis, 2002, 22, 821. 258. P. Checkland and J. Scholes, Soft Systems Methodology in Action, John Wiley and Sons, Chichester, UK, 1990, p.36. 259. R. Münch, Cultures of Democracy: Historical Formation and Contemporary Challenges, Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, USA, 2000, p.155. 260. C. Lahusen and R. Münch, Introduction: Democratic Politics and Ecological Challenges, Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger Publishers, Westport, CT, USA, 2000, 6, p.18. 261. K. Geiser and J. Tickner, (2003) New Directions in European Chemicals Policies: Drivers, Scope and Status, Lowell Center for Sustainable Production, Lowell, MA, USA, 2003. 262. R. Löfstedt, Swedish Chemical Regulation: An Overview and Analysis, Risk Analysis, 2003, 23, 2, 411. 263. Council Directive 91/689/EEC of 12 December 1991 on hazardous waste, Official Journal of the European Communities, EEC, 1991, NoL.377. 264. Council Directive 90/394/EEC of 28 June 1990 on the protection of workers from the risks related to exposure to carcinogens at work (Sixth individual Directive within the
325
Framework for Chemical Risk Management under REACH meaning of Article 16 (1) of Directive 89/391/EEC), Official Journal of the European Communities, EEC, 1990, NoL.196. 265. Council Directive 76/464/EEC of 4 May 1976 on pollution caused by certain dangerous substances discharged into the aquatic environment of the Community. Official Journal of the European Communities, EEC, 1976, No.L.129/23. 266. R. Donkers, First Reading Approval of REACH in the EU Parliament and by the Council of Ministers, CMS Insider, 2005. 267. Two Decades of Responsible Care: Credible Response or Comfort Blanket? ENDS Report, 2005, 360, 19. 268. REACH Questions and Answers, Future EU Policy, Defra, 2006. http://www.defra.gov.uk/environment/chemicals/reach/qanda/ index.htm. 269. Directive 98/34/EC of the European Parliament and of the Council of 22 June 1998 laying down a procedure for the provision of information in the field of technical standards and regulations, Official Journal of the European Union, EC, 1998, No.L.204. 270. Political Agreement on REACH, Summary Text, Defra, 2005, p.3. www.defra.gov.uk 271. Political Agreement on REACH, Summary Text, Defra, 2005. www.defra.gov.uk 272. European Parliament legislative resolution on the proposal for a regulation of the European Parliament and of the Council on the Registration, Evaluation, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency and amending Directive 1999/45/EC and Regulation (EC) No …/… [on Persistent Organic Pollutants]
326
References (COM(2003)0644 - C5-0530/2003 - 2003/0256(COD)) P6_ TA-PROV(2005)0434, EP, European Parliament, 2005. 273. F. Pedersen, J. de Bruijn, S. Munn and K. van Leeuwen, Assessment of Additional Testing Needs under REACH: Effects of (Q)SARS, Risk Based Testing and Voluntary Industry Initiatives, European Commission Joint Research Centre, Institute for Health and Consumer Protection, 2003. http://ihcp.jrc.ec.europa.eu 274. H. Tyle, H.S. Larsen, E.B. Wedebye, D. Sijm, T. Pedersen Krog and J. Niemela, Identification of potential PBTs and vPvBs by use of QSAR, Danish Environmental Protection Agency document QSAR PBT rev8, Copenhagen, Denmark, 2002. 275. Assessment of the Business Impacts of New Regulations in the Chemicals Sector Phase 2: Availability of Low Value Products and Product Rationalisation, Final Report prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, UK, 2003, p.9. 276. Revised Business Impact Assessment for the Consultation Document, European Commission Environment Directorate-General, Risk and Policy Analysts, London, UK, 2003, p.4. 277. Proposal for a Regulation of the European Parliament and of the Council concerning the Registration, Authorisation and Restriction of Chemicals (REACH), establishing a European Chemicals Agency and amending Directive 1999/45/EC and Regulation (EC) {on Persistent Organic Pollutants, COM(2003)644. Brussels, Belgium, CEC, 2003, p.71. 278. J. Canton and C.H. Allen, A Microeconomic Model to Assess the Economic Impacts of the EU’s New Chemicals Policy, Commission of the European Communities: Brussels, Belgium, 2003.
327
Framework for Chemical Risk Management under REACH 279. Feasibility Study on the Resource Requirements for a Central Entity, Report for the Commission of the European Communities, Ref. B4-3040/2001/329289/MAR/C3, Deloitte and Touche, Brussels, Belgium, 2002. 280. Directive 2000/60/EC of the European Parliament and of the Council of 23 October 2000 establishing a framework for Community action in the field of water policy, Official Journal of the European Union, EC, 2000, No.L327. 281. J. Steiner and L. Woods, Textbook on EC Law, 8th Edition, Oxford University Press, Oxford, UK, 2003, p.481. 282. International Chemical Secretariat, Cry Wolf – Predicted Costs by Industry in the face of New Regulations, ChemSec, Gothenburg, Sweden, 2004, Report 6.04, p.17. 283. International Chemical Secretariat, Cry Wolf – Predicted Costs by Industry in the face of New Regulations, ChemSec, Gothenburg, Sweden, 2004, Report 6.04. 284. G. Haq, P.D. Bailey, M. Chadwick, J. Forrester, J. Kuylenstierna, G. Leach, D. Villagrasa, M. Fergusson, I. Kinner and S. Oberthur, Determining the Costs to Industry of Environmental Regulation, European Environment, 2001, 11, 125. 285. REACH – A Leap Forward for Industry: Nordic Concerns and Benefits, Nordic Council of Ministers (NCM), 2004. 286. Study of the Impact of the Future Chemicals Policy: Additional Study following the Regulation Proposal of 29 October 2003, Study Prepared for UIC (French Union for the Chemicals Industry), Mercer Management Consulting and NERA Economic Consulting, Paris, France, 2004.
328
References 287. EU Officials Seek to Make Sense of 36 REACH Reports, 3 November, EurActiv, 2004. www.euractiv.com/Article?tcmuri=tcm:29-13172516andtype=news. 288. S. Waygood, S. Erler, W. Wehrmeyer and H. Jeswani, Integrated Investment Analysis: Investment Implications of the REACH Regulation, Responsible Investment Eds., R. Sullivan and C. Mackenzie, Greenleaf Publishing, London, 2006, p.322. 289. M. Fleischer, Documented personal communication. 290. K. Pflug and S. Rüdiger, Active Pricing - The Smart Way to Higher Profits, Chemicals Industry, ADL International, 2003. 291. Chemicals for Beginners: An Investing Guide, Deutsche Bank AG, London, UK, 2003, p.23. 292. J.L. Gross and S. Rayner, Measuring Cultures: A Paradigm for the Analysis of Social Organization, Columbia University Press, New York, NY, USA, 1985. 293. P. Checkland and J. Scholes, Soft Systems Methodology in Action, John Wiley and Sons, Chichester, UK, 1990. 294. V. Minichiello, R. Aroni, E. Timewell and L. Alexander, In-Depth Interviewing: Principles, Techniques, Analysis, Longman Pry, Melbourne, Australia, 1995, p.84. 295. V. Minichiello, R. Aroni, E. Timewell and L. Alexander, In-Depth Interviewing: Principles, Techniques, Analysis, Longman Pry, Melbourne, Australia, 1995, p.80. 296. M. Abrahamson, Social Research Methods, Prentice Hall. Englewood Cliffs, NJ, USA, 1983, p.339. 297. V. Minichiello, R. Aroni, E. Timewell and L. Alexander, In-Depth Interviewing: Principles, Techniques, Analysis, Longman Pry, Melbourne, Australia, 1995, p.85.
329
Framework for Chemical Risk Management under REACH 298. A.N. Oppenheim, Questionnaire Design, Interviewing and Attitude Measurement, Pinter Publishers, London, UK, 1992, p.65. 299. S. Titscher, M. Meyer and E. Vetter, Methods of Text and Discourse Analysis, Sage Publications, London, 2000, p.55. 300. S. Titscher, M. Meyer and E. Vetter, Methods of Text and Discourse Analysis, Sage Publications, London, UK, 2000, p.94. 301. V. Minichiello, R. Aroni, E. Timewell and L. Alexander, In-Depth Interviewing: Principles, Techniques, Analysis, Longman Pry, Melbourne, Australia, 1995, p.94. 302. Council Directive 76/769/EEC of 27 July 1976 on the approximation of the laws, regulations and administrative provisions of the Member States relating to restrictions on the marketing and use of certain dangerous substances and preparations, Official Journal of the European Communities, EEC, 1976, No.L.262. 303. European Chemical Substances Information System, ECB, 2006. http://ecb.jrc.it/esis. 304. Commission Recommendation of 29 April 2004 on the results of the risk evaluation and the risk reduction strategies for the substances: acetonitrile; acrylamide; acrylonitrile; acrylic acid; butadiene; hydrogen fluoride; hydrogen peroxide; methacrylic acid; methyl methacrylate; toluene; trichlorobenzene (2004/394/EC) 30.4.2004, Official Journal of the European Union, EC, 2004, No.L.144. 305. Commission Recommendation of 4 July 2002 on the results of the risk evaluation and the risk reduction strategies for the substances: o-anisidine, 1,4-dioxane (2002/575/EC), Official Journal of the European Union, EC, 2002, No.L.181/29.
330
References 306. Commission Recommendation of 4 July 2002 on the results of the risk evaluation for the substances: ethyl acetoacetate, 4-chloro-o-cresol, dimethyldioctadecylammonium chloride, (2002/576/EC) Official Journal of the European Union, EC, 2002, No.L.181. 307. Commission Recommendation of 16 September 2002 on the results of the risk evaluation and the risk reduction strategy for the substance: diphenyl ether, octabromo derivative (notified under document number C(2002) 3394) (Text with EEA relevance) (2002/755/EC), Official Journal, EC, 2002, No.L.249. 308. Commission Recommendation of 5 March 2001 on the results of the risk evaluation and the risk reduction strategies for the substances: diphenylether/pentabromo derivative and cumene (2001/194/EC), Official Journal of the European Union, EC, 2001, No.L.069. 309. Commission Recommendation of 7 November 2001 on the risk evaluation and on the risk reduction stragies for the substances: acrylaldehyde; dimethyl sulphate, nonylphenol phenol, 4-nonyl-, branched; tert-butyl methyl ether (2001/838/EC) 4.12.2001), Official Journal of the European Union, EC, 2001, No.L.319. 310. Commission Recommendation of 12 October 1999 on the results of the risk evaluation and the risk reduction strategies for the substances: 2-(2-butoxyethoxy)ethanol; 2-(2methoxyethoxy)ethanol; Alkanes, C10-13, chlorobenzene, C10-13-alkyl derives, (1999/721/EC), Official Journal of the European Union, EC, 1999, No.L.292. 311. J. de Bruijn, B. Hansen and S. Munn, Role of the Precautionary Principle in the EU Risk Assessment Process on Industrial Chemicals, 2003, 75, 11-12, 2535.
331
Framework for Chemical Risk Management under REACH 312. Risk Reduction Strategy for DEHP, Draft January 2003, KemI, Solna, Sweden, 2003. 313. Proposal for a Directive of the European Parliament and of the Council amending for the 22nd time Directive 76/769/ EEC on the approximation of the laws, regulations and administrative provisions of the Member States relating to restrictions on the marketing and use of certain dangerous substances and preparations (phthalates) and amending Council Directive 88/378/EEC on the approximation of the laws of the Member States concerning the safety of toys, COM 1999 577, CEC, Brussels, Belgium, 1996. 314. Trilateral Letter from Blair, Chirac and Schröder, Dated 8 September 2003. 315. A. Ahlers, D. Jepsen, K. Heitmann and F. Claus, The SPORT Report, Making REACH Work in Practice, Final Report. Ökopol and iku GmbH publication, Hamburg, Germany, 2005. 316. Strategy on Management of Substances, Accepted by the Cabinet on March 16th 2001, TK, Dutch Parliament, The Hague, The Netherlands, 2001. 317. VROM, Spatial Planning and the Environment, Dutch Substances Policy in an International Perspective, Memorandum on implementation of SOMS, Dutch Ministry of Housing, The Hague, The Netherlands, 2004. www.vrom.nl/international. 318. C. Lees, Environmental Policy: The Law of Diminishing Returns? PAS Annual Conference, University of Lincoln, Lincoln, UK, 2004. 319. K. Hentschel, Chemicals Industry Sector Analysis, Industry Canada, 2003, Publication ID: 110184. www.ic.gc.ca
332
References 320. UK Chemical Industry, Facts and Figures 2006, Chemical Industries Association (CIA), London, UK, 2006. 321. Profile of French Chemical Industry, Key Figures, Union des Industries Chimiques (UIC), 2005. 322. Chemical and Plastic Industry in Sweden, Facts and Figures, Plast- and Kemiföretagen, Stockholm, Sweden 2003. 323. Swedish National Chemicals Inspectorate, Management of Chemicals and Pesticides in Sweden, KemI, National Profile, 2000. 324. Eurostat New Release, 105/2004 - 31 August 2004, European demography in 2003, EU25 population up by 0.4% to reach 456 million, One in 14 people in world live in the EU25, Eurostat, 2004. 325. Swedish Industry, SI, Swedish Institute, 2003, Swedish Institute Fact Sheet No. 124. 326. D. Bunke, Indicators on Chemicals: Impacts, Sources and Policy Performance, Sustainable Chemistry - Integrated Management of Chemicals, Products and Processes Conference, Dessau, Germany, 2004. 327. Facts and Figures: The European Chemical Industry in a Worldwide Perspective, CEFIC, 2004. 328. U. Stirba, U. Kowalski and U. Schlottmann, National Profile: Chemicals Management in Germany, Schriftenreihe der Bundesanstalt Für Arbeitsschutz und Arbeitsmedizin (German Institute of Occupational Health and Medicine), Dortmund/ Berlin, Germany, 2000. 329. Profil Français pour Evaluer les Capacités Nationales de Gestion des Produis Chimiques, République Française (RF), 2000.
333
Framework for Chemical Risk Management under REACH 330. S. Milmo, Rise After Storm, Chemistry and Industry, 2005, 24, p.15. 331. Chemicals for Beginners: An Investing Guide, Deutsche Bank AG, London, UK, 2003, p.23. 332. Swedish Industry, Swedish Institute, 2003, Swedish Institute Fact Sheet No.124. 333. Biotechnology and Pharmaceuticals in Sweden, Swedish Institute, 2003, Swedish Institute Fact Sheet No.132. 334. UK Chemical Industry, Chemical Industries Association, London, UK, 2001. 335. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, 2001, p.111. 336. Environmental Data Compendium, OECD Publications, Paris, France, 1999, p.321. 337. Environmental Outlook for the Chemicals Industry, OECD Publications, Paris, France, 2001, p.27. 338. C. Borgards, France: Rule by Virtue of Elite Position. In: Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, USA, 2000, p.73. 339. Environmental Performance Reviews: Germany, OECD Publications, Paris, France, 2001, p.114. 340. W. Cremer and A. Fisahn, New Environmental Policy Instruments in Germany, New Instruments for Environmental Policy in the EU, Ed., J. Golub, Routledge, London, UK, 1998, p.56. 341. B. Wahlström and B. Lundqvist, Risk Reduction and Chemical Control: Lessons from the Swedish Chemicals
334
References Action Programme, Clean Production Strategies: Developing Preventative Environmental Management in the Industrial Ecology, Ed., T. Jackson, Lewis Publishers, London, UK, 1993, p.237. 342. Economic Evaluations in Environmental Policies and Legislation, Final Report prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, UK, 1998, p.13. 343. Britain in Europe: Themes and Issues in National Environmental Policy, British Environmental Policy and Europe: Politics and Policy in Transition, Eds., P. Lowe and S. Ward, Routledge, London, UK, 2998, p.24. 344. Environmental Charter Adopted. Environment, Planning and Regulatory News, Freshfields Bruckhaus Deringer, 2005, p.1. 345. The Impact of the Pioneers on EU Environmental Policy, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.19. 346. Environmental Performance Reviews: Germany, OECD Publications, Paris, France, 2001, p.66. 347. A. Kronsell, Sweden: Setting a Good Example, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.40. 348. Swedish Chemical Policy Committee Towards a Sustainable Chemicals Policy, En Hållbar Kemikaliepolitick. Riskpanorama för kemikalier (with a summary in English), Fritzes, Offentliga Publilationer, Stockholm, Sweden, 1997. 349. D. Walters, Working Safely in Small Enterprises in Europe, European Trade Union Confederation, Brussels, Belgium, 2002, p.112. 335
Framework for Chemical Risk Management under REACH 350. Summary of: Documentation for In-Depth Evaluation of the Environmental Quality Objective of a Non-Toxic Environment, Ref no.342-531-03. KemI, Solna, Sweden, 2003, p.9. 351. The Swedish Environmental Objectives: A Summary of the Government Bill, ME, 1997/98, p.145. 352. Managing Chemicals for a Better Environment: The Environment Agency’s Strategy, Environment Agency (EA), Bristol, UK, 2003. 353. The Government’s Chemicals Strategy: Sustainable Production and Use of Chemicals - A Strategic Approach, UK Department of the Environment, Transport and the Regions, DETR, London, UK, 1999. 354. Cultures of Democracy: Historical Formation and Contemporary Challenges, Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, USA, 2000, p.160. 355. P. Rawcliffe, Environmental Pressure Groups in Transition, Manchester University Press, Manchester, UK, 1998, p.53. 356. J. Jordan, The Impact on UK Environmental Administration, British Environmental Policy and Europe: Politics and Policy in Transition, Eds., P. Lowe and S. Ward, Routledge, London, UK, 1998, p.183. 357. The Impact of the Pioneers on EU Environmental Policy, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.6. 358. P. Rawcliffe, Environmental Pressure Groups in Transition, Manchester University Press, Manchester, UK, 1998, p.54.
336
References 359. Economic Evaluations in Environmental Policies and Legislation, Final Report prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, UK, 1998, p.11. 360. Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities: Luxembourg, CEC, 1998, p.93. 361. Economic Evaluations in Environmental Policies and Legislation, Final Report prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, 1998, p.15. 362. P. Hinchcliffe, The Role of Precaution in Chemicals Policy - The UK Approach, The Role of Precaution in Chemicals Policy, Eds., E. Freytag, T. Jackl, G. Loibl and M. Wittmann, Diplomatische Akademie, Vienna, Austria, 2001, p.70. 363. The Swedish Society for Nature Conservation and Ecolabelling, Swedish Society for Nature Conservation, 2003. 364. D. Walters, Working Safely in Small Enterprises in Europe, European Trade Union Confederation, Brussels, Belgium, 2002, p.113. 365. J. Tickner and K. Geiser, New Directions in European Chemicals Policies: Drivers, Scope and Status, Lowell Center for Sustainable Production, Lowell, MA, USA, 2003, p.43. 366. U. Stirba, U. Kowalski and U. Schlottmann, National Profile: Chemicals Management in Germany, Schriftenreihe der Bundesanstalt Für Arbeitsschutz und Arbeitsmedizin (German Institute of Occupational Health and Medicine), Dortmund/ Berlin, Germany, 2000, p.68.
337
Framework for Chemical Risk Management under REACH 367. U. Stirba, U, Kowalski and U. Schlottmann, National Profile: Chemicals Management in Germany, Schriftenreihe der Bundesanstalt Für Arbeitsschutz und Arbeitsmedizin (German Institute of Occupational Health and Medicine), Dortmund/ Berlin, Germany, 2000, p.74. 368. C. Borgards, France: Rule by Virtue of Elite Position, Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, 2000, p.74. 369. A. Kronsell, Sweden: Setting a Good Example, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.49. 370. H. Pehle, Germany: Domestic Obstacles to an International Forerunner, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.177. 371. Pollution Handbook: The Essential Guide to UK and European Pollution Control Legislation, National Society for Clean Air and Environmental protection, Brighton, UK, 1997, p.5. 372. Pollution Handbook: The Essential Guide to UK and European Pollution Control Legislation, National Society for Clean Air and Environmental protection, Brighton, UK, 1997, p.19. 373. J. Skea and A. Smith, Integrating Pollution Control, British Environmental Policy and Europe: Politics and Policy in Transition, Eds., P. Lowe and S. Ward, Routledge, London, UK, 1998, p.270. 374. S. Tindale and C. Hewett, New Environmental Policy Instruments in the UK, New Instruments for Environmental Policy in the EU, Ed., J. Golub, Routledge, London, UK, 1998, p.34. 338
References 375. H.K. Colebatch, Policy, 2nd Edition, Open University Press, Milton Keynes, UK, 2002, p.14. 376. J.J. Richardson, Policy Styles in Western Europe, Allen and Unwin, London, UK, 1982, p.13. 377. H.K. Colebatch, Policy, 2nd Edition, Open University Press, Milton Keynes, UK, 2002, p.13. 378. H.K. Colebatch, Policy, 2nd Edition, Open University Press, Milton Keynes, UK, 2002, p.1. 379. Occupational Exposure Limit Values and Measures Against Air Contaminants, Statute Book of the Swedish National Board of Occupational Safety and Health, AFS, Arbetsmiljoverket, Solna, Sweden, 2000, p.3. 380. R. Nilsson, Misguided Precaution: Chemicals Control and the Precautionary Principle in Sweden, Perilous Precaution: The Folly of Disregarding Science, The European Science and Environment Forum, Cambridge, UK, 2002, p.6. 381. V. Palm and K. Jonsson, Including Chemical Products in Environmental Accounts. Eurostat Working Papers Prepared for DG Environment and Eurostat by Statistics Sweden 2/2001/B/7 22, Statistics Sweden, Stockholm, Sweden, 2001. 382. J. Arrowsmith, K. Sisson and P. Marginson, What Can ‘Benchmarking’ Offer the Open Method of Co-ordination? Journal of European Public Policy, 2004, 11, 311. 383. P. Hampton, Reducing Administrative Burdens: Effective Inspection and Enforcement, HM Treasury, UK, 2004. 384. J. Tickner and K. Geiser, New Directions in European Chemicals Policies: Drivers, Scope and Status, Lowell Center for Sustainable Production, Lowell, MA, USA, 2003, p.134.
339
Framework for Chemical Risk Management under REACH 385. Draft Minutes of a meeting held on the 2-3 September 2004 at Church House, Royal Commission on Environmental Pollution (RCEP), London, UK, 2004. 386. U. Stirba, U. Kowalski and U. Schlottmann, National Profile: Chemicals Management in Germany, Schriftenreihe der Bundesanstalt Für Arbeitsschutz und Arbeitsmedizin (German Institute of Occupational Health and Medicine), Dortmund/ Berlin, Germany, 2000, p.22. 387. J. Scherer, Environmental Regulation in the Federal Republic of Germany, Regulating the European Environment, Ed., T. Handler, 2nd Edition, John Wiley and Sons, London, UK, 1997, p.101. 388. P. Welfens, The German Economy and EU Eastern Enlargement, American Institute for Contemporary German Studies at Johns Hopkins University, Washington, DC, USA, 2004. 389. Priorities and Strategies in Occupational Safety and Health Policy in the Member States of the European Union, European Agency for Safety and Health at Work, OSHA, Bilbao, Spain, 2000, p.19. 390. Techno-Economic Aspects of Measures to Reduce Water Pollution in Small and Medium Sized Enterprises on the Basis of an Industry and Market Driven Approach, Contract No, B4-3040/93/001170. European Commission, DirectorateGeneral XI.C5 Environment, Nuclear Safety and Civil Protection, Krüger Consult, Luxembourg, 1995, p.29. 391. D. Walters, Working Safely in Small Enterprises in Europe, European Trade Union Confederation, Brussels, Belgium, 2002, p.116. 392. D. Walters, Working Safely in Small Enterprises in Europe, European Trade Union Confederation, Brussels, Belgium, 2002, p.130. 340
References 393. D. Walters, Working Safely in Small Enterprises in Europe. European Trade Union Confederation, Brussels, Belgium, 2002, p.131. 394. Safety Representatives, The Swedish Trade Union Confederation Regional LO, Stockholm, Sweden, 2001. 395. D. Walters, Working Safely in Small Enterprises in Europe, European Trade Union Confederation, (EUTC), Brussels, Belgium, 2002, p.32. 396. The Workers Safety Advisors Pilot, HSE Research Report, Her Majesty’s Stationary Office, London, UK, 2003, p.144. 397. The Legislative History of Deca-BDE in the European Union and within the Restriction on Hazardous Substances (RoHS) Directive, CPA, 2005. 398. J. Ahlers, Substitution and Precaution – Both on Principle? The Role of Precaution in Chemicals Policy, Eds., E. Freytag, T. Jackl, G. Loibl and M. Wittmann, Diplomatische Akademie, Vienna, Austria, 2001, p.85. 399. F.M. Christensen S.I. and Olson, Risk assessment of Short and Medium Chain Chlorinated Paraffin’s (SCCP and MCCP), Technical Report, Technical University of Denmark, 2002. 400. Summary Record: Ad-hoc Meeting of the Commission Working Group on the Classification and Labelling of Dangerous Substances, KEMI, Solna, Sweden, 1999, ECBI/39/99 – Rev. 1. 401. M. Darvill, Update on European Occupational Safety and Health Issues and the UK, Paper presented at the UK Health and Safety Commission Meeting, 2005, HSC/05/106. 402. UK Parliament Select Committee on European Legislation, Protection of Workers Against Carcinogens, Proceedings of the Committee, SCEL, European Legislation - 27th Report, 1998. 341
Framework for Chemical Risk Management under REACH 403. Update on the Official EU Test Methods for Azocolorants, Sparkle, Intertek Labtest (IL), 2004, 173, 1. 404. L. Krämer, Casebook on EU Environmental Law, Hart Publishing, Oxford, UK, 2002, p.181. 405. European Union Network for the Implementation and Enforcement of Environmental Law, Second Annual Survey on the Implementation and Enforcement of Community Law (January 1998 to December 1999), Commission of the European Communities, IMPEL, Brussels, Belgium, 2000, p.26. 406. R. Zubek, Implementation of Community Legislation in Poland and Institutional Configurations at the Centre of Government, Department of Government Working Paper, London School of Economics and Political Science, London, UK, 2002. 407. R. Zubek, Complying with Transposition Commitments in Poland: Collective Dilemmas, Core Executive and Legislative Outcomes, West European Politics, 2005, 28, 3, 592. 408. O. Renn, Commentary on ‘The Changing Character of Regulation: A Comparison of Europe and the United States,’ Risk Analysis, 2001, 21, 406. 409. R.E. Quinn and J. Rohrbaugh, A Spatial Model of Effectiveness Criteria: Towards a Competing Values Approach to Organizational Analysis, Management Science, 1983, 29, 363. 410. J. Jordan, The Impact on UK Environmental Administration, British Environmental Policy and Europe: Politics and Policy in Transition, Eds., P. Lowe and S. Ward, Routledge, London, UK, 1998, p.187. 411. Cultures of Democracy: Historical Formation and Contemporary Challenges, Democracy at Work: A
342
References Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, USA, 2000, p.193. 412. H. Pehle, Germany: Domestic Obstacles to an International Forerunner, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.162. 413. C. Lees, Environmental Policy: The Law of Diminishing Returns? PAS Annual Conference, University of Lincoln, UK, 2004, p.5. 414. G. Loibl, The Precautionary Principle in International Law, The Role of Precaution in Chemicals Policy, Eds., E. Freytag, T. Jackl, G. Loibl and M. Wittmann, Diplomatische Akademie, Vienna, Austria, 2001, p.13. 415. A. Kronsell, Sweden: Setting a Good Example, European Environmental Policy: The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.53. 416. Manual for Investigation of HPV Chemicals: Description of OECD Work on Investigation of High Production Volume Chemicals, 2005, p.3. 417. European Agency for Safety and Health at Work, Priorities and Strategies in Occupational Safety and Health Policy in the Member States of the European Union, OSHA, Bilbao, Spain, 2000, p.15. 418. S. Boehmer-Christiansen and J.F. Skea, Acid Politics: Environmental and Energy Policies in Britain and Germany, Belhaven Press, London, UK, 1991, p.15.
343
Framework for Chemical Risk Management under REACH 419. Cultures of Democracy: Historical Formation and Contemporary Challenges, Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, USA, 2000, p.159. 420. UK Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment, Statement on Adverse Trends in Development of the Male Reproductive System - Potential Chemical Causes, COT, 2004. 421. J. Ahlers, T. Beulshausen, T. Bigalke, H. Eggers, A. Gies, P. Greiner, K. Henseling, B. Melhorn, H. Merkel, I. Paulini, K. Steinhäuser, H. Stolzenberg, K. Vormann and S. Wiandt, Precautionary Risk Assessment and Risk Management of Chemical: New strategies for the ecological risk assessment and risk management of substances, Umweltbundesamt (German Federal Environment Agency) Report, 2001. 422. Phthalates, The Facts Behind the Issues, Health Issues, UK Food Standards Agency (FSA), London, UK,. 423. C.M. Carruthers and P.M.D. Foster, Critical Window of Male Reproductive Tract Development in Rats Following Gestational Exposure to Di-n-butyl Phthalate, Birth Defects Research (Part B), 2005, 74, 277. 424. Chemicals Known to the State to Cause Cancer or Reproductive Toxicity, Office of Environmental Health Hazard Assessment Safe Drinking Water and Toxic Enforcement Act of 1986, State of California Environmental Protection Agency, 2005, p.65. http://www.oehha.org/prop65/prop65_list/120205list.html. 425. E. Hildebrandt and E. Schmidt, Industrial Relations and Environmental Protection in Europe, European Foundation for the Improvement of Living and Working Conditions, Office for Official Publications of the European Communities, Luxembourg, 1994, p.115. 344
References 426. N. Eisberg, Biorefinery Revolution, Chemistry and Industry, 2005, 19, 13. 427. S. Milmo, Heads You Win, Chemistry and Industry, 2006, 3, 25. 428. Bioctechnology for Clean Industrial Products and Processes, OECD Publications, Paris, France, 1998, p.35. 429. Pan European Survey: Image of the Chemical Industry, CEFIC 2004, p.6. 430. J. Boler, A Question of Trust, Chemistry and Industry, 2005, 23, 14. 431. Environmental Performance Reviews: Germany, OECD Publications, Paris, France, 2001, p.141. 432. CEFIC Pan European Survey: Image of the Chemical Industry, 2004. 433. The Public Image of the Chemical Industry, Research Conducted for the Chemical Industries Association, Final Report, MORI, London, UK, 2002. 434. A. Boholm, Comparative Studies of Risk Perceptions: A Review of Twenty Years of Research, Journal of Risk Research, 1998, 1, 2, 135. 435. E. Hildebrandt and E. Schmidt, Industrial Relations and Environmental Protection in Europe, European Foundation for the Improvement of Living and Working Conditions, Office for Official Publications of the European Communities, Luxembourg, 1994, p.150. 436. EU-Chemiepolitik muss Schutz der Gesundheit und Umwelt in den Mittelpunkt Stellen: Stellungnahme von IG Meall, ver. di und IG BAU, Documentation Brief, Arbeit and Ökologie, 2005, 39, 9.
345
Framework for Chemical Risk Management under REACH 437. J. Tickner and K. Geiser, New Directions in European Chemicals Policies: Drivers, Scope and Status, Lowell Center for Sustainable Production, Lowell, MA, USA, 2003, p.133. 438. A. Kronsell, Sweden: Setting a Good Example, European Environmental Policy, The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.50. 439. Y. Mény and A. Knapp, Government and Politics in Western Europe: Britain, France, Italy, Germany, 3rd Edition, Oxford University Press, Oxford, UK, 1998, p.129. 440. Sweden: Economy, Politics and Government, World of Information, Business Intelligence Report, 2001. 441. Environmental Performance Review: Sweden, OECD Publications, Paris, France, 2004, p.132. 442. A. Kronsell, Sweden: Setting a Good Example, European Environmental Policy, The Pioneers, Eds., M.S. Andersen and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.48. 443. C. Borgards, France: Rule by Virtue of Elite Position, Democracy at Work: A Comparative Sociology of Environmental Regulation in the United Kingdom, France, Germany, and the United States, Ed., R. Münch, Praeger, Westport, CT, USA, 2000, p.89. 444. C. Crouch, Industrial Relations and European State Traditions, Oxford University Press, Oxford, UK, 1994, p.303. 445. C. Crouch, Industrial Relations and European State Traditions, Oxford University Press, Oxford, UK, 1994, p.338.
346
References 446. E. Hildebrandt and E. Schmidt, Industrial Relations and Environmental Protection in Europe, European Foundation for the Improvement of Living and Working Conditions, Office for Official Publications of the European Communities, Luxembourg, 1994. 447. I. Cartier, J.P. Cazeneuve, D. Dupré and D. Le Page, An Innovative Economic Incentive Model to Improve the Working Environment: Testing in France, European Foundation for the Improvement of Living and Working Conditions, Dublin, Republic of Ireland, 1996. 448. C. Crouch, Industrial Relations and European State Traditions, Oxford University Press, Oxford, UK, 1994, p.325. 449. M. Douglas, Risk and Blame: Essays in Cultural Theory, 3rd Edition, Routledge, London, UK, 1996. 450. M. Casson, A Cultural Theory on Industrial Policy, European Industrial Policy: The Twentieth-Century Experience, Eds., J. Foreman-Peck and G. Federico, Oxford University Press, Oxford, UK, 1999, p.405. 451. E. Hildebrandt and E. Schmidt, Industrial Relations and Environmental Protection in Europe, European Foundation for the Improvement of Living and Working Conditions, Office for Official Publications of the European Communities, Luxembourg, 1994, p.111. 452. J. Dormois, The Idiosyncrasies of Volontarisme, European Industrial Policy: The Twentieth-Century Experience, Eds., J. Foreman-Peck and G. Federico, Oxford University Press, Oxford, UK, 1999, p.92. 453. What We Need from REACH: Views on the proposal for a new chemical legislation within the EU, ChemSec, International Chemical Secretariat, 2005. 347
Framework for Chemical Risk Management under REACH 454. Shop Till You Drop? A Survey of High Street Retailers and Risky Chemicals, UK Friends of the Earth Report, 2004. 455. European Union Network for the Implementation and Enforcement of Environmental Law, Second Annual Survey on the Implementation and Enforcement of Community Law (January 1998 to December 1999), Commission of the European Communities, IMPEL, Brussels, Belgium, 2000, p.37. 456. P.M. Barnes and I.G. Barnes, Environmental Policy in the European Union, Edward Elgar Publishing, Cheltenham, Gloucester, UK, 1999, p.102. 457. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.141. 458. Council Regulation (EEC) No 793/93 of 23 March 1993 on the evaluation and control of the risks of existing substances, Official Journal of the European Communities, EEC, 1993, No.L.84. 459. P. Libreros, French View of the Chemicals Agency in REACH – SAFE Proposal, Presentation at the European Environmental Bureau Seminar on Chemicals, Brussels, Belgium, 2005 460. M. Seychell, Small Member States, the Agency and SMEs, Presentation at the European Environmental Bureau Seminar on Chemicals, Brussels, Belgium, 2005. 461. Assessment of the Business Impacts of New Regulations in the Chemicals Sector Phase 2: Availability of Low Value Products and Product Rationalisation, Final Report prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, UK, 2003. 462. J. Solbé, PRODUCE: Piloting REACH on Downstream Use and Communication in Europe, Final Report, 2006. www.producepartnership.be 348
References 463. M. Führ, S. Merenyi, K. Bizer, D. Lewin, G. Cichorowski and S. Kleihauer, Schnittstellenprobleme Zwischem Gemeinschaftichem Soffrecht und anderem Sektoralen Umweltrecht (Umsetzungehmmnisse bei der Riskiominderung von Altstoffen nach 793/93/EG – “Instrumentenlücke”), Vorläufiger Abschlussbericht, Sofia – Sonderforschungsgruppe Institutionenanlyse, Darmstadt, Germany, 2004. 464. Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities, CEC, Luxembourg, 1998, p.2. 465. Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities, CEC, Luxembourg, 1998, p.6. 466. Technical Guidance Document on the Development of Risk Reduction Strategies, Luxembourg, Office for Official Publications of the European Communities, CEC, Luxembourg, 1998, p.56. 467. M. Glachant, The Need for Adaptability in EU Environmental Policy Design and Implementation, European Environment, 2001, 11, 5, 239. 468. L. Perenius, The Marketing Restriction Regime under Directive 76/769/EEC and Its Implementation, Risk Asessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, p.54. 469. S. Lanser and T. Pless-Mulloli, Integrated Pollution Prevention and Control: A Review of Health Authorities’ Experience, Journal of Public Health Medicine, 2003, 25, 3, 234.
349
Framework for Chemical Risk Management under REACH 470. European Union Network for the Implementation and Enforcement of Environmental Law, Second Annual Survey on the Implementation and Enforcement of Community Law (January 1998 to December 1999), Commission of the European Communities, Brussels, Belgium, 2000. 471. L. Krämer, Casebook on EU Environmental Law, Hart Publishing, Oxford, UK, 2002, p.184. 472. European Commission Expert Team for Vapour Retarding Additives Effectiveness of Vapour Retardants in Reducing Risks to Human Health from Paint Strippers containing Dichloromethane, Final Report published by the European Commission, Brussels, Belgium, 2004. http://europa.eu.int/comm/enterprise/chemicals/docs/ studies/ etvaread-paint_stripper_dcm.pdf. 473. A.W. van der Wielen, Acceptable Risk Assessments by Industry: Lessons from Experience and Challenge for Future, 23rd Annual Chemical Control Conference – The EU Draft Chemical Control Regulations, Charles Simeons Conferences, London, UK, 2002. 474. Derivation of Assessment Factors for Human Health Risk Assessment, European Centre for Ecotoxicology and Toxicology of Chemicals, Brussels, Belgium, 2003. 475. L. Krämer, Introduction into the European chemicals regulation: Basic structures and performance, Risk Assessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, p.27. 476. M. Führ, Industrial Self-Control in the Regulation of Chemicals, Risk Assessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft: BadenBaden, Germany, 2000, p.165.
350
References 477. Practical Experience from Enforcement of Chemicals Legislation (PEECL), Umweltbundesamt (Austrian Federal Environment Agency), Chemical LegislationbEuropean Enforcemnet Network, Austria, 2002. 478. Study on the Impact of the Revision of the Council Directive 88/378/EEC on the Safety of Toys, Final Report - prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, UK, 2004, p.42. 479. Study on the Impact of the Revision of the Council Directive 88/378/EEC on the Safety of Toys, Final Report - prepared for the European Commission Directorate-General Enterprise, Risk and Policy Analysts, London, UK, 2004, p.55. 480. Corporate Environmental and CSR Issues Management: An Environmental Resources Management Study, Executive Summary, ERM, Oxford, UK, 2003. 481. L. Krämer, Casebook on EU Environmental Law, Hart Publishing, Oxford, UK, 2002, 159. 482. P. Eder and M. Sotoudeh, Innovation and Cleaner Technologies as a Key to Sustainable Development: The Case of the Chemical Industry, Report for the European Commission, Joint Research Centre Institute for Prospective Technological Studies: Seville, Spain, 2000. 483. S. Erler, H. Jeswani and W. Wehrmeyer, Business Impacts of REACH: Costs, Obstacles and Opportunities, CES Final Report, Centre for Environmental Strategy, University of Surrey, Guildford, UK, 2005. 484. S.O. Hansson, Replacing the No-Effect Level (NOEL) with Bounded Effect Levels (OBEL and LEBEL), Statistics in Medicine, 2002, 21, 3071.
351
Framework for Chemical Risk Management under REACH 485. Directive 98/34/EC of the European Parliament and of the Council of 22 June 1998 laying down a procedure for the provision of information in the field of technical standards and regulations. Official Journal of the European Union, 1998 No.L204, p.14. 486. Directive of the European Parliament and of the Council or 27 February 2003 amending Council Directive 76/768 on the approximation of the laws of the Member States relating to cosmetic products, Official Journal, EC, 2003, No.L.66. 487. List of Approved Products and Processes, Department for Environment, Food and Rural Affairs, Drinking Water Inspectorate, DWI, London, UK, 2003. 488. Council Directive 96/62/EC of 27 September 1996 on ambient air quality assessment and management, Official Journal of the European Union, EC, 1996, No.L.296. 489. G. Winter, Redesigning Joint Responsibility of Industry and Government, Risk Assessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, p.182. 490. J. Golub, New Instruments for Environmental Policy in the EU: Introduction and Overview, New Instruments for Environmental Policy in the EU, Ed., J. Golub, Routledge, London, UK, 1998, p.5. 491. H. von Holleben, Comment: Responsible Care in the German Chemical Industry, Risk Assessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, 187.
352
References 492. L. Perenius, The Marketing Restriction Regime under Directive 76/769/EEC and Its Implementation, Risk Asessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, 53. 493. P.M. Barnes and I.G. Barnes, Environmental Policy in the European Union, Edward Elgar Publishing, Cheltenham, Gloucester, UK, 1999, p.158. 494. Commission Communication on Trade and the Environment, COM(96)54. CEC, Brussels, Belgium, 1996, p.10. 495. Institute of Environment and Health (IEH) and RPA (2006) Draft Annex XIV Guidance: Guidance for the Production of an Annex XIV Dossier, Produced by the RIP 4.4 Consortium under Service Contract CCR.IHCP.C430320. X0 BRE, 2006. 496. B. Ballantine, Enhancing the Role of Science in the DecisionMaking of the European Union, The European Policy Centre, Brussels, Belgium, 2005, p.3. 497. Council Directive 98/24/EC of 7 April 1998 on the protection of the health and safety of workers from the risks related to chemical agents at work (fourteenth individual Directive within the meaning of Article 16(1) of Directive 89/391/EEC,Official Journal of the European Communities, EC, 1998, No.L.31. 498. Directive 2001/95/EC of the European Parliament and of the Council of 3 December 2001 on general product safety. Official Journal of the European Union, EC, 2001, No.L.01. 499. Existing Chemicals: II Priority Setting, European Chemicals Bureau, 2005.
353
Framework for Chemical Risk Management under REACH 500. J. Ranke, Ecotoxicological Risk Profiles of Chemicals: Concept and Application to Antifouling Biocides, Dissertation, University of Bremen, Germany, 2001, p.25. 501. Chemicals in Products: Safeguarding the Environment and Human Health, 24th Report, The Stationary Office, RCEP, London, UK, 2003, p.106. 502. Select Committee on Science and TechnologyRelevant documents: Within REACH, the EU’s new chemicals strategy—Sixth Report from the Science and Technology Committee, Session 2003–04, HC 172-I, and the Government’s response thereto, 7th Special Report, Session 2003–04, HC 895, Westminster Hall, Thursday 9 September 2004, 9 Sept 2004, Column 357WH, HC, UK House of Commons, London, UK, 2004. 503. Commission Working Group of Specialised Experts in the fields of Reprotoxicity, Ispra, Italy, 2004, ECBI/132/04 Rev. 2 504. Borax Europe Ltd’s Comments on the European Chemical Bureau’s Summary Record of 5-6 October 2004 meeting of Specialised Experts in the Field of Reprotoxicity, 2004, ECBI/118/04 Add.80. 505. R. Kasperson, The Social Amplification of Risk: Progress in Developing an Integrative Framework, Social Theories of Risk, Eds., Krimsky and Golding, Praeger Publishers, Westport, CT, USA, 1998, p.153. 506. R.E. Alcock and J.S. Busby, Risk migration and scientific advance: the case of flame retardant compounds, Risk Analysis, 2006, 26, 369. 507. Handbook of Biodiversity Valuation: A Guide for Policy Makers, OECD Publications, Paris, France, 2002. 508. Technical Guidance Document on the use of Socio-economic Analysis in Chemical Risk Management Decision Making, 354
References Series on Risk Management, OECD Publications, Paris, France, 2002, 14, 80. 509. J. Adams, Risk, 5th Edition, Routlege, London, UK, 2001, p.95. 510. Technical Guidance Document on the use of Socio-economic Analysis in Chemical Risk Management Decision Making, Series on Risk Management, OECD Publications, Paris, France, 2002, 14. 511. Safer Chemicals Within Reach: Using the Substitution Principle to Drive Green Chemistry, REACH Report prepared for the Greenpeace Environmental Trust by Clean Production Action, Clean Production Action, 2004. www.cleanproduction.org/library/SafeChem.pdf. 512. 23rd Meeting of the UK Chemicals Stakeholder Forum, UK Department for Environment, Food and Rural Affairs, Chemical Stakeholder Forum, London, UK, 2006. 513. Dutch Ministry of Housing, Spatial Planning and the Environment (2004) Dutch Substances Policy in an International Perspective, Memorandum on implementation of SOMS, VROM, The Hague, The Netherlands, 2004, p.42. 514. Dutch Ministry of Housing, Spatial Planning and the Environment Dutch Substances Policy in an International Perspective, Memorandum on implementation of SOMS, VROM, The Hague, The Netherlands, 2004, p.14. www.vrom.nl/international 515. Commission Directive 2004/98/EC of 30 September 2004 amending Council Directive 76/769/EEC as regards to restrictions on the marketing and use of pentabromodiphenyl ether in aircraft emergency evacuation systems for the purpose of adapting its Annex I to technical progress, 1.10.2004, Official Journal of the European Union, EC, 2004, No.L.305. 355
Framework for Chemical Risk Management under REACH 516. L. Krämer, Casebook on EU Environmental Law, Hart Publishing, Oxford, UK, 2002, p.221. 517. L. Perenius, The Marketing Restriction Regime under Directive 76/769/EEC and Its Implementation, Risk Asessment and Risk Management of Toxic Chemicals in the European Community: Experiences and Reform, Ed., G. Winter, Nomos Verlagsgesellschaft, Baden-Baden, Germany, 2000, p.55 518. European Pollutant Emission Register, EPER Review Report, Final Report, Commission of the European Communities, 2004, p.113. 519. Risk Assessment 1-4 Dichlorobenzene, Draft Report March, National Institute of Research and Security, 1999. 520. Risk Reduction Strategy and Analysis of the Advantages and Drawbacks for Perfluorooctane Sulphonate (PFOS), Final Report, Risk and Policy Analysts and BRE Environment, 2004. 521. European Pollutant Emission Register, EPER Review Report, Final Report, Commission of the European Communities, 2004, p.83. 522. European Pollutant Emission Register, EPER Review Report, Final Report, Commission of the European Communities, 2004, p.89. 523. Assessment of the Business Impacts of New Regulations in the Chemicals Sector, Final Report prepared for the European Commission Directorate-General Enterprise, RPA and Swedish Statistics, London, UK, 2002, p.11. 524. About Vision Zero, Stockholm Environment Institute, 2005. 525. Risk: Leadership and Communication in Government. EPSRC Risk Network Presentation, HM Treasury, Risk Network, London, UK, 2005.
356
References 526. Sustainable Development Report, Rhodia, 2005, p.11. www.rhodia.com 527. Chemical and Plastic Industry in Sweden, Facts and Figures, Plast- and Kemiföretagen, Stockholm, Sweden, 2003. 528. European Region Product Integrity Department Request for Data for Regulatory Compliance, Rohm and Haas Company, 2005. 529. Health and Environmental Impact of Toxic Chemicals: Chemical Management Policies of Russia and EU Countries, Moscow Seminar, ChemSec, Brussels, Belgium 2005. 530. N. Eisberg, Amending the REACH Legislation, Chemistry and Industry, 2005, 18, 4. 531. C. Hogue, Chemicals Management: Countries are on the Verge of Finishing New International Agreement, Chemical and Engineering News, 2005, 83, 38, 27. 532. Development of Risk Reduction Strategies for Priority Chemicals: A Guidance Document (Pilot Version), Published in Conjunction with the International Programme on Chemical Safety (IPCS), UNITAR, Geneva, Switzerland, 1999. 533. J. de Bruijn, The EU Interim Strategy for Management of PBT and VPVB Substances, Presentation at Thirteenth Meeting of the UK Chemicals Stakeholder Forum, 2003. 534. D.J.H. Phillips, Bioaccumulation, Handbook of Ecotoxicology, Ed., P. Calow, Blackwell Science, Oxford, UK, 1998, p.392. 535. J. Timbrell, Introduction to Toxicology, 3rd Edition, CRC Press, London, UK, 2002, p.64. 536. F.M. Christensen and S.I. Olson, The Potential Role of Life Cycle Assessment in Regulation of Chemicals in the
357
Framework for Chemical Risk Management under REACH European Union, International Journal of Life Cycle Assessment, 2004, 9, 5, 327. 537. A. Wegener Sleeswijk, GLOBOX: a Global Multimedia Fate and Exposure Model. 538. de Kroning and co-workers, 2002. 539. J. Guinée, R. Heijungs, L. van Oers, D. van de Meent, T. Vermeire and M. Rikken, LCA Impact Assessment of Toxic Releases: Generic Modelling of Fate, Exposure and Effect for Ecosystems and Human Beings with Data for about 100 Chemicals, Dutch Ministry of Housing, Spatial Planning and Environment (VROM), The Hague, The Netherlands, 1996. 540. G. Finnveden, On the Limitations of Life Cycle Assessment and Environmental Systems Analysis Tools in General, International Journal of LCA, 2000, 5, 229. 541. J. de Bruijn, Risk Management of Chemicals: The Challenges of REACH, School of Architecture and the Built Environment, News Swedish Royal Institute of Technology, 2005. 542. S. Maguire and J. Ellis, The Precautionary Principle and Global chemical risk management: Some insights from POPs, Greener Management International, 2003, 41, 33. 543. J. Weiner, Precaution in A Multi Risk World, The Risk Assessment of Environmental and Human Health Hazards, Ed., D.J. Paustenbach, John Wiley and Sons, New York, NY, USA, 2002, p.1509. 544. Summary of: Documentation for In-Depth Evaluation of the Environmental Quality Objective of a Non-Toxic Environment, (Ref no.342-531-03, 13.) KemI, Solna, Sweden, 2003.
358
References 545. The Swedish Environmental Objectives: A Summary of the Government Bill, Swedish Ministry for the Environment, 1997/98, 145, 25. 546. Summary of: Documentation for In-Depth Evaluation of the Environmental Quality Objective of a Non-Toxic Environment, Document Reference No.342-531-03, KemI, KemI: Solna, Sweden, 2003. 547. Summary of: Documentation for In-Depth Evaluation of the Environmental Quality Objective of a Non-Toxic Environment, (Ref no.342-531-03 14) KemI, Solna, Sweden, 2003. 548. The Swedish Environmental Objectives: A Summary of the Government Bill, Swedish Ministry for the Environment, 1997/98, 145, 27. 549. The Swedish Environmental Objectives: A Summary of the Government Bill, Swedish Ministry for the Environment, 1997/98, 145, 28. 550. A. Kronsell, Sweden: Setting a Good Example, European Environmental Policy: The Pioneers, Eds., M.S. Andersen M.S. and D. Liefferink, Manchester University Press, Manchester, UK, 1997, p.43. 551. B. Ballantine, Improving the Quality of Risk Management in the European Union, Risk Communication, The European Policy Centre, Brussels, Belgium, 2003, p.20. 552. B. Ballantine, Improving the Quality of Risk Management in the European Union, Risk Communication, The European Policy Centre, Brussels, Belgium, 2003, p.21. 553. The Swedish Chemicals Inspectorate has Not Proposed to Have the Teeth of Deceased Persons Extracted, KemI, Press Release, 2004.
359
Framework for Chemical Risk Management under REACH 554. J. Cramer, Environmental Management: From ‘Fit’ to ‘Stretch’, Business Strategy and the Environment, 1998, 7, 162. 555. H. Mintzberg, Mintzberg on Management, The Free Press, New York, NY, USA, 1989.
360
A
ppendix
Appendix 1.1 - Definitions Terms in italics correspond to elements of the systems framework presented in Chapter 5. Adequate control
Following the definition of the European Commission, “exposure of humans and the environment can be considered to be ‘adequately controlled’ if the Derived No Effect Levels (DNEL) or the Predicted No Effect Concentrations (PNEC) are not exceeded” [199].
Article
An object which during production is given a special shape, surface or design which determines its function to a greater degree than does its chemical composition.
Bioaccumulative
Bioaccumulation is the uptake and sequestration of synthetic substances by biological organisms that results in increasing exposure levels. It is a biological phenomenon that combines bioconcentration and biomagnification. Bioconcentration refers to the distribution of a substance in the fat tissues of an organism, which increase in concentration when tissue absorption levels from an external environment exceed rates for metabolism and excretion. Biomagnification
361
Framework for Chemical Risk Management under REACH specifically refers to bioconcentration process that result from a trophic food chain. For organic molecules, bioaccumulation is often modelled using the octanol–water partition VivwViÌÃÊ vÊ >Ê ÃÕLÃÌ>ViÆÊ Ì
iÊ ÀiÃÕÌÃÊ >ÀiÊ then combined with monitoring data. Predicting the potential bioaccumulation of inorganic molecules, such as metals, is more difficult and therefore generally requires comparatively more detailed toxicological experimentation and exposure monitoring studies [534]. Chemical safety
With regards to chemical risk management, ‘safety’ refers to the application of knowledge on protective and preventive measures to limit, reduce and avoid chemical risks.
Chemical Safety Report under REACH
A Chemical Safety Report that documents chemical risk assessments must be prepared for substances that a company manufactures or imports at q10 tonnes per year. If a substance is identified as ‘dangerous’ or ‘very persistent and very bioaccumulative’ (VPVB), a set of exposure scenarios detailing the relevant risk management measures necessary to reduce exposures must be attached as an Annex to existing Safety Data Sheets. Hazardous properties of substances or preparations are classified under Directive 67/548 as ‘dangerous’ if exposure can result in a risk to human health or the environment [272]. Classification of preparations depends on the hazardous properties and concentrations of its constituent ingredients according to Directive 99/45. This book
Dangerous
362
Appendix uses the term ‘dangerous’. However, the term ‘dangerous’ will be changed in the REACH Regulation to ‘hazardous’ by Classification, Labelling and Packaging Regulation in 2010. Directive 67/548 and Directive 99/45 will also be replaced with Classification, Labelling and Packaging Regulation 1272/2008. Exposure
Exposure describes the integral of a concentration of a synthetic substance in a biological or environmental system resulting from a particular use, or a combination of uses, of a chemical or group of chemicals over a specified period of time. Whether measured or modelled, exposures are usually grouped according to sets of physical and biological parameters corresponding to certain environmental emissions from industrial processes or releases of chemicals during the use of manufactured products.
Hazard
Hazard refers to the intrinsic properties of a chemical that establish its potential to harm humans or the environment following exposure.
Listed uses
Under the systems framework, the chemical identity of substances contained in a specified set of consumer preparations (i.e., mixtures) and articles (i.e., finished manufactured products) would be made directly accessible to the general public via the European Chemicals Agency REACH-IT website.
Mobile
Mobility describes the transportation of chemicals through environmental media.
363
Framework for Chemical Risk Management under REACH Partition coefficients of a substance between air, water, and organic phases provide indicators of the distance a molecule will travel when released into the environment. Substances with ‘high mobility’ have a particularly high potential to contaminate groundwater and pollute oceans. Monitoring Network
Under the systems framework, a Monitoring Network would comprise representatives from Member State regulatory authorities, EU agencies and the Directorates-General of the European Commission. In addition to co-ordinating activities across different EC legislative framework, the network would: (1) review REACH decision-making rules; and, (2) promote the harmonisation of administrative procedures for companies demonstrating compliance across REACH and other EU legislative frameworks.
Persistent
A ‘persistent’ substance undergoes very slow rates of degradation in environmental media and/or metabolism in biological systems. The term therefore refers to the ability of a substance to remain unchanged in the wider environment.
Preparation
A mixture or solution composed of two or more substances. This book uses the term ‘preparation’. However, the term ‘preparation’ was changed in the REACH Regulation to ‘mixture’ by the Classification, Labelling and Packaging Regulation in 2009.
Permissible uses
The systems framework proposed that these permissible uses would have low priority
364
Appendix under REACH and could also be issued with time-limited exceptions from restrictions and authorisations. Recommendations
Under the systems framework, regulatory recommendations for professional and consumer products should guide regulators and industry through REACH. Companies would be able to develop recommendation equivalents that could incorporate a wider set of communication and management tools than currently envisaged under REACH.
Reprotoxic
As well as potential effects on fertility and genetic alterations in chromosomes caused by exposure, reproductive toxicity can cause alterations to the developmental characteristics of progeny [109, 535].
Safe uses
To minimise the number of substances and uses subject to REACH, the systems framework proposes that companies should be able identify safe chemical uses prior to registration.
Safety Data Sheet
Safety Data Sheets must be supplied to companies upon first business transaction for ‘dangerous’ substances and preparations. The Safety Data Sheet summarises hazard data and the appropriate safety measure for use and disposal of the given chemical. Thereafter, suppliers and users share responsibility for updating a Safety Data Sheet. A Safety Data Sheet must usually be made available upon demand for professional users but not members of the general public.
365
Framework for Chemical Risk Management under REACH Stakeholder Inputs
A formal structure for incorporating stakeholder inputs during REACH, as described by the systems framework.
Substance
A chemical element and its compounds in the natural state or obtained by any manufacturing process, including any additive necessary to preserve its stability and any impurity deriving from the process used, but excluding any solvent which may be separated without affecting the stability of the substance or changing its composition.
Target-setting
Under the systems framework, a method of target-setting would co-ordinate action across various legislative frameworks outside the scope of REACH. The process would apply when risk reduction has been identified as necessary across the EU but the socio-economic, environmental or political factors involved in decision-making are identified as varying significantly between Member States. Devising, monitoring and reporting targets would be co-ordinated by a Monitoring Network.
Volatile
Volatility describes the rate at which liquids and solids turn to vapour at standard temperature and pressure.
366
Appendix
Appendix 1.2 - Substances outside the scope of the research project Food additives and flavourings in foodstuffs Unlike the chemicals investigated by this research study, these additives and flavourings are intended to be ingested orally. Current EC legislation on these substances requires specific risk assessment and risk management procedures for these substances. Medicinal (and veterinary products) Medicines are administered directly to a patient (e.g., oral doses, topical applications, or parenteral injections). There are also many issues relating to the risk management of medicinal products that do not apply to chemicals, such as antibiotic resistance and release into the environment of metabolites that are excreted through urine. Although drug precursors to medicines fall within the scope of REACH, these are generally fine intermediates consumed in closed systems with low exposures to human health or the environment. Medicinal products often have strong and obvious beneficial socio-economic uses by, for example, directly assisting the combat of disease, this strongly affects risk management options for drug precursors. Biocides These toxic substances can serve some necessary hygienic purposes, such as to maintain sterile conditions in hospitals. Although some of the chemicals used in formulations used in biocides fall under REACH, the active ingredients present within biocides do not (e.g., bacteriocides), but are regulated instead under Directive 98/8/EC. In addition, the use of active ingredients may result in the development >`Ê Ã«Ài>`Ê vÊ `ÀÕ}Ê ÀiÃÃÌ>ViÊ Ê VÀÀ}>ÃÃÆÊ Ì
ÃÊ ÃÊ >Ê specific biochemical risk management consideration that is not pertinent to this research project. 367
Framework for Chemical Risk Management under REACH Plant protection products and fertilisers These products are applied in large quantities during agricultural activities. Humans may also ingest residues on food. Risk assessment of these products therefore requires specific toxicological studies and environmental studies that involve complex modelling (far more detailed than any risk assessment envisaged under REACH) and heavy reliance on monitoring data. Separate authorisation schemes for the active substances contained in plant protection products and fertilisers exist in several EC Directives. The use of plant protection products and fertilisers generally falls within the scope of agricultural policy (which is distinct from chemicals policy) as they involve questions pertaining to intensified food production and agricultural subsidies. Furthermore, the controls for the proper use of these substances involve specific training programmes and technology in the agriculture sector along with extensive monitoring programmes. Detergents Thousands of tonnes of detergents are used every year in the EU alone. Modern-day detergents are generally not considered to be toxic to the environment following ÃiÜ>}iÊÜ>ÌiÀÊÌÀi>ÌiÌÆÊÀ>Ì
iÀ]ÊÌ
iʫ
ë
>ÌiÊVÌiÌÊ>`Ê biological oxygen demand are the focus of risk management controls as these cause eutrophication in rivers and lakes. Detergents are regulated by a number of EC directives that will be consolidated after the adoption of a recently proposed EC regulation. Fuels (primarily used in combustion engines) Today’s societal energy and transport needs heavily depend on fuels. Regulating
368
Appendix fuels therefore poses significantly wider socio-economic impacts than most of the chemicals considered in this research project. In addition, risk management control options relating to fuel storage, distribution and filling in petrol stations differ from the chemicals considered in this project. Radioactive substances The very particular nature of the risks involved with their use, which may span thousands of years, means that risk management of these substances presents a unique field on its own.
369
Sectors
Basic Chemicals
Divisions
Organic
Inorganic
Primary inputs
Oil, gas, biobased products
Mines, sea water
Products
Industrial chemicals, incl. intermediates, solvents and plastics
Customers
Fine Chemicals Composition products Basic chemicals
Specialty Chemicals Performance products (other than coatings)
Coatings
Basic chemicals
Performance and basic
Bulk chemicals for Single substances industrial use for life science products, incl. intermediates, food supplements, vitamins
Multi-purpose: personal care, electronics, food, textiles, general manufacturing
DIY, trade, inks
General manufacturing, construction, packaging
Industrial use
Pharmaceutical, agriculture
Textiles, leathers, paper, paints, construction, agrochemicals, pharma, water, coatings, plastics
Automobile, aeronautic, professional users, consumers
Product value
High volume, low value
Medium volume, high value
Low volume, medium value
Medium volume, high value
Medium volume, medium value
Product life
Long
Long
Medium-long
Medium-short
Short
Product diversity
Low
Very low
Very low
High
Medium
Framework for Chemical Risk Management under REACH
370
Appendix 1.3 - Chemical industry activity (information based on [24])
R&D focus
Process development
Application development/ new products
Capital intensive industry dependent on oil and gas
Capital intensive Feedstock influences location of production Independent of oil
High number of R&D products Products for industrial use
Products for industrial use, but often with final consumer exposure
Products for large scale industry use or small traders or consumers
Customer relations
Distributors are used to service small customers Contract sales
Close contact for specific needs in science
Close contact Long-term contracts for certain intermediates and building blocks
Technical assistance very important Close contact with clients
Close contact for industrial uses
Implications of REACH
Wide number of downstream users across many sectors
Wide number of downstream users across many sectors
Limited number of industrial uses and exposure scenarios
Consumer products may pose high exposure scenarios
Consumer products may pose high exposure scenarios
Competition
Often substitutes available Competition is mostly between very large companies
Highly competitive prices
Competitive prices Difficult to keep customer
High competition Constant innovation
New technology threatens Not much price competition
Appendix
371
Description
Framework for Chemical Risk Management under REACH
Appendix 2.1 - Risk Phrases Risk phrases following Directive 67/548 and Directive 99/45 Italics indicate flammable and explosive properties. R1
Explosive when dry
R2
Risk of explosion by shock, friction, fire or other sources of ignition
R3
Extreme risk of explosion by shock, friction, fire or other sources of ignition
R4
Forms very sensitive explosive metallic compounds
R5
Heating may cause an explosion
R6
Explosive with and without contact with air
R7
May cause fire
R8
Contact with combustible material may cause fire
R9
Explosive when mixed with combustible material
R10 Flammable R11 Highly flammable R12 Extremely flammable R13 Extremely flammable liquefied gas R14 Reacts violently with water R15 Contact with water liberates extremely flammable gases R16 Explosive when mixed with oxidising substances R17 Spontaneously flammable in air
372
Appendix R18 In use, may form flammable/explosive vapour-air mixture R19 May produce explosive peroxides R20 Harmful by inhalation R21 Harmful in contact with skin R22 Harmful if swallowed R23 Toxic by inhalation R24 Toxic in contact with skin R25 Toxic if swallowed R26 Very toxic by inhalation R27 Very toxic in contact with skin R28 Very toxic if swallowed R29 Contact with water liberates toxic gas R30 Can become very flammable in use R31 Contact with acids liberates toxic gas R32 Contact with acids liberates very toxic gas R33 Danger of cumulative effects R34 Causes burns R35 Causes severe burns R36 Irritating to eyes R37 Irritating to respiratory system R38 Irritating to skin
373
Framework for Chemical Risk Management under REACH R39 Danger of very serious irreversible effects R40 Limited evidence of carcinogenic effect R41 Risk of serious damage to eyes R42 May cause sensitisation by inhalation R43 May cause sensitisation by skin contact R44 Risk of explosion if heated under confinement R45 May cause cancer R46 May cause heritable genetic damage R47 May cause birth defects R48 Danger of serious damage to health by prolonged exposure R49 May cause cancer by inhalation R50 Very toxic to aquatic organisms R51 Toxic to aquatic organisms R52 Harmful to aquatic organisms R53 May cause long-term adverse effects in the aquatic environment R54 Toxic to flora R55 Toxic to fauna R56 Toxic to soil organisms R57 Toxic to bees R58 May cause long-term adverse effects in the environment
374
Appendix R59 Dangerous for the ozone layer R60 May impair fertility R61 May cause harm to the unborn child R62 Possible risk of impaired fertility R63 Possible risk of harm to the unborn child R64 May cause harm to breastfed babies R65 Harmful: May cause lung damage if swallowed R66 Repeated exposure may cause skin dryness or cracking R67 Vapours may cause drowsiness and dizziness R68 Possible risk of irreversible effects
375
Framework for Chemical Risk Management under REACH
Appendix 2.2 - Lifecycle Assessment (LCA) A product lifecycle ranges from resource extraction through manufacture to use and final waste disposal or processing. Under the current EU chemical risk assessment procedure, lifecycle thinking considers the risks at each stage of a substance’s lifecycle, with the possible summation of multiple sources of exposures to a single substance [144]. LCA that quantitatively evaluate the overall environmental and health impacts of processes, services or products can complement risk assessment as a useful decision-support tool for chemical risk management [183, 188, 536]. LCA covers not only the chemical’s hazardous properties, but also the nature of the production processes involved. For example, to produce 2 g of chemicals for an electronic chip requires 72 g of chemicals. Such considerations have considerable consequences for chemical risk management approaches, such as when setting environmental taxes and overall pollution prevention. The basis for LCA is the ‘functional unit’ which essentially defines a quantifiable function of a process, product or service (e.g., ‘1 square metre of painted surface area during 10 years’ in a LCA of paint). LCA considers many burdens to health and the environment, expressing these on the basis of their contribution per functional unit. In a LCA, chemical emissions and the potential for damage to human health or the environment are generally summed according to broad spatial and temporal characteristics (Figure 1). In this way, exposures are not evaluated in any detail and risks are not fully characterised. Numerous methodologies are being developed so that LCA results will represent actual risk levels rather than overall burdens [537, 538]. Carrying out LCA on substances for regulatory decision-making will usually not be necessary. If direct risks to the environment or human health are imminent and high for a specific substance, then regulatory control measures will be necessary regardless of overall impacts. For instance, indirect contributions to eutrophication, acidification,
376
Appendix ozone formation and climate change may often be small compared with direct impacts on health or the environment for a single highly toxic substance (see [399, 536, 539]). Moreover, collection of LCA data can be complex and time-consuming [539], as well as requiring careful communication to avoid misinterpretations of results [540]. One conclusion is that LCA and its application should not be done on individual substances for regulatory risk management decisionmaking of high-level risks, but rather at the functional level of a process or final product [536]. Another conclusion is that LCAs should be used limited to risk management at the local rather than EU level [541]. Because products assessed in a LCA are often characterised quantitatively in terms of function, comparisons can be made of product lifetime or durability and efficiency. For example, one paint may have greater covering power than the other or one paint may be more durable. In some cases, product performance can be an important variable for determining chemical exposure during the various stages of the product’s lifecycle, thereby having implications to risks. Comparative risk assessments between two or more products or processes may therefore also be done on a functional unit basis.
377
Framework for Chemical Risk Management under REACH
France
Cradle
Germany Belgium Netherlands Grave
Figure 2.2.1 Example of a LCA of a product where emissions of different substances at different locations that arise during its production, manufacture, use and disposal are indicated according to location (taken from [539]).
Appendix 2.3 - The Precautionary Principle Three aspects of precaution (following [542, 543]) 1. Uncertainty does not justify inaction (prevention) This aspect of the precautionary principle simply states that scientific uncertainty is not a reason to delay taking action. An uncertain or inconclusive result in a chemical risk assessment does not present a justification for inaction. The uncertainty of a given risk can be analysed through various methods, and if evidence then indicates that the risk is high enough, action can be taken. Alternatively, further risk assessment data can be collected to reduce uncertainty. Methods for screening to identify potential chemicals of concern are therefore seen as an application of the precautionary principle [542]. While further risk assessment is being done, risk reduction strategies can be
378
Appendix developed and formulated so that regulatory measures can be rapidly implemented. 2. Uncertain risk justifies action (protection) The literature identifies four cases when the potential risk from a chemical may be used to justify immediate action: ®Ê «ÀÀÊÌÊV«iÌÊvÊ>ÊvÕÊÀÃÊ>ÃÃiÃÃiÌÆ ®Ê vÊÌ
iÀiÊ>ÀiÊ
}
ÊiÛiÃÊvÊÕViÀÌ>ÌÞÊÊ>ÊÀÃÊ>ÃÃiÃÃiÌÆ (iii) if it is not possible to establish a causal link between exposure >`ÊivviVÌÆ (iv) if there is a risk of severe or irreversible damage to health or the environment. In short, each relates to measures necessary to achieve adequate protection. The suitability of each of the above applications of the second version of the precautionary principle is subject to intense debates. 3. Shifting the burden of proof (responsibility) The third and final aspect of the precautionary principle refers to the burden of proof for improving damage to health and the environment. It is typically regulators who are responsible for conducting risk assessments and ensuring a high level of health and environmental protection. The new EU chemicals policy seeks to reverse the ‘burden of proof’, requiring companies to do risk assessments and thereby demonstrate safe use for all chemicals (Section 2.4.1).
379
Framework for Chemical Risk Management under REACH
Appendix 2.4 - National Factors Definition of national factors following Lahusen and Munch [260]– see Figure 2.8 Policy National policy establishes regulatory goals and directs courses vÊ>VÌÆÊÌÊÛÛiÃ\ UÊ conflict settlement: redefining the roles and responsibilities of actors UÊ innovation:ÊVÕÃÊvÊiÜÊÃV>ÊÛiiÌÃÆÊiÃÌ>LÃ
iÌÊ of new relations between actors UÊ inclusion: channels of resources and communication between actors UÊ consensus formation: redefinition of networks and professional competences Rules Formal (e.g., legal) and informal institutional decisionmaking rules guide the administration of decision-making and implementation within a country. This includes the networks between actors that determines who is competent and accepted in decision-making and institutional (administrative) structures that determine how the rules are followed and enable participation of actors. Practice Practice may be defined as the practical implementation of national policy and application of decision-making rules. Practice must account for the experience, perspectives and resources of actors, especially in terms of cultures and
380
Appendix professions. Culture involves the ideas of what constitutes effective, efficient and equitable decision-making and the roles and responsibilities of the various actors. Professions describe the actors in terms of their roles, resources and responsibilities in identifying and controlling chemical risks. Risk Reduction Strategies The development, selection and implementation of risk management measures depend on aspects of policy, rules and practice. For instance, the risk reduction process requires evaluating if existing control measures are sufficient to adequately prevent or control risks [465]. Risk management strategies will vary in terms of their efficacy and efficiency in controlling a risk, which may depend on the roles and resources of the implementing actors. Risk management measures will also differ in terms of their effectiveness in achieving overall policy goals. Finally, stakeholders will have different interpretations of the equity of any risk management measure.
381
Framework for Chemical Risk Management under REACH
Appendix 3.1 - List of Interviews France Fédération Nationale CFE-CGC du Personnel d’ Encadrement des Industries Chimiques (CFE-CGC Chemie) - National Federation of Personnel in the Chemical and Allied Industries Institut National de Recherche et de Sécurité (INRS) - National Institute of Research and Security Ministère de l’Ecologie et du Développement Durable (MEDD) Ministry of Ecology and Sustainable Development Union des Industries Chimiques (UIC) - Chemical Industry Association Germany Bundesanstalt für Arbeitsschutz und Arbeitsmedizin (BAuA) - Federal Institute for Occupational Safety and Health Bundesministerium für Umwelt, Naturschutz und Reaktorsicherheit (BMU) - Federal Ministry for the Environment, Nature Conservation and Nuclear Safety Bundesministerium für Wirtschaft und Arbeit (BMWA) – Federal Ministry of Economics and Labour IG Metall Umweltbundesamt (UBA) - Federal Environment Agency Verband der Chemischen Industrie (VCI) - Association of Chemical Industries Verbraucherzentrale Budesverband (VB) - Federation of Consumer Organisations
382
Appendix Sweden Arbetsmiljoverket (AV) - Work Environment Authority Kemikalieinspektionen (KemI) - Chemicals Inspectorate Landsorganisationen i Sverige (LO) - Swedish Trade Union Confederation Naturvårdsverket - Environmental Protection Agency Plast- and Kemiföretagen - Plastics and Chemicals Federation Sveriges Konsumentråd - Swedish Consumers’ Association United Kingdom Amicus British Occupational Hygiene Society (BOHS) Department for Environment, Food and Rural Affairs (Defra) Environment Agency for England and Wales (EA) Health and Safety Executive (HSE) Health and Safety Laboratory (HSL) LGC WWF-UK EU Bureau Européen des Unions de Consommateurs (BEUC) European Consumers’ Organisation WWF International
383
Framework for Chemical Risk Management under REACH
Appendix 3.2 - Interview Questions A sample set of interview questions and probes are listed next, formulated for an interview with Germany’s Federal Institute for Occupational Health and Safety (BAuA). Policy and Administrations 1. So to start off, how would you describe current German chemicals policy? - In other words, what is its general approach to regulating chemicals? - Are there any over-riding principles? 2. Could you please describe how the current policy has come about?/Why doesn’t Germany have any set policy? - Has it recently developed or changed? - Do you think it has improvements for all stakeholders involved (i.e., who gains, who loses)? - How has Germany’s chemicals policy changed over recent years? - Why exactly do you think that Germany has adopted this current policy? 3. How would you describe Germany’s regulatory infrastructure? - In other words, what are the roles and responsibilities of the Ministries, the federal institutions, the Länder? - Are there any other relevant authorities responsible for regulating chemicals? - What about the Institute for Risk research?
384
Appendix - What’s your overall view of how the system works? 4. How does BAuA decide on how to allocate its resources? - Does this resource allocation and prioritisation often change? - What are the determining factors? - Do you think that the prioritisation mechanism is effective and efficient? Risk Reduction Strategy 5. After the results of a risk assessment, how well does the current process for developing and selecting a risk reduction strategy work within Germany? - Does the process work well? - And what would you say are generally the most influential factors? 6. How are stakeholders consulted? - Is there a stakeholder forum? 7. Could you please tell me a little more about the role and responsibility of different stakeholder groups? - What sort of information do they input into any decision-making process? - How is dialogue, communication and negotiation between various stakeholder groups achieved? (in other words, how are their concerns accounted for) - How are differences in opinion generally resolved? Are there any unresolved issues? - Are there conflicts or problems with this process?
385
Framework for Chemical Risk Management under REACH Monitoring and Enforcement 8. What are the current limitations and difficulties with monitoring and enforcing chemical risks in Germany? - Are there difficulties with regulating chemical use in SME? What about controlling professional or consumer use? What consideration is given in Germany to environmentally proactive consumer behaviour? - How well do the authorities view CAD, IPPC, WFD, etc. 9. How is responsibility in managing risks divided – between say regulators, companies, the Länder, NGO and other groups/the general public? 10. What experience is there with company non-compliance? 11. What is the general view and experience in Germany of voluntary initiatives? 12. What about taxes, are these ever considered to control chemicals? – Why or why not? EU Decision-Making 13. When it comes to decision-making regarding developing and selecting a risk reduction measure at the EU level... how is consensus reached between the Member States? What do you see as the main obstacles and opportunities? - How do you think that these obstacles can be overcome/ opportunities be promoted? - What strategies does Germany adopt for the EU decision-making process (e.g., co-operation)?
386
Appendix - How does Germany support its positions? - Does Germany see itself as having any sort of particular ‘role’? 14. Very briefly, do you think REACH is a good idea in terms of actual risk management? Why (not)? What about changes to the current structure? Will they be major?
387
Framework for Chemical Risk Management under REACH
Appendix 4.1 - National Policy of Sweden, UK and Germany This Appendix provides an overview of the national chemicals policy of Sweden, the UK and Germany. Neither France nor Germany has a national chemicals policy, but a policy document is available from the German Environment Agency (UBA). Sweden In 1999, the Swedish Parliament (Riksdag) adopted a bill establishing 15 Environmental Quality Objectives [351]. The bill follows SEPA proposals and only one objective specifically relates to chemicals, that of a non-toxic environment. The primary objective of a non-toxic environment states that ‘The environment must be free from man-made substances and metals that represent a threat to human health or biological diversity’ [544]. With regards to chemical risk management the following should be achieved within one generation [544]: UÊ /
iÊVViÌÀ>ÌÃÊvÊÃÕLÃÌ>ViÃÊÌ
>ÌÊ>ÌÕÀ>ÞÊVVÕÀÊÊÌ
iÊ environment are close to the background concentrations. UÊ /
iÊ iÛiÃÊ vÊ vÀi}Ê ÃÕLÃÌ>ViÃÊ °i°]Ê >>`i®Ê Ê Ì
iÊ environment are close to zero. UÊ "ÛiÀ>Ê iÝ«ÃÕÀiÊ Ê Ì
iÊ ÜÀÊ iÛÀiÌ]Ê Ì
iÊ iÝÌiÀ>Ê environment and the indoor environment to particularly dangerous substances is close to zero and, as regards other chemical substances, to levels that are not harmful to human health. To develop a strategy for achieving a non-toxic environment, the Swedish government established a Committee on New Guidelines on Chemicals Policy. The Committee primarily consists of academic scientists and KemI personnel. However, prior to publishing its official Chemicals Policy, the Swedish government required an analysis of the further need for control
388
Appendix instruments, such as licensing reviews and prohibitions [545]. The government also specified that the policy should be refined according to an assessment of socio-economic impacts of the proposals while accounting for industry’s own experiences [545]. Details on how this process occurred are not readily available and warrant further investigation. Examining the process of its policy formulation could yield a deeper insight into the Swedish political process. The official Chemicals Policy adopted by Parliament is not available in English. The analysis of Swedish chemicals policy has therefore been carried out on the most recent document relating to the Swedish chemicals policy, Summary of Documentation for In-Depth Evaluation of the Environmental Quality Objective of a Non-Toxic Environment [546] and the Swedish Environmental Quality Objectives [351]. In 2001, the Swedish Parliament clarified interim targets for achieving a non-toxic environment [546]. First and foremost, newly manufactured articles should be [544]: ‘….to the [fullest?] extent possible free from: UÊ V>ÀV}iV]ÊÕÌ>}iVÊ>`ÊÀi«À`ÕVÌÛi`ÃÌÕÀL}Ê ,®Ê substances by 2007, if the products are intended to be used in such a way that they are released to the eco-cycle UÊ iÜ]ÊÀ}>VÊÃÕLÃÌ>ViÃÊÌ
>ÌÊ>ÀiÊ«iÀÃÃÌiÌÊ>`ÊL>VVÕÕ>Ì}Ê (PB) as soon as possible, but not later than 2005 UÊ >``Ì>ÊÀ}>VÊÃÕLÃÌ>ViÃÊÌ
>ÌÊ>ÀiÊÛiÀÞÊ«iÀÃÃÌiÌÊ>`ÊÛiÀÞÊ bioaccumulative (VPVB), by 2010 UÊ Ì
iÀÊÀ}>VÊÃÕLÃÌ>ViÃÊÌ
>ÌÊ>ÀiÊ«iÀÃÃÌiÌÊ>`ÊL>VVÕÕ>ÌÛi]Ê by 2015 UÊ iÀVÕÀÞÊLÞÊÓääÎÆÊV>`ÕÊ>`Êi>`ÊLÞÊÓä£ä°½Ê
389
Framework for Chemical Risk Management under REACH The previous target then specifies that the products ‘containing substances’ with the properties listed above must be handled in such a way that these chemicals are not released to the environment [547]. The report continues to state explicitly that chemicals listed should not be used in production processes ‘unless the company can prove that human health and the environment will not be harmed’ [547]. In other words, the overall target advocates a mix between hazard and risk-based approaches. In cases of handling, no exposure should occur at all. An example would be in the disposal or disassembly of an electronic piece of equipment where actual substances are contained within the articles. In cases of production, manufacture and use, Sweden advocates a risk approach through exposure assessment. Regulatory instruments available to achieve the target depend on whether the definition of ‘newly manufactured’ articles only applies to products manufactured in Sweden. If so, the number of products covered is greatly reduced. In addition, voluntary agreements between regulators and Swedish industry would be far easier. Alternatively, setting eco-label criteria or applying other instruments such as Extended Producer Responsibility could be prioritised according to those products. Future interviews must ascertain the precise definition and scope of this interim target, as well as its exact relation to any longterm target. Another interim target concerns guideline values for 100 selected chemicals indicating the concentrations in the environment below which no harm is ‘expected’ to occur [350]. The longterm aim for the guideline values is to establish Environmental Quality Standards [350]. To date, 68 guideline values have been presented, but it is unclear whether this was by SEPA or KemI [350]. The next step of the research project should include a brief overview of the intrinsic hazardous properties of these chemicals. Identifying any chemicals meeting the criteria of CMR or VPVB would indicate a contradiction between the two targets.
390
Appendix A third interim target calls for the phase-out of chemicals that interfere with recycling articles. Attaining the target requires a system for identifying these chemicals. Under existing EU legislation, this falls within the scope of Extended Producer Responsibility. The final interim target relevant to the analysis seeks for articles to carry health and environmental information on included dangerous substances. Such a label is distinguishable from the eco-label and other existing Environmental Product Declarations because it presents a regulatory rather than voluntary requirement. No formal proposal for ‘hazard labels’ currently exists at the EU level, which would fall under the scope of Integrated Product Policy. Although the current REACH proposals call for industry declarations on the content of hazardous materials included in articles imported at volumes above 1 tonne per year to be in conformity with REACH registration requirements, ‘hazard labels’ on articles represent a significant burden to companies and regulators. Basically, companies would need to redesign many existing product labels. Distributors and retailers would be responsible for conducting conformity checks. Finally, regulators would be responsible for enforcing the hundred thousands of products on the EU market. Demonstrating the proportionality of these costs with the improved health and environmental benefits appears an impossible task. A further limitation to the ‘hazard labels’ arises from the fact that the labels must communicate the associated risks of the hazards to the general public. Otherwise, there is the potential that the public misinterprets the hazards as actual risks. The Swedish Environmental Quality Objectives contain a section on restrictions and phase-out, referring to current EU proposals on marketing and use restrictions [548]. The report recognises a difficulty with realising the Swedish objective of phasing-out lead, stating that there are no environmentally superior alternatives to lead in batteries [549]. Secondly, the
391
Framework for Chemical Risk Management under REACH review suggests that the phase-out of some chemicals, phthalates and other ‘harmful or potentially harmful’ plasticisers, should be on a voluntary basis from industry [549]. UK The UK Chemicals Strategy focusses on production, marketing >`ÊÕÃiÆÊÌÊ`iÃÊÌÊVÕ`iÊ>ëiVÌÃÊÀi>Ì}ÊÌÊVVÕ«>Ì>Ê health or pollution control [353]. Basically, the strategy proposes a seven-step process for the risk assessment and management of existing chemicals: 1. A Chemical Stakeholder Forum selects ‘chemicals of concern’ that require a ‘fast-track approach’ to risk assessment and/or management. 2. Industry summarises information on the priority chemicals. 3. After consultation with the Stakeholder Forum, industry prepares a draft risk reduction strategy for each chemical. 4. The forum discusses the draft risk reduction strategy of each chemical 5. Industry refines its risk reduction strategy and proposes a final version to the Stakeholder Forum. 6. The Stakeholder Forum reports its views on the final industry strategy to Ministers within six months of receiving the summary of information from industry. 7. Government decides on the necessity for risk management measures or voluntary action by industry. From an analysis of the strategy, circumstances that mandate action from industry or Government need refining. The strategy specifies that the submission of a draft risk reduction strategy from industry immediately follows submission and discussion of the available data on the chemical at the stakeholder forum.
392
Appendix The timing from the initial summary of available information (step 2) presented by industry to the final stakeholder report (step 6) is six months. Within this six-month period, industry must present a draft risk reduction strategy for the chemical (step 3) which it must then refine following consultation with the Stakeholder Forum (step 5). The final stakeholder report then consists of its views on the final risk reduction strategy (step 6). In an apparent contradiction, the strategy then states that for ‘chemicals selected through the fast track approach, industry guided by the Stakeholder Forum will produce risk management strategies by 2005’. As the Stakeholder Forum meets only twice a year, either (a) the Stakeholder Forum does not meet often enough to meet its obligations, or (b) a section of the strategy incorrectly refers to ‘summary of available information’ (step 2) instead of ‘final risk reduction strategy’ (step 5). Assuming the latter conclusion would void the apparent contradiction within the strategy that industry risk reduction strategies for all the relevant chemicals selected must be proposed by 2005. Industry is granted a maximum period of five years to implement its final proposal for a risk reduction strategy. In cases where voluntary measures are insufficient or impracticable, the strategy specifies that Government will consider alternative measures at a national or European level. The UK strategy states that in some cases hazard criteria are sufficient ‘to move immediately to consideration of risk reduction strategies’. Given that this statement implies avoiding risk assessment or further gathering of data (step 2), a brief review of Stakeholder Forum meeting reports finds no evidence for this approach. Criteria for selecting chemicals of concern for priority review include the intrinsic properties of persistence and bioaccumulation. However, the UK strategy notes that
393
Framework for Chemical Risk Management under REACH persistence and bioaccumulation are not infallible indicators of hazard. The UK supports this position via reference to naturally persistent and bioaccumulative chemicals, such as vitamins and minerals that are ‘vital for life’. Overall, the UK policy advocates a risk-benefit approach to chemical regulation. The UK chemical strategy places considerable emphasis on the important function chemicals serve in modern society. Germany Germany does not have an official chemicals policy, nor is a chemicals policy available from any of the Ministries. The only publicly available document specifically relating to national chemicals policy is a report on the Precautionary Risk Assessment and Risk Management of Chemicals produced by the Federal Environment Agency (UBA). The UBA report does not appear to receive strong support from the ministries as it fails to appear on the Ministry for the Environment’s website, although an interviewee stated that the UBA has to receive permission prior to any official publication. It is not unusual for the Ministry for Environment and the UBA to form separate environmental policy [370]. Neither is it uncommon for the UBA staff to publicly criticise the concepts put forward by the ministry [370]. The UBA publication establishes five targets for chemical management. In summary: UÊ Û`Ê Ì
iÊ `ÃV
>À}iÊ vÊ «iÀÃÃÌiÌÊ >`Ê L>VVÕÕ>Ì}]Ê or persistent and highly mobile, xenobiotics into the environment, as well as chemicals whose metabolites exhibit these properties. UÊ Û`ÊÌ
iÊ`ÃV
>À}iÊvÊV>ÀV}iV]ÊÕÌ>}iVÊÀÊÀi«À`ÕVÌÊ toxic chemicals, as well as chemicals whose metabolites exhibit these properties. UÊ ÌÀÊ Ì
iÊ >Ì
À«}iVÊ Àii>ÃiÊ vÊ >ÌÕÀ>ÞÊ VVÕÀÀ}Ê
394
Appendix persistent and bioaccumulating, or persistent and highly mobile, carcinogenic, mutagenic or reproduction toxic chemicals into the environment so that there is not an increase in geogenic or biogenic background concentration. UÊ ,i`ÕViÊ ÌÊ >ÃÊ ÜÊ >ÃÊ ÌiV
V>ÞÊ «ÃÃLiÊ Ì
iÊ Àii>ÃiÊ vÊ other (eco-)toxic chemicals (including naturally occurring substances) which do not fall into the above categories, as well as chemicals whose metabolites exhibit these properties. UÊ Û`Ê>ÊVÀi>ÃiÊvÊV
iV>Ê`ÃV
>À}iÃÊÌÊÌ
iÊiÛÀiÌÊ regardless of the effects known so far and other intrinsic properties, where high distribution and/or low exchangeability make recovery practically impossible. The first two targets correspond to controlling pollution at source, in other words through pollution prevention. The third target recognises that all chemical emissions resulting from human activities should be considered, even if the chemicals exist in the natural environment (i.e., not man-made chemicals, but can be produced as the result of human activities). In the case of the final two targets, the UBA suggest an authorisation procedure for ‘unavoidable procedures’ which should not be based on risk assessment of PEC/PNEC ratios. Essentially, the authorisation corresponds to the system proposed under REACH, but requires significantly fewer testing requirements. A particular basis of the targets relates to the fact that existing EU risk assessment procedures do not consider the persistence and accumulation of chemicals in the marine environment. To control these potential risks, the UBA identifies the need to adopt a comprehensive strategy for the protection of the marine environment. In addition to action specifically aimed at protecting the marine environment and establishing an authorisation procedure, the UBA proposes a combination of the following measures.
395
Framework for Chemical Risk Management under REACH UÊ ÝÌi`}Ê Ì
iÊ ÃÕLÃÌÌÕÌÊ «ÀV«iÊ Õ`iÀÊ Ì
iÊ iÀ>Ê Hazardous Substances Ordinance (i.e., the EC Chemical Agents Directive) to cover environmental risks. UÊ >ÀÀÞ}ÊÕÌÊÃÕLÃÌ>ViÊyÜÊ>>ÞÃÃÊvÊ«ÀÌ>ÌÊiVVÊ sectors to provide users and consumers with information necessary to develop less polluting and more resource efficient product and processes. UÊ 1«`>Ì}ÊÌ
iÊVÀÌiÀ>ÊvÀÊiV>Li}ÊÜÌ
Ê>ÊV«Ài
iÃÛiÊ and uniform framework. UÊ Ìi}À>Ì}Ê iÛÀiÌ>Ê ÀiµÕÀiiÌÃÊ Ê V
iV>ÃÊ Ê product standards (e.g., building products). UÊ ÝÌi`}Ê>Li}ÊÀiµÕÀiiÌÃÊÌÊVÕ`iÊ`iV>À}ÊViÀÌ>Ê chemical ingredient in articles. The above measures make maximum use of existing EU legislative frameworks. Carrying out substance flow analysis essentially supplements current use of risk assessment and LCAs. Although aspects of substance flow analysis feature in risk assessments, it is not specifically required under any existing framework of EU legislation, including the future implementation of REACH. The UBA proposal for its application also seeks to extend its application from regulators to companies and consumers. Implementing this measure appears to necessitate a reconfiguration of an existing framework or the development of a new one.
396
Appendix
Appendix 4.2 - Trust, Media and Actors: Swedish Chemical Wars Sweden is a veteran of chemical wars. While the interviews did not yield evidence of hardline campaigning by Swedish NGO, the distribution of T-shirts by the Swedish Nature Conservation Society during debates in Brussels on penta-BDE stating ‘I am a guinea pig for the chemical industry: ban brominated flame retardants’ suggests a ‘no-holds barred’ approach to challenging the industry. Swedish industry association interviewees recalled with chagrin that a landmark offensive for campaigning against the chemical industry dates to the physical destruction of a chemical factory. Around the same time as the publication of the Swedish translation of Silent Sprint in 1963 [550], the government decided to physically explode a chemical site that had been producing Agent Orange. Responding to media attention that Agent Orange produced in some countries was contaminated with POP substances even more dangerous than the active chemical itself, ignoring the fact that the site was not responsible for producing any such contaminated product, and irrespective of environmentally ‘friendly’ methods of disassembling the plant, media attention overrode any political rationality. The destruction of the plant was aired on national television. The Swedish industry interviewees also described that the media will present news of chemical risks based on limited facts. For instance, a family living near a chemical facility featured in the news as being exposed to a chemical that was causing uncomfortable skin rashes. It later emerged that the family was suffering from scabies, but this was deemed too embarrassing to the family to report in any future media coverage. In 2002, studies indicated that a significant percentage of the Swedish population is frequently exposed to acrylamide in food (see [551]). To limit public panic, the Swedish National Food Administration (SLV) held a press conference. One hundred and fifty journalists and the National Television subsequently covered the press conference. Unsurprisingly, a public scare ensued [552]. The public response to Sweden’s phase-out of mercury provides another example of poor risk communication. One Swedish regulator reported that
397
Framework for Chemical Risk Management under REACH KemI’s renowned stringent regulatory requirements on the use of mercury resulted in public rumors circulating in the mass media and subsequent outrage that dead bodies may be exhumed to remove mercury fillings in teeth [553]!
398
Appendix
Appendix 5.1 - Regulatory Options EU Authorisations Scope UÊ /
iÊ>ÕÌ
ÀÃ>ÌÊ«ÀViÃÃÊÃ
Õ`ÊLiÊ«À>ÀÞÊÕÃi`ÊÌÊi>VÌÊ immediate bans as necessary to stabilise or reduce current concentrations of specific substances in environmental media across Member States. UÊ ÕÌ
ÀÃ>ÌÊÃ
Õ`ÊÀi}Õ>ÌiÊÕÌ«iÊÃÕÀViÃÊvÊiÛÀiÌ>Ê emissions from manufacturing or products. UÊ ÀÊ`ÕÃÌÀ>Ê«ÀViÃÃiÃ]ÊÌ
iÊ>ÕÌ
ÀÃ>ÌÊ«ÀViÃÃÊÃ
Õ`ÊLiÊ limited to instances when the reporting of monitoring data from point source emissions with a high potential for causing trans-boundary pollution cannot demonstrate ‘adequate control’ – see Definitions in Appendix 1.1. Procedure 1. The Annex XV dossier should contain data that characterises the exposed ecosystems and/or populations with monitoring data indicative of current exposure levels. 2. Evidence should be provided demonstrating that current levels of exposure to the environment and/or human health are increasing. 3. If environmental concentrations result from point sources of trans-boundary pollution, a programme of environmental monitoring corresponding to set emission level values should be proposed to control the point source emission. 4. If environmental concentrations result from product use or disposal, data should be provided indicating that production or use levels are anticipated in the near future.
399
Framework for Chemical Risk Management under REACH 5. Once submitted, the Annex XV dossier should be compared with other proposals for substance inclusion in the authorisation process. If a substance is consequently identified as being of comparatively very low priority, other regulatory options should be proposed and the Annex XV dossier revised accordingly. Effectiveness
As the subsequent issuing authorisations will depend on the provision of data to demonstrate that a substance use is adequately controlled, authorisation under REACH can be considered to be effective. For substances that may not meet the adequate control criteria – i.e., VPVB, PBT, non-threshold CMR or endocrine disruptors – but serve critical purposes to manufacturing or product use that might enable them to be authorised, effectiveness of stringent regulatory control would need to be demonstrated based on monitoring schemes.
Practicality
The practicality of inclusion of a substance under the authorisation process will be subject to the results of a socio-economic analysis.
Monitorability Products not intended for consumer use should be clearly marked (e.g., ‘for professional use only’). If the authorisation includes products for consumer or professional use, reference to product test methods and at least one supplier source of pure material for product testing should be provided unless the ban is set at detection limit values. SEA
400
Socio-economic analysis would follow the REACH Annex XVI procedure.
Appendix Member State Permitting Scope UÊ iLiÀÊ -Ì>ÌiÊ ¼«iÀÌÌ}½Ê ÀiviÀÃÊ ÌÊ iÝÃÌ}Ê ÀÊ >ÌV«>Ìi`Ê authorisation, approval, certification or licensing schemes. UÊ i>ÃÕÀiÃÊ«iiÌi`ÊÌ
ÀÕ}
Ê>Ì>Ê«iÀÌÌ}ÊÃV
iiÃÊ could include occupational exposures for industrial and/or professional uses. In particular, Member States should always have the option of operating national permitting schemes when this is deemed necessary to achieve levels of protection for occupational exposures in industrial settings (following Article 138 of the EC Treaty). UÊ iÝÊ 86Ê `ÃÃiÀÃÊ Ã
Õ`Ê >VVÕÌÊ vÀÊ iÝÃÌ}Ê «iÀÌÌ}Ê schemes operated by Member States. Otherwise, variations in existing national legislation could delay reaching consensus on the need for EU action and even act against the subsidiarity principle. The permitting process would be particularly relevant when evidence of the risk criterion for ‘national dimension’ is fulfilled (see step 1a). UÊ *iÀÌÌ}ÊÃV
iiÃÊÜÕ`ÊVÕ`iÊ«À`ÕVÌÊViÃ}ÊÃV
iiÃÊ notified under Directive 98/34, for example processes and products authorised for water treatment processes [487]. Procedure 1. The Annex XV dossier should detail the relevant existing national permitting schemes. 2. The Annex XV dossier should provide evidence that the ‘national dimension’ risk criterion in step 1 applies. Equally, Member States should provide this evidence when reviewing the dossier. 3. Consideration should be given to the necessity for harmonisation of the identified national permitting schemes by the relevant DG of the European Commission. 401
Framework for Chemical Risk Management under REACH Effectiveness
Operation of Member State permitting schemes is primarily dependent on national implementation, especially if legislation based on Articles 138 or 175 of the EC Treaty. National permitting may be supplemented by a proposed target-setting mechanism to ensure overall risk reduction (see Target-setting). Any trans-boundary pollution resulting from the operation of a national permitting scheme would need to be challenged or justified through other EU regulatory or legislative processes.
Practicality
Permitting schemes relating to technical standards must be notified to, and approved by, the European Commission under Directive 98/34. Co-ordinating this process with decision-making under REACH is therefore not expected to be administratively burdensome to Member States or the Commission.
Monitorability Monitoring national permitting schemes would not be considered under REACH. SEA
402
Socio-economic considerations for measures aimed at achieving occupational or environmental protection set under legislative frameworks based on Articles 138 or 175 of the EC Treaty would be outside the scope of REACH. For standards under Directive 98/34 proportionality must already be demonstrated by the Member State through the notification procedure.
Appendix Restrictions on Industrial Uses Scope UÊ ,iÃÌÀVÌÃÊÊ`ÕÃÌÀ>ÊÕÃiÃÊÃ
Õ`Ê>««ÞÊÌÊVÌÀÊ«ÌÊ source emissions resulting in trans-boundary pollution that have been demonstrated to be a major contribution to the cause of reported negative impacts to human health or the environment. UÊ ÀÊ«iÀ>LÌÞ]ÊvÊÌ
iÊV>ÌÊvÊ«ÌÊÃÕÀViÃÊiÝVii`ÃÊÃiÛiÀ>Ê Member States or if several Member States or countries outside the EU are experiencing effects of the pollution, the emissions should be controlled through the authorisation process. UÊ Ì
Õ}
ÊiÃÌ>LÃ
}ÊÌ
iÊiViÃÃ>ÀÞÊVÌÀÃÊvÀÊ«ÌÊÃÕÀViÊ emissions should only occur at national levels, the results of these activities should be reported to decision-making under REACH. Procedure 1. The point sources should be detailed in terms of specific locations. 2. Any Member State(s) experiencing the results of the transboundary pollution would be responsible for preparing the risk characterisation section of the Annex XV dossier (unless located outside the EU). 3. Member State(s) from which the point source emission(s) emanate should propose monitoring and reporting programmes that demonstrate adequate risk reduction. 4. If compliance cannot be demonstrated, within a certain time period, then the relevant use(s) should be subject to the authorisation procedure.
403
Framework for Chemical Risk Management under REACH Effectiveness
The process would contain a mechanism to ensure monitoring and reporting duties.
Practicality
Monitoring and reporting that would be established by the relevant (polluting) Member State(s) could include data from the facility and/or environmental monitoring data from the relevant regulatory authorities.
Monitorability The Member State(s) experiencing the impact of the pollution should have the option of submitting its own monitoring data, including epidemiological data, to verify that the risks are being adequately controlled at source. SEA
Socio-economic analysis would follow the procedure detailed in REACH Annex XVI which could then serve as the basis for deciding on the emission limit values and frequency of reviewing reported monitoring data. Restrictions on Professional and Consumer Uses
Scope Restrictions on the marketing and use of substances for professional and consumer uses would follow the set of recommendations in Table 5.4, Section 5.3.1, page 149. Procedure 1. Restrictions would follow the general format and structure of the recommendations. 2. During socio-economic analysis of the restriction, companies would be able to provide evidence of use of personal protective equipment (PPE), existing licensing or training schemes, or
404
Appendix product take-back activities. Any exceptions would be subject to a review of the potential trade barriers caused by enacting regulatory requirements that depend on such schemes. Effectiveness
Most restrictions would involve bans on substance uses. Several possible restrictions could include requirements for provision of correct PPE which would reduce any chance of incorrect PPE usage. Safety measures necessary for following the restrictions would need to be detailed or referenced in Safety Data Sheets providing a method for communicating regulatory obligations. Being preventive measures, the effectiveness of any licensing, certification, or take-back schemes would however depend on compliance and correct product use.
Practicality
Recommendations would be communicated to companies prior to registration and provide a general format for companies submitting substance registration dossiers for professional and consumer uses as well as apply as a general rule for safe use of preparations. This can be expected to facilitate the prioritisation of compliance activities by companies and regulators at an early stage of REACH.
Monitorability As with authorisation, products not intended for consumer use should be clearly marked (e.g., ‘for professional use only’). If the restriction includes products for consumer or professional use, reference to product test methods and at least one supplier source of pure material for product testing should be provided unless the ban is set at detection limit values.
405
Framework for Chemical Risk Management under REACH SEA
Socio-economic analysis would follow the procedure detailed in REACH Annex XVI and must include an assessment of potential trade barriers that could be caused by enacting any restrictions that depend on licensing, certification or take-back schemes. Targets
Scope UÊ />À}iÌÃiÌÌ}ÊÜÕ`Ê>««ÞÊÜ
iÊÌ
iÀiÊ>ÀiÊ>ÞÊÃÕÀViÃÊvÀÊ exposure to a single substance or chemical grouping and a corresponding high number of potential uses. UÊ Ì
Õ}
Ê Ì>À}iÌÃÊ ÜÕ`Ê LiÊ «iiÌi`Ê ÕÃ}Ê i}Ã>ÌÊ outside REACH, decisions on whether to set targets would affect decisions on restrictions amd authorisations. (note that restrictions can be applied in combination with target-setting.) UÊ />À}iÌÃiÌÌ}ÊVÕ`Ê>Û`Ê`i>ÞÃÊÊ`iVÃ>}ÊÀiÃÕÌ}Ê from the need to (i) gather further data necessary to resolve different interpretations of risk assessment data, or (ii) conduct detailed socio-economic analyses of risk reduction measures. Procedure 1. Target-setting and subsequent reporting should reflect national differences in production, use, and environmental conditions, as well as alternative regulatory measures and policy instruments that can be implemented to achieve risk reduction. 2. Targets would be based on use-specific volumes as a proxy for human and environmental exposures, occupational health statistics, industry performance indicators, emission registers or other benchmarking schemes.
406
Appendix 3. Publishing substances subject to targets on the REACH-IT website could promote the communication and co-ordination of a wide set of risk management and policy measures within industry subgroups or NGO at national and EU levels. Effectiveness
Achieving overall risk reduction would be left to the devices of the Member States who would retain the ability to enact measures beyond EU target by using different regulatory tools, including taxes and voluntary initiatives. Targets would establish minimal levels of risk reduction.
Practicality
The practicality and acceptability of implementing measures to achieve targets depends on the regulatory administrative structure and industrial landscape of a country.
Monitorability Reporting of performance towards targets must reflect different regulatory tools which could facilitate reporting under other existing EU legislative frameworks. SEA
Socio-economic analysis would only be conducted at the national level for selecting cost-effective strategies to meet targets, allowing Member States to decide on the level of data necessary to set or justify any decision.
407
A
bbreviations and Acronyms
4CC
4-Chloro-o-cresol
AA
Acrylic acid
ADMET
Absorption, distribution, metabolism, excretion and toxicity
AF
Assessment Factor(s)
AISE
International Soap, Detergent and Maintenance Products Association
AP
Tris(aziridinyl) phosphinoxide
BA
Benzene, C10–13-alkyl derivatives
BAA
Volatile esters of bromoacetic acids
BAM
Federal Institute for Materials Research and Testing, Germany
BAT
Best-available technique/technology
BAuA
Federal Institute for Occupational Safety and Health, Germany
BBP
Butylbenzyl phthalate
BfR
Federal Institute for Risk, Germany
BMAS
Ministry of Labour and Social Affairs
BMU
Ministry for the Environment, Germany
BMV
Ministry for Consumer Protection, Germany
BMW
Ministry of Economy
BMWA
Ministry of Economics and Labour, Germany
BSE
Bovine spongiform encephalopathy
CAD
Chemical agents directive
409
Framework for Chemical Risk Management under REACH Cefic
European chemical industry council
CEN
European Committee for Standardisation
CMR
Carcinogenic, mutagenic or reprotoxic
CONSEXPO Model to assess consumer exposure CSR
Chemical safety report(s)
DBP
Dibutyl phthalate
DBPP
Tris (2, 3-dibromopropyl) phosphate
DCB
Dichlorobenzene
DCM
Dichloromethane
DDAC
Dimethyldioctadecylammonium chloride
deca-BDE
Decabromodiphenyl ether
Defra
Department of the Environment, Food and Rural Affairs (UK)
DEGBE
2-(2-butoxyethoxy)ethanol
DEGME
2-(2-methoxyethoxy)ethanol
DEHP
Di (2-ethylhexyl) phthalate
DERMAL
Model to assess dermal exposure
DG
Directorate General
DGS
Directorate General for Consumer Health
DIDP
Di-iso-decyl phthalate
DINP
Di-iso-nonyl phthalate
DINP
Di-iso-nonyl phthalate
DIREN
Regional environment departments, france
DIY
Do-it-yourself
DMEL
Derived minimal-effect level
DMS
Dimethyl sulfate
DNA
Deoxyribonucleic acid
DNEL
Derived no-effect level(s)
DNOP
Di-n-octyl phthalate
410
Abbreviations and Acronyms DRIRE
Regional Departments for Industry, Research and the Environment, France
DTI
Department of Trade and Industry (UK)
EA
Environment Agency of England and Wales (UK)
EAA
Ethyl acetoacetate
EASE
Model to assess occupational inhalation
EC
European community
ECA
European Court of Auditors
EEA
European Environment Agency or a chemical?
EHS
?
EINECS
European Inventory of Existing Commercial Chemical Substances
EMAS
Ecomanagement and Audit Scheme
EMPL
Employment, Social Affairs & Equal Opportunities DG
ENTR
Enterprise & Industry DG
ENV
Environment DG
EPER
European pollution emission register
EU
European union
EURAM
EU Risk Ranking Method
EUSES
EU System for the Evaluation of Substances
FECCIA
European Federation of Managerial Staff in the Chemical and Allied Industries
GDP
Gross domestic product
GHS
Globally harmonised system
H2O2
Hydrogen peroxide(s)
HERA
Human and Environmental Risk Assessment
HF
Hydrogen fluoride
HPV
High production volume (i.e., production q1000 tonne per year)
HSE
Health and Safety Executive (UK) 411
Framework for Chemical Risk Management under REACH IG BAU
Industriegewerkschaft Bauen-Agrar-Umwelt
IG Metall
Union for those working in the metal-working industries
IGBCE
Chemical industry trade union
IMC
Inter-ministerial committee, france
INERIS
National Institute for Industrial Environment and Safety, France
INRS
National Institute of Research and Security, France
IPPC
Integrated Pollution Prevention and Control
IQ
Intelligence quotient
KemI
Chemicals inspectorate, sweden
LCA
Life cycle assessment(s)
LD50
Lethal dose for 50% of a test population
LEV
Local exhaust ventilation(s)
MAA
Methacrylic acid
MCC
Ministry of Con sumption
MCCP
Medium chained chlorinated paraffin(s) (C14-C16-alkyl derivatives)
MEDD
Ministry for Ecology and Sustainable Development, France
MEFI
Ministry of Economics, Finance and Industry, France
MESA
Ministry of Employment & Social Affairs, France
MI
Ministry of the Interior, France
MMA
Methyl methacrylate
MTBE
Methyl tert-butyl ether
N/A
Not available
NGO
Non-Governmental Organisation(s)
NO(A)EL
No observable (adverse) effect level
NP
Nonylphenol
NPE
Nonylphenol ethoxylate
412
Abbreviations and Acronyms octa-BDE
Octabromodiphenyl ether
OECD
Organisation for Economic Co-operation and Development
OSHA
European Agency for Safety and Health at Work
PB
Persistent and bioaccumulating
PBB
Polybromobiphenyl(s)
PBT
Persistent, bioaccumulative and toxic
PEC
Predicted environmental concentration
penta-BDE
Pentabromodiphenyl ether
PFOS
Perfluorooctane sulfonate
PNEC
Predicted no-effect concentration
POP
Persistent organic pollutant
PPE
Personal protective equipment
PPORD
Process/product-oriented research and development
PRODUCE
Piloting REACH on downstream use communication in Europe
QSAR
Quantitative Structure-Activity Relationship(s)
R&D
Research and Development
RCEP
Royal Commission on Environmental Pollution (UK)
REACH
Registration, Evaluation And Authorisation of Chemicals
REACH-IT
REACH-Information Technology
RIP
REACH implementation project(s)
RRS
Risk reduction strategy(ies)
SAICM
UNEP Strategic Approach to International Chemicals Management
SANCO
Directorate General for Health and Consumer Affairs
SCCP
Short-chained chlorinated paraffin(s) C10–C13-alkyl derivatives 413
Framework for Chemical Risk Management under REACH SDS
Safety data sheet(s)
SEA
Socio-economic analysis
SEPA
Swedish Environmental Protection Agency
SLV
Swedish National Food Administration
SME
Small and medium-sized enterprise(s)
SOMS
Dutch Strategy on Management of Substances
SPORT
Strategic Project on REACH testing
SVHC
Substance(s) of very high concern
SWEA
Swedish Work Environment Agency
TA
Trade association
TCB
Trichlorobenzene
TCE
Trichloroethylene
TEGEWA
German Textile Trade Association
TGD
Technical guidance document(s)
tpy
Tonnes per year
TRGS
Technical Rules for Hazardous Substances
UBA
Federal Environment Agency, Germany
UK
United Kingdom
UNEP
United Nations Environmenal Programme
VCI
Verband der Chemischen Industrie eV (German Chemical Industry Federation)
VER.DI
Vereinte Dienstleistungsgewerkschaft
VPVB
Very persistent and very bioaccumulative
VROM
Dutch Ministry of Housing, Spatial Planning and the Environment
VZBZ
German Federation of Consumer Associations
WFD
Water Framework Directive
WTO
World Trade Organisation
WWK-UK
Worldwide Fund for Nature - UK
414
INDEX
Index Terms
Links
A ADMET PROGRAMMES
33
45
ACTION INDICATOR RATING AMES TEST
258 186
ANALYTICAL FRAMEWORK ASSESSMENT FACTORS
88 34
B BIOACCUMULATIVE SUBSTANCES
23
BIOASSAYS
31
BIOMARKERS
36
40
128
133
BOTTOM-UP APPROACHES
147
BOTTOM-UP IMPLEMENTATION
149
BOTTOM-UP REACH IMPLEMENTATION
150
This page has been reformatted by Knovel to provide easier navigation.
283
Index Terms
Links
BOTTOM-UP REGULATORY APPROACHES
237
C CAP-AND-TRADE SYSTEM
53
CARCINOGENIC MUTAGENIC OR REPROTOXIC SUBSTANCES
73
186
188
203
113
244 CARCINOGENICITY
60
CARCINOGENS
31
63
65
135
186
263
264
269
278
CASE-BY-CASE APPROACH CE MARKING
169
CENTRE FOR ENVIRONMENTAL STRATEGY
78
CHEMICAL EXPOSURE
25
CHEMICAL INDUSTRY
12
21
23
77
84
91
106
108
139
144
152
168
170
212
286
CHEMICAL INTOLERANCE
40
This page has been reformatted by Knovel to provide easier navigation.
Index Terms CHEMICAL PRODUCTS
Links 10
CHEMICAL REGULATORY SYSTEMS CHEMICAL RISK
172 23
38
41
47
50
60
65
80
88
89
105
128
142
156
161
167
171
175
190
215
233
239
277
278
76
91
229
5
9
13
16
24
62
67
84
88
91
104
109
116
124
126
128
133
135
141
151
207
215
240
273
277
283
286
290
70
88
139
149
167
169
190
238
269
275
285
292
69
169
286 ASSESSMENT
45
MEASURES
15
MANAGEMENT
292 MANAGEMENT DECISION-MAKING CHEMICAL SAFETY
120
CHEMICAL SAFETY REPORT
This page has been reformatted by Knovel to provide easier navigation.
Index Terms CHEMICAL AGENCY
POLICY
RISKS
Links 30 148
151
169
171
178
205
24
111
114
116
143
146
149
114
146
75
CHEMICALS REGULATION EXISTING CHEMOPHOBIC CIVIL ‘CODAL’ LAW
44 41 137
CLASSIFICATION LABELLING AND PACKAGING REGULATION
28
COMMAND AND CONTROL
15
REGULATION
51
COMMON LAW
137
CONSUMER EXPOSURE (CONSEXPO)
37
CONSUMER RISK MANAGEMENT
123
CONSUMER RISKS
199
CORPORATIST MODEL
116
COST-BENEFIT ANALYSES
133
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
D DECISION-MAKING MATRIX
197
202
229
242
254
264
269
284
291 DECISION-MAKING PROCESS
233
242
175
177
205
234
239
243
265
269
275
277
279
283
35
39
72
200
34
35
38
39
40
72
73
184
192
193
196
200
213
219
249
DECISION-MAKING RULES
291 DECISION-MAKING WHEEL
161
DEPOSIT-REFUND SCHEMES
57
DERIVED MINIMALEFFECT LEVEL DERIVED NO EFFECT LEVEL
DERMAL EXPOSURE
37
DICHLOROMETHANE
164
281
DI(2-ETHYLHEXYL) PHTHALATE
48
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
DOSE SPACING
34
DOSE-RESPONSE CURVES
32
DUTCH CHEMICALS POLICY
241
DUTCH DECISIONMAKING RULES
241
DUTCH MINISTRY OF HOUSING SPATIAL PLANNING AND THE ENVIRONMENT
234
DUTCH STRATEGY ON MANAGEMENT OF SUBSTANCES
172
210
211
241
190
E EC LEGISLATIVE DRAFTING
175
EC RECOMMENDATION
184
EC TREATY
237
EC TREATY
3
243
EC LAW
89
EC LEGISLATION
58
94
3
52
164
193
263
282
EC TREATY
ECOMANAGEMENT AND AUDIT SCHEME
54
ECOSYSTEMS NORMAL
27
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
EFFECT LEVELS BOUNDED EMISSIONS SPECIFIC ENDOCRINE DISRUPTER
185 199 33
182
186
188
258 ENVIRONMENTAL EXPOSURE
36
ENVIRONMENTAL PROTECTION
111
117
26
40
164
248
255
261
164
248
255
258
ENVIRONMENTAL RISK ASSESSMENT ETHANOL 2-(2BUTOXYETHOXY) ETHANOL 2-(2METHOXYETHOXY)
260 EU AUTHORISATION EU CHEMICAL INDUSTRY
EU CHEMICAL LAW
247
254
24
50
80
109
244
274
278
280
59
139
150
235
277
288
1
15
69
EU CHEMICAL REGULATION
157
EU CHEMICAL RISK MANAGEMENT
DECISION-MAKING EU CHEMICALS AGENCY
145
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
EU CHEMICALS CONTROL
42
EU CHEMICALS LEGISLATION EU CHEMICALS POLICY
155 1
19
144
147
150
156
168
274
14
16
20
43
66
81
83
85
90
93
98
104
106
121
129
134
147
150
153
157
159
163
167
174
179
210
228
242
245
252
254
255
261
263
267
274
275
280
290
282 EU CHEMICALS STRATEGY EU DECISION-MAKING
167
EU ENVIRONMENTAL LEGISLATION
95
EU ENVIRONMENTAL POLICY EU INDUSTRY EU LEGISLATION
145 99
238
3
66
79
85
139
169
177
193
209
234
243
252
260 265 273 This page has been reformatted by Knovel to provide easier navigation.
Index Terms EU LEVELS
EU MARKET
EU MEMBER STATES
Links 1
2
83
89
92
94
105
115
120
134
145
150
153
157
159
164
168
170
190
193
203
206
218
229
235
243
251
253
255
260
262
267
268
271
274
282
287
292
44
73
167
187
209
233
244
100
206
218
280
84
93
EU NATIONAL LEGISLATION EU POLICY EU REGULATION
115 279
281
57
81
95 EU REGULATORY DECISION-MAKING
155
EU REGULATORY PROCESS
164
EU REGULATORY RISK MANAGEMENT
242
EU REGULATORY RISK REDUCTION PROCESSES
181
This page has been reformatted by Knovel to provide easier navigation.
Index Terms EU RISK ASSESSMENT
Links 37
42
123
165
200
244
251
262
18
98
163
164
171
269
274
277
DECISION-MAKING
18
97
214
266
POLICY
18
134 EU RISK MANAGEMENT
EU RISK RANKING MODEL EU RISK REDUCTION
210 250
254
93
98
156
158
163
164
244
277
148
151
156
166
177
179
185
196
205
234
239
277
EU RISK REDUCTION STRATEGIES
291 TECHNICAL GUIDANCE
196
EU SUBSTANCEBY-SUBSTANCE REGULATORY APPROACH
283
EUROPEAN CHEMICALS AGENCY
291 EUROPEAN POLLUTION EMISSION REGISTER
107
EUROPEAN UNION DECISION-MAKING 273 This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
EUROPEAN UNION REGULATORS
1
EUROPEAN UNION SYSTEM FOR THE EVALUATION OF SABSTANCES
37
EXPOSURE ACTION INDICATOR
259
EXPOSURE ASSESSMENTS
36
EXPOSURE INDICATOR
221
223
227
EXPOSURE-BASED APPROACHES
281
F FUNNELLING TECHNIQUE FUTURE RISKS
92 199
G GENERAL RISKS
199
GENOTOXIC CARCINOGENS
182
GERMAN CHEMICAL INDUSTRY
145
GERMAN RISK MANAGEMENT
112
This page has been reformatted by Knovel to provide easier navigation.
230
Index Terms
Links
GLOBALLY HARMONISED SYSTEM
28
GROSS DOMESTIC PRODUCT
21
106
H HAMPTON REVIEW RISKBASED STRATEGY
128
HAZARD (TOXICITY) INDICTOR
221
231
275
HAZARD AND EXPOSURE
25
HAZARD ASSESSMENTS
26
42
156
158
166
172
181
217
256
274
281
226
230
126
152
HAZARD CHARACTERISATION
27
HAZARD IDENTIFICATION HAZARD INDICATOR
27 222
HAZARD STATEMENTS
28
HAZARD TESTING
45
HAZARD-BASED APPROACH
113 262
HAZARD-BASED RISK ASSESSMENTS
239
This page has been reformatted by Knovel to provide easier navigation.
157
Index Terms
Links
HIGH PRODUCTION VOLUME CHEMICALS
43
I IMMEDIATE (HIGHLEVEL) RISKS IN SILICO
IN VITRO
199 72
173
256
287
31
33
183
256
183
185
45
72
INSTITUTE FOR THE INDUSTRIAL ENVIRONMENT AND SAFETY
118
INSTITUTE OF RISK ASSESSMENT
123
INTEGRATED POLLUTION PREVENTION AND CONTROL
69
INTERVIEW SELECTION
91
INTERVIEW TECHNIQUE
92
192
INTERVIEWEE RESPONSE VALIDATION ISO 14001
96 54
K KEML’S DECISION-MAKING PROCESSES
124
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
L LEGAL SYSTEMS
137
149
LEVEL-1 IN VITRO MUTAGENICITY TESTING
60
LEVEL-2 IN VIVO STUDIES
60
LIFE CYCLE ASSESSMENTS
47
113
LOCAL EXHAUST VENTILATION
63
M MANAGEMENT OBSTACLES
41
MANAGEMENT STRUCTURE
41
MANAGEMENT TOOL HANDS-ON RISK
268
MAN-MADE CHEMICALS
22
MATRIX SYSTEM
125
MEDIUM-CHAIN CHLORINATED PARAFFINS
134
250
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
MINISTRY FOR EMPLOYMENT AND SOCIAL AFFAIRS
118
MINISTRY OF ECOLOGY AND SUSTAINABLE DEVELOPMENT
118
MINISTRY OF ECONOMICS FINANCE AND INDUSTRY
118
MONITORING NETWORK
192
206
207
229
234
242
245
260
31
128
135
186
MUTAGENICITY
N NANOTECHNOLOGY
41
NATIONAL INSTITUTE OF RESEARCH AND SECURITY
118
NATIONAL REGULATORY AUTHORITIES
122
NATURALLY OCCURRING CHEMICAL
22
NO OBSERVABLE (ADVERSE) EFFECT LEVEL (NO(A)EL) ACUTE
30
34
32
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
NO OBSERVABLE (Cont.) CHRONIC NO-EFFECT LEVEL
32 32
38
O OCCUPATIONAL INHALATION
37
ORGANISATION FOR ECONOMIC COOPERATION AND DEVELOPMENT
42
139
157
175
178
181
192
233
242
252
292
71
73
186
287
35
46
132
186
189
258
279
HAZARD ASSESSMENT METHODOLOGIES TONNAGE SYSTEM
31 44
P PERMISSIBLE USES
PERSISTENT BIOACCUMULATIVE AND TOXIC CHARACTERISTICS PERSONAL PROTECTIVE EQUIPMENT
PHTHALATES PLACEBO EFFECT
139 41
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
PRECAUTIONARY PRINCIPLE
137
PRECAUTIONARY STATEMENTS
28
PREDICTED NO-EFFECT CONCENTRATION
34
38
53
72
193
196
200
219
210
211
227
280
259
291
PREVENTION AND RECYCLING OF WASTE STRATEGY
56
PRINCIPLE OF SUSTAINABLE DEVELOPMENT
5
PRIORITISATION METHOD PRIORITISATION SCHEME PRIORITSATION RANKINGS
255
PROBABILITY INDICATORS
223
PROFESSIONAL USER RISKS
199
Q QUANTITATIVE STRUCTURE-ACTIVITY RELATIONSHIPS
26
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
R REACH
1
16
39
67
71
79
83
85
90
95
97
105
111
120
130
143
150
153
155
158
166
182
190
196
202
205
209
215
218
220
224
229
233
237
242
244
249
251
253
256
260
262
268
273
277
281
286
288
180
REACH AUTHORISATION PROCESS
191
REACH DECISIONMAKING
148
277
19
34
97
183
195
291
159
171
242
REACH PROPOSAL
78
275
REACH REGULATION
67
69
181
283
REACH IMPLEMENTATION PROGRAMMES
REACH LEGISLATION
71
REACH RESTRICTION SYSTEM
249
This page has been reformatted by Knovel to provide easier navigation.
75
Index Terms
Links
REACH RISK REDUCTION DECISION-MAKING
192
REACH TECHNICAL GUIDANCE DOCUMENTS
210
287
RECENT DEVELOPMENTS IN REGULATION RECHTSSTAAT
55 112
RECOMMENDATION EQUIVALENTS
RECOMMENDATIONS
185
187
190
245
251
265
179
184
239
243
265
268
270
277
279
283
291
220
223
226
231
255
258
184
264
REGULATORY ACTION INDICATORS
REGULATORY APPROACHES
59
REGULATORY DECISIONMAKING WHEEL REGULATORY PROCESS
57 160 51
REGULATORY RECOMMENDATIONS REGULATORY REFORM
175 155
REGULATORY RISK COMMUNICATION 169 This page has been reformatted by Knovel to provide easier navigation.
229
Index Terms
Links
REGULATORY RISK MANAGEMENT
80
RELATIVISM CONSTRAINED
7
89
RELATIVISM UNCONSTRAINED
7
REPRODUCTIVE TOXICITY
29
REPROTOXIC
32
128
RESEARCH DESIGN
83
86
86
98
135
186
RESEARCH METHODOLOGY RESEARCH PROJECT SCOPE
10
RISK ANALYSIS
6
8
42
284
RISK ASSESSMENT
4
9
12
16
26
29
33
37
38
41
43
47
51
58
61
69
71
86
93
96
99
111
113
126
134
146
149
157
164
168
172
177
183
194
196
206
209
214
222
228
234
237
239
241
245
248
251
254
264 281 287 This page has been reformatted by Knovel to provide easier navigation.
292
Index Terms
Links
RISK-BENEFIT BRITISH APPROACHES
163
RISK CHARACTERISATION RISK COMMUNICATION RISK DATA GENERATION RISK DEFINITIONS RISK DESCRIPTORS RISK MANAGEMENT
38 43
76
241 6 199 1
8
11
13
16
19
29
39
41
47
50
58
63
65
69
76
81
83
88
95
99
101
109
115
121
124
126
128
146
149
152
156
158
161
163
165
171
174
177
181
184
190
192
196
205
209
212
214
216
224
233
235
237
240
245
248
259
264
269
271
274
279
287
290
292 RISK MODEL
8
RISK OBJECTIVE
6
RISK PERCEIVED 7 This page has been reformatted by Knovel to provide easier navigation.
Index Terms RISK REDUCTION
Links 19
50
53
60
62
89
92
128
170
179
191
194
199
200
203
205
209
233
247
250
261 RISK REDUCTION DECISION-MAKING
197
RISK REDUCTION MEASURES
160
167
13
98
161
195
4
19
66
84
110
112
134
151
157
262
265
275
62
64
93
118
155
158
160
196
203
252
264
268
270
287
292
67
69
RISK REDUCTION STRATEGY RISK TOLERABILITY RISK-BENEFIT
290 REGULATION
114
RISK-REDUCTION MEASURES
RISK-REDUCTION STRATEGY RISK-REDUCTION
90 62 165
This page has been reformatted by Knovel to provide easier navigation.
121
Index Terms
Links
S SAFETY COMMUNICATION PROGRAMMES
240
SAFETY COMMUNICATION SCHEMES
241
SAFETY DATA SHEETS
49
SAFETY FACTORS
34
SAFETY PHRASES
28
169
234
286
SHORT-CHAINED CHLORINATED PARAFFINS
134
250
SCIENTIFIC RISK ASSESSMENTS SOCIAL AMPLIFICATION SOCIAL MOBILISATION
168 8
212
212
217
220
229
47
60
109
115
144
151
202
205
222
229
253
267
268
282
88
94
90
104
234
245
258 SOCIO-ECONOMIC ANALYSIS
SOFT SYSTEMS
239
SOFT SYSTEMS ANALYTIC FRAMEWORK STAKEHOLDER
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
STAKEHOLDER INPUTS
206
209
239
254
42
181
233
249
264
275
278
283 STRATEGIC PROJECT ON REACH TESTING
99
SUBSTANCE-BYSUBSTANCE (CASE-BYCASE) EU APPROACH
14
SUBSTANCE-BYSUBSTANCE APPROACH
SUBSTITUTION PRINCIPLE
109
SWEDISH CHEMICAL INDUSTRY
141
SWEDISH CHEMICAL LAW
137
SWEDISH CHEMICAL RISK MANAGEMENT
113
SWEDISH CHEMICALS POLICY SWEDISH POLICY
113 149
SWEDISH RISK MANAGEMENT
127
SWEDISH SOCIETY FOR NATURE CONSERVATION’S ECO-LABELLING
240
This page has been reformatted by Knovel to provide easier navigation.
Index Terms SYNTHETIC CHEMICALS
SYSTEMS FRAMEWORK
Links 22
40
55
142
211
278
1
97
101
155
176
178
180
183
192
197
205
214
216
229
235
237
242
247
250
253
262
264
268
270
274
284
292
259
261
175
177
192
197
200
202
206
233
240
243
250
253
260
262
276
282
248
255
261
DECISION-MAKING MATRIX
260
PRIORITISATION METHOD RECOMMENDATIONS
243 250
REGULATORY OUTCOMES
247
REGULATORY RECOMMENDATIONS
265
T TARGET ORGAN SYSTEMIC TOXICITY TARGET SETTING
TRICHLOROBENZENE
29
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
TECHNICAL APPROACHES
264
TECHNICAL GUIDANCE FOR THE DEVELOPMENT OF RISK REDUCTION STRATEGIES TESTING FRAMEWORK
159 104
TECHNICAL GUIDANCE DOCUMENT
97
101
115
159
171
177
183
195
235
239
270
287
233
262
147
268
TESTING ACUTE
31
TESTING CHRONIC
31
TESTING SUB-CHRONIC
31
TICK BOX APPROACH
184
TOLERABLE
178
192
268
282
TONNAGE SYSTEM TOP-DOWN
43 126
128
283
288
LEGISLATION
169
REGULATION
148
TOXICITY CHRONIC
258
TOXICITY TESTING
40
TOXICITY ACUTE
29
TOXICOLOGICAL TESTING 42 45 59 This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
TREND ANALYSES
115
TRICHLOROETHYLENE
165
U UK CHEMICALS POLICY
129
UK RISK-BENEFIT APPROACH
114
163
172
174
234
235
UK ROYAL COMMISSION ON ENVIRONMENTAL POLLUTION
SCREENING PROCESS
173
UNCERTAINTY FACTORS
34
210
211
UNEP STRATEGIC APPROACH TO INTERNATIONAL CHEMICALS MANAGEMENT
287
V VERY PERSISTANT AND VERY BIOACCUMULATIVE SUBSTANCES
23
182
186
37
71
73
240
243
249
259
263
This page has been reformatted by Knovel to provide easier navigation.
Index Terms
Links
Z ZERO-RISK APPROACH
285
This page has been reformatted by Knovel to provide easier navigation.