2. Background, theory and method

In 2002, the EU Commission launched a Multi-Stakeholder Forum (MSF) on CSR with an aim of formulating a European guiding framework for CSR. At the outset, the Commission suggested strongly that the European CSR framework which was to be developed by the MSF be based on internationally agreed principles on human and labour rights. The final product, however, was much less strong in references to international human rights law than what could have been expected based on indications from the Commission and other stakeholders and the strong role which human rights law and practice generally hold in Europe. The MSF was technically headed by the Commission but industry, employers, employees and NGOs were main participants.

In 2006, the Commission launched a new CSR initiative, the CSR Alliance. The CSR Alliance was announced to be a learning forum comprising the Commission and European companies with the aim of exchanging best practice. At the launch of the Alliance, the Commission again suggested international human rights law as a strong normative source for CSR.

The launch of the MSF was based on a Commission Green Paper on a European Framework for CSR (Commission, 2001a), and a Commission Communication (Commission, 2002). The Green Paper and the Communication were formulated, at least in part, against the background of a 1999 Resolution from the European Parliament (European Parliament, 1999), which had called for Codes of Conduct for European Multinational Corporations, as well as the European Strategy on Sustainable Development (‘the Gothenburg strategy’) (Commission, 2001b). The 1999 Resolution mentioned core labour standards, and human and children's rights as topics in need of attention through the regulation of European companies when acting outside the EU, particularly in developing countries. The CSR Alliance was announced in a 2006 Communication (Commission, 2006a), the Commission's first Communication on CSR after the final report of the MSF in 2004. None of those instruments – the Green Paper, the Communications or the European Parliament Resolution – are hard law. However, they have a certain soft law effect pertaining to emerging EU regulation of CSR.Footnote ³

Human rights are often considered part of CSR.Footnote ⁴ However, the insistence by the EU Commission and many others that CSR is ‘voluntary’, without nevertheless clearly stating what precisely this entails with regard to law and legal requirements, tends to shroud the contributions that international human rights law may make to CSR normativity.

For the purposes of this case-study based analysis, international human rights law is the overall normative point of reference for the purpose of analysing the EU Commission's approach to making companies take sources of international human rights law (including international labour law) into account as a normative foundation for their self-regulation on CSR, and to draw lessons on how this may be done effectively through soft public–private regulatory schemes. The analysis particularly considers the normative role of international hard and soft law instruments on human rights with regard to the EU MSF and CSR Alliance. It also discusses the interrelationship between objectives of a political nature, which are addressed directly through the EU initiatives, and specific human rights standards. These are argued to be complementary for the purpose of Commission CSR objectives as well as elements of human rights implementation in EU member states and beyond.

With the exception of law-based requirements on CSR reporting which are found in some European and other states, CSR regulation is generally not subject to regulation in conventional legislative texts. For this reason, ordinary preparatory works for legislation are also scarce. For a study of the development of CSR normativity in a public–private regulatory context, statutory texts and case-law are so scarce that other forms and sources of data materials must be identified to provide an empirical basis for analysis. This is particularly so in an EU context because, so far, the EU has shunned the development of conventional EU law on CSR.Footnote ⁵ The study on which this article builds therefore takes inspiration from discourse theory and method as a supplementary method for data collection and analysis. At an overall level, discourse theory and discourse analysis are concerned with the analysis of power relations and the role (and use) of discourse to shape society, including politics and ideas on what is appropriate conduct in a social setting. Discourse theory ranges from very abstract theories of power to very concrete methods for textual analysis, often applied in some type of combination. Collection of the empirical data (texts) which form the body of the two cases on which this article builds has been based on a discourse theory inspired textual reading of official EU documents related to the MSF and the establishment of the CSR Alliance, as well as speeches and reports produced as part of these initiatives. The analysis is also discourse analysis inspired in the sense that power relations are a main point of focus and serve as an explanation for the outcomes of the MSF in relation to the construction of the concept of CSR. This analysis focuses on the role to be played by international human rights law for a European framework on CSR as a constitutive element of the CSR construct, and on the construction of CSR as ‘voluntary’ or ‘mandatory’. It should be noted that the study is not, however, a fully-fledged discourse analysis.

The application of discourse analysis to legal studies is a relatively novel but not untested method in legal scholarship (see for example Holdgaard, Reference Holdgaard2008; Kjær, Reference Kjær, Bentzon, Bjørst and Koch2002; Koskenniemi, Reference Koskenniemi1989; Kennedy, Reference Kennedy1987; and compare Linderfalk, Reference Linderfalk2009; Perez, Reference Perez2004). A discourse analytical approach to studies of CSR in a legal scholarship based context has also been undertaken in a couple of other instances, notably a legal–anthropological study of language at CSR meetings and CSR reports by Conley and Williams (Reference Conley and Williams2005). The relevance of the approach is also underscored in a recent study which recognises the speech act quality of corporate codes of conduct (Herberg, Reference Herberg, Dilling, Herberg and Winter2008, p. 24). With a related but different approach from the one undertaken here, another recent study by Koudiadaki on reflexive law and the implementation of EC Directive rules on employee consultation in Britain applies textual analysis to provide contextual understanding (Koudiadaki, Reference Koudiadaki2009).

In the current article, the textual data serves as empirical data much the same as preparatory texts for statutory or other regulation in many other legal studies. The application of discourse theory and method on which the current article draws has served as a method for reading statements produced and presented during the MSF process. The method of textual reading is inspired by Norman Fairclough's critical discourse theory and textual focused method (see for the general approach which has inspired this study, inter alia Fairclough Reference Fairclough2003; Reference Fairclough, Wodak and Meyer2001; Reference Fairclough1995; Reference Fairclough1992). In the current context, however, it does not focus on specific linguistic features but on those linguistic parts of the statements which specifically seek to shape the construction of the CSR concept in the European context, with particular focus on the role to be played by international human rights law as part of the normative construction of CSR and CSR as ‘voluntary’ or ‘mandatory’.

Fairclough's theory is normally applied in a political science context with a view to analysing power relations and how language creates or shapes social events and features. In the context of the analysis informing this article, this has been complemented by the approach of Ernesto Laclau and Chantal Mouffe, who argue that discourses are modalities for acting out ‘struggles’ to achieve hegemony in terms of deciding the substantive contents of terms, etc. (see for example Andersen, Reference Andersen2003; Collin, Reference Collin2003, esp. pp. 98–110). In the context of constructing CSR within the EU MSF, CSR as a concept is a ‘floating signifier’. It is a value-laden concept which, however, lacks a precise definition. Various stakeholders, some of which have been included in the EU MSF, struggle to gain impact to fix the normative substance to be connected to CSR in the EU context. In this article, discourse analysis is employed in a legal scholarship context which sets it apart from the political science context but still contributes to understanding the Commission's and MSF participants' struggles to determine what is to be understood normatively by CSR in the EU and for EU based companies. Discourse theory and analysis provides a way to identify and read texts to understand how their positions and arguments impact on the end normative result, that is, the way in which CSR as a normative concept is constructed through the EU MSF process. In the context of the theme of the study, the discourse theory inspired approach also adds theoretical and informing perspective to the analysis of the argumentative struggle between stakeholders. That is of particular relevance in view of the attention paid by reflexive law theory to balancing power relations (see below in this section and the section on reflexive law and below in Sections 6–7).

In terms of specific method, empirical data has been gathered through initial identification of texts produced as part of the MSF and the CSR Alliance set-up and processes. The texts have been identified through the website which the EU Commission has established on CSR.Footnote ⁶ The texts have been subjected to a discourse analysis inspired textual reading in which the statements have been studied, particularly with regard to their arguments in favour or against international law as a normative source for a European guiding framework on CSR, general use of reference to international human rights law, and the form of regulation which CSR in Europe should be subjected to. The analysis here is based on a larger study.Footnote ⁷ Space restraints do not allow the rendering of all relevant quotes in this article. Instead, selected quotes and summaries of longer text sections substantiate the analysis of the case-studies.

Following the text-based description and preliminary analysis of the MSF and the CSR Alliance in Sections 5.2 and 5.3, the EU Commission's approach to regulating CSR is analysed in Section 6 from the perspective of reflexive law. Reflexive law is a both a theory and a regulatory strategy, based in legal/socio-legal theory. Reflexive law is a process-oriented legal theory. The terminology of reflexive law refers to learning and exchange between social subsystems of their expectations and demands of each other as well as best practice, and other modalities to support reflection within and between social subsystems. Reflexive law assumes reflection within public authorities on their own regulatory capacities, followed by their setting up procedural institutions to promote reflection among the corporate sector and/or others, to internalise (and further reflect) societal concerns and demands through self-regulation. The theory of reflexive law, which is also the foundation for the regulatory strategy, was developed by German legal theorist Gunther Teubner in the 1980s and has been debated and applied analytically in later writings by Teubner as well as by other authors (see further Section 6). It was originally developed as a response to what was seen as a regulatory crisis of the interventionist state (typically the welfare state) to meet or handle societal challenges and needs through ordinary top-down substantive law. It targeted regulatory challenges and the effectiveness of law within nation states, providing an option to supplement formal substantive legislation with a regulatory technique which allowed for greater participation of those whose actions were necessary in order to handle the societal challenges and demands. Such participation could be achieved through procedural co-regulatory modalities which could be established by authorities. The procedural framework was seen to provide a forum for dialogue and mutual learning on expectations and experience between different social subsystems. Through this, it was expected to support mutual reflection between those subsystems in relation to regulation and self-regulation on the topics concerned, and their ability to internalise societal concerns, needs and expectations.

Applying reflexive law in the context of CSR is not of itself new. Teubner has addressed CSR from the perspective of reflexive law, and others have followed (see also Section 6 below). What is new in this article is the application of reflexive law to understanding the regulatory processes undertaken outside the context of ordinary law-making in an EU setting, and to draw conclusions on how such processes may be rendered more effective and legitimate.

The theory of reflexive law as applied in this article is based on the early writings on reflexive law, in which the theory was proposed as a regulatory technique to supplement formal and substantive law to meet challenges related to environmental degradation, inequality and unemployment in (European) welfare states. While not disregarding later contributions or applications of reflexive law, drawing on the origins of the theory has been found to be the most useful approach for the current article and the application of reflexive law in a different regulatory context and at a different systemic level than the national level for which it was originally intended – in casu the efforts of the EU Commission as an authority at inter-governmental level to procedurally and substantively promote corporate self-regulation in order to meet societal challenges. Reflexive law presents a way to understand emerging EU regulation of CSR in terms of regulatory strategy and technique to encourage the private sector to self-regulate, to analyse the results and challenges of the Commission's initiatives, and draw lessons for more effective public regulation of private–sector self-regulation on CSR to meet political and other societal concerns.

Reflexive law theory draws on systems theory, but combines that with other theory-based approaches to law and regulation, such as Habermasian theory and Nonet and Selznick's responsive law. Systems theory focuses on systems and does not normally discuss actors. For the purposes of the current article it should be noted that Teubner's reflexive law theory recognises interaction between representatives of economic, political and legal systems. Referring to these representatives as ‘actors’ in the current context accords with international law's recognition of ‘state actors’ and ‘non-state actors’ and its compartmentalisation of different rights, duties and access to law-making, depending on the ‘state’ or ‘non-state’ status. Reference to actors also accords with discourse analytical approaches (compare Hajer, Reference Hajer1995).

With developments in societal, political and legal concerns on human rights, labour conditions (especially in developing and transitional countries), environmental and climate challenges and related regulatory efforts at inter-governmental levels to target the conduct of companies, the need for understanding soft and public–private regulatory modalities of a reflexive type has been pushed from the national level onto the inter-governmental level. This is particularly so with regard to efforts by inter-governmental organisations, e.g. within the EU or the UN. This observation underscores the need for increased understanding of and insight into strategies through which public institutions may promote corporate self-regulation on CSR and related issues at inter-governmental level. Such a need is further underscored by public and political expectations that various inter-governmental bodies act to influence corporate behaviour on human rights. The need for insight is strengthened because of limited direct or formal legislative powers and state backing in many cases, due to the general structure of international law as well as, in the case of the EU, limited independent legislative competences of the EU with regard to human rights and labour standards rights, especially as they pertain to business action. The latter case for insight into such regulatory modalities combines with, and is further stressed by, the tendency over the past decade for inter-governmental bodies to launch relatively informal public–private regulatory initiatives precisely to address societal concerns that cannot easily be met through ordinary legislative measures (see Buhmann, Reference Buhmann2009; and compare Krisch and Kingsbury, Reference Krisch and Kingsbury2006; Kingsbury, Krisch and Steward, Reference Kingsbury, Krisch and Steward2005).

While the institutional context for the use of reflexive regulatory strategies for inter-governmental CSR regulation, such as that which frames the EU Commission's efforts, differs from the national law context which led Teubner to propose the theory, the concerns are the same: societal needs (including needs based on similar concerns: inequality, poverty, unemployment, environmental degradation) which translate into policy objectives and require being addressed through effective regulatory action, combined with the inability or inefficiency of formal legislation or law-making to do so, and the potential role of the private sector to contribute to solving the problems. For this reason, reflexive law has been found for the purposes of the current analysis to provide a useful theoretical framework for the analysis and explanation of the functioning and results of the MSF and the launch of the CSR Alliance as shown in Sections 6–7.

Given the relative immaturity of legal-theory-based CSR analysis of contemporary and emerging CSR schemes, and particularly the scarcity of empirical research on the process of EU Commission initiatives to regulate CSR, some delimitations are in place: the aim of the current article is the relatively modest one of describing and analysing the EU MSF and the launch of the CSR Alliance in terms of processes that weave together international law, particularly on human rights, with policy goals, and analysing the implications of this for regulatory strategy and procedure. The main focus of the discourse theory inspired analysis (Section 7) is on negotiations and other linguistic exchanges that took place as part of the MSF and the Commission's launch of the Alliance, and the role played by international human rights law as a normative source for CSR in the EU context throughout these processes. The concluding Section 8 draws on this to formulate recommendations on reflexive law inspired public–private regulation on CSR.

3. The concept of CSR and its relation to law

CSR is a dynamic and open term. Although neither scholars nor practitioners have one single definition of CSR, there is general consensus that CSR means that companies take responsibility for their social and environmental impact (Crane, Matten and Spence, Reference Crane, Matten and Spence2008, pp. 4–7; and compare on the debate on the concept Carroll, Reference Carroll1979; Reference Carroll1991; Schwartz and Carroll, Reference Schwartz and Carroll2003; Newell and Frynas, Reference Newell and Frynas2007; Hopkins, Reference Hopkins2006; Wan, Reference Wan2006). In practice, this means that companies take action to mitigate or prevent negative social and environmental impact or to maximise positive impact, generally beyond the obligations of directly applicable statutory law. With variations, this is reflected in definitions of CSR employed by various business organisations, NGOs and (inter-)governmental organisations.

Corporate responsibilities for human rights and labour rights are often embodied in CSR as ‘social issues’. This had already been reflected in the so-called ‘Triple Bottom Line: People, Planet, Profit’ (Elkington, Reference Elkington1998), which set in motion the process of CSR-reporting in many companies in Europe and beyond. The human and labour rights informed understanding of social issues as part of CSR remains reflected in the widely used CSR reporting tool Global Reporting Initiative (GRI) and its third-generation G3 reporting guidelines launched in 2006,Footnote ⁸ and in the CSR policies and strategies of a number of companies. The understanding of CSR, however, differs somewhat between regions. Whereas corporate philanthropy is a relatively prominent part of the conception of CSR in the United States and in some parts of Asia, in Europe CSR is generally perceived as corporate practice directly related to the employees of the company, and its community, suppliers or customers.

CSR is often, although in practice not exclusively, understood to be ‘voluntary’. Just as there is no commonly agreed precise definition of CSR, there is also no specific understanding of what exactly ‘voluntary’ means with regard to CSR. Various definitions and discussions indicate a narrow understanding of ‘voluntary’. This comes out, for example, in the EU's definition (below). Set in contrast to ‘mandatory’ action, in practice this idea of CSR is understood to mean that CSR is to act beyond immediate compliance with national public law applying to the specific business entity. Indeed, as further discussed below, the alleged quality of CSR to be ‘beyond law’ creates some confusion and is not by all means conducive to the development and implementation of CSR. The dichotomy appears to assume that ‘mandatory’ is distinguished from ‘voluntary’ by the presence or absence of public law and coercion related to the non-observance of such law. It therefore also assumes a highly positivistic approach to law, according to which legal obligation implies effective coercive sanctions. As demonstrated below, the understanding is not in fact in accordance with significant parts of management theory, which assumes that CSR entails at least abiding by the letter and perhaps even the spirit of the law, nor with business and emerging legislative practice in many states.

Instruments and standards of public international law serve as normative guidance for much CSR action, and also inform the social expectations of business. The assertion that CSR is ‘voluntary’ therefore also assumes that the pertinent international law instruments and standards are perceived – and the international legal order construed – to not establish obligations for companies. For companies that prefer minimal legal constraint and who build on CSR as ‘voluntary’ as a marketing parameter (‘do-gooding’), a construction of international law as not creating obligations for companies is therefore paramount. As we shall see, that approach had a strong impact in the context of the EU emerging regulation of CSR in the 2002–2004 MSF on CSR.

Overall, the dichotomy between mandatory and voluntary action, understood as compliance with law respectively acting beyond the requirements of law, has played a relatively prominent role in later years' efforts to define corporate responsibilities for social and human rights. This was evident in debates within the EU MSF as well as on corporate responsibilities for human rights in the UN Human Rights Commission 2004–2005 and (since 2006) the UN Human Rights Council. The dichotomy is reflected in formal presentations of the EU's CSR Alliance as ‘not a regulatory instrument’, and a similar representation of the UN Global Compact. The dichotomy has been fuelled by certain business organisations and NGOs, but is perceived by many others as unfruitful and as having slowed down progress in developing normative consensus and guidelines (see for example Chandler, Reference Chandler2008).

Indeed, an understanding of CSR which is well entrenched in organisational scholarship assumes that compliance with law is part of CSR. In an article which is often cited in management-oriented CSR studies and applied at business schools, Archie B. Carroll in 1979 provided a definition according to which “[t]he social responsibility of business encompasses the economic, legal, ethical, and discretionary expectations that society has of organizations at a given point in time” (Carroll, Reference Carroll1979, p. 500). In Carroll's understanding of CSR, those four categories are not mutually exclusive, nor are they intended to portray a continuum with economic concerns on one end and social concerns on the other. Legal responsibilities are understood as business's fulfilment of its economic mission within the framework of legal requirements. They are “the ground rules – the laws and regulations – under which business is expected to operate” (p. 500).

Carroll's 1979 definition assumes that the social responsibilities of business include compliance with law in the narrow sense of directly applicable legal obligations (what we may refer to in this context as lex), but does not seem to recognise a wider role for law in CSR, such as for legal instruments that do not apply directly to the pertinent business to inform CSR normatively. Moreover, it does not appear to recognise law as a regulatory strategy to promote certain political or societal objectives that may be met through corporate CSR action (both of the latter may be referred to in this context as ius). A later model by the same author, the ‘CSR Pyramid’, also includes legal responsibility, understood as obeying the law (Carroll, Reference Carroll1991). In a 2003 article with Mark S. Schwartz, Carroll introduced a three-domain approach, developed partly in response to weaknesses in the 1979 definition and the 1991 model. One of the major points of the 2003 article is an elaboration of the ‘legal domain’ that goes beyond simple legal compliance and also includes action undertaken in order to avoid civil litigation and in anticipation of (new) law. In addition, compliance is elaborated to comprise three forms: ‘passive’ (the idea that a company does what it wants and just happens to be acting within the boundaries of the law, i.e. no specific intention to comply with the law); ‘restrictive’ (the idea that the legal system limits, constrains or modifies otherwise intended behaviour in a restrictive fashion to promote conduct within the boundaries of legal compliance); and ‘opportunistic’ (when a company takes advantage of loopholes in legislation, in many cases complying with the letter but not the spirit of the law, or chooses to place operations in a particular jurisdiction because its weaker legal standards are found to be favourable to the company's objectives) (Schwartz and Carroll, Reference Schwartz and Carroll2003). The last category of opportunistic compliance is of particular relevance in an (international) human rights and labour law context. Research suggests that the category of anticipation, i.e. when a company acts in a particular way to pre-empt new or amended law and therefore to keep the idea of voluntary action, is also of high practical relevance in a number of contexts (for example, Hess and Warren, Reference Hess and Warren2008, with references; Adams, Reference Adams2002).

Although corporate social responsibilities were debated from as early as the 1930s in US legal scholarship (Berle, Reference Berle1931; Dodd, Reference Dodd1932), with a European attempt at revival in the 1980s by Gunther Teubner in a private law Corporate Governance context (Teubner, Reference Teubner, Hopt and Teubner1985), CSR tended for the majority of the twentieth century to be considered a topic outside the realm of relevance to the legal field. One reason for this is likely to be the allegation that CSR is beyond the requirements of law, although this to some extent disaccords with Carroll's understanding. Taking on the line of thinking strongly promoted by American economist Milton Friedman (Reference Friedman1970), popularly rendered as ‘the business of business is business’, lawyers tended to debate whether corporate social responsibility was at all within the mandate of a company's management and its fiduciary duties, rather than what actions a socially responsible company could or should be engaging in. The change in interest for CSR which the legal field has seen since the mid 1990s is due to a number of trends and possible reasons. Among these, cases lodged in national courts over the past decade (in the United States under the Alien Torts Claims Statute and in the United Kingdom and some other countries under their national statutes) have demonstrated that companies' impact on labour rights, human rights and the environment is of legal relevance with regard to torts, compensation and legal practice. Codes of Conduct are increasingly incorporated into contracts between supplier and buyer companies, with CSR commitments in such Codes becoming legally binding on the supplier in the intra partes relationship. Work undertaken in the context of various inter-governmental, national and NGO fora to clarify the extent of corporate social and human rights responsibilities and encourage corporate self-regulation also touch on normative and regulatory issues of relevance to lawyers and draw on sources of law, particularly international law. Grievance mechanism schemes, like the National Contact Points (NCPs) established under the OECD Guidelines for Multinational Enterprises, draw on legal method for investigation and complaints handling.Footnote ⁹ Some nation states, including the United Kingdom, France, Sweden, Australia and, most recently, Denmark,Footnote ¹⁰ have promulgated statutes with legally binding requirements on certain companies and/or institutional investors to report on the social impact of their business and investments.

At the same time, international society has moved towards regulating CSR-related topics through conventional instruments of international law: at inter-governmental level, the first major attempt by the UN to formulate a code of conduct for transnational corporations was initiated in 1977. It was finalised in 1990 (UN Draft Code of Conduct on Transnational Corporations (TNCs), United Nations, 1990) but abandoned within two years due to opposition from certain governments and companies, partly as a result of the North–South conflict and the Cold War. The draft code incorporated obligations for TNCs to, inter alia, respect host countries' development goals, observe their domestic law, respect fundamental human rights, and observe consumer and environmental protection objectives. The OECD's Guidelines for Multinational Enterprises were adopted in 1976, followed by the ILO [International Labour Organisation] Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy in 1977 (both revised in 2000). A new effort at UN level to regulate the behaviour of TNCs with regard to CSR-relevant issues was made with the draft Norms on Human Rights Responsibilities of the Transnational Corporations and other Business Enterprises (‘the draft UN Norms’) prepared between 1998 and 2003 by a working group set up by the then UN Human Rights Commission (United Nations, 2003). Following this, the UN Secretary General appointed a Special Representative on business and human rights in 2005Footnote ¹¹ (SRSG John Ruggie), with the original mandate of 2005–2007 extended in 2007 by the UN Human Rights Council. Following the Human Rights Council's unanimous ‘welcome’ in June 2008 of the final reportFootnote ¹² of the Special Representative proposing a ‘Respect, Protect and Remedy’ policy framework (Ruggie, Reference Ruggie2008), the mandate was extended by the Human Rights Council for another three-year term, 2008–2011, with the objective of operationalising the 2008 framework (Human Rights Council, 2008). Although the work of the Special Representative has made it clear that there are distinctions between corporate responsibilities for human rights and CSR in general, and particularly with regard to state duties, there are important overlaps in terms of societal moves pushing for corporate responsibility and resulting types of regulatory efforts at various levels.

The UN Global Compact was established in 2000 as an initiative of the then UN Secretary General, Kofi Annan.Footnote ¹³ The Global Compact is also strongly supported by the current UN Secretary General, Ban Ki-moon. The Compact is run by a small secretariat in New York and funded primarily through contributions from donor states. Participation is primarily for companies but has recently also been opened to public institutions and cities, and has been growing steadily, especially in recent years. As of early 2010, company participants number around 6,000. Participants commit to the ten principles on human rights, labour standards, environmental protection and anti-corruption. Among the Compact's ten principles, Principles 1–2 and 3–6 refer directly to international law instruments on human and labour rights, in particular the Universal Declaration of Human Rights and the 1998 ILO Declaration on Fundamental Principles and Rights at Work, which refers to the core labour principles of freedom of association and collective bargaining, non-discrimination in employment, elimination of forced labour and elimination of child labour.

Also testifying to the increasing role of international law in informing CSR, the 2007 G8 Summit in Germany agreed to promote corporate and other forms of social responsibility (Chair's Summary, 2007;Footnote ¹⁴ G8, 2007), specifically referencing the OECD Guidelines for Multinational Enterprises, the ILO Tripartite Declaration of Principles concerning Multinational Enterprises and Social Policy, and the UN Global Compact as normative bases.

The above developments, which have taken place over the past decades, demonstrate that CSR is becoming a field of interaction between society and law. This has been particularly sparked by the increased focus on the human rights responsibilities of business. The Special Representative of the UN Secretary General on Human Rights and Business has considered the interrelationship between social expectations and new or emerging legal regulation and has highlighted the significance of societal processes which lead to a gradual hardening or ‘juridification’ of societal CSR expectations to become legally binding requirements on companies and states, and related to companies' ‘social licence to operate’ (Ruggie, Reference Ruggie2007, inter alia paras. 44, 62, 72, 80; Ruggie, Reference Ruggie2008, para. 54).

As the overview above shows, law and CSR are related in a number of ways. On several grounds, the alleged ‘voluntary’–‘mandatory’ dichotomy surrounding much of the CSR debate may be characterised as false and as alleging a distinction between legal compliance and voluntary action that does no justice to the law-based normativity which today permeates CSR in a number of ways, as indicated above. First, CSR has been defined by organisational scholarship as also encompassing compliance with law. Second, in order to go beyond the law, one has at least to know what is required by law. Third, more and more states, especially in Europe but also in industrialised as well as developing countries outside Europe, are imposing CSR reporting requirements on certain companies and/or investors with the objective of affecting their actions prior to reporting. Fourth, in developing countries, CSR is increasingly coming to be understood as compliance, with a new term, ‘CSR + ’, coined to suggest voluntary action beyond what is mandated by law.Footnote ¹⁵ In fact, as described elsewhere, for some time some companies and CSR consultants acting in countries with weak enforcement of CSR-relevant legislation, e.g. on labour rights, have considered simply acting in accordance with the law to be CSR-compliant (Buhmann, Reference Buhmann2005). In these contexts, CSR includes observing legal requirements that are not likely to be enforced and ensuring that the company and its suppliers do not abuse their power, personal contacts or funds, including through bribes, to avoid enforcement. Last in this context, but not least, links between CSR and law are also indicated by several studies published in recent years which place CSR in a legal context. Such texts have demonstrated the relevance of CSR to various fields of national and international law, such as Corporate Governance and company law, economic law, contract law, torts law, international trade law, international labour law, international human rights law, arbitrated settlements of transnational disputes and combined approaches of soft and hard regulation to dealing with the regulatory challenges of a globalised society (see for example Addo, Reference Addo1999; Jägers, Reference Jägers2002; Clapham, Reference Clapham2006; Mares, Reference Mares2006; Zerk, Reference Zerk2006; Marrella, Reference Marrella, Benedek, de Feyter and Marrella2007; Muchlinski, Reference Muchlinski2007; MacLeod, Reference MacLeod2007; McBarnet, Reference McBarnet, McBarnet, Voiculescu and Campbell2007).

To sum up, the normative links between CSR and law are numerous and of a character that negate an inflexible understanding of CSR as only requirements beyond law. At least from a legal scholar's perspective, there is a need for an understanding of CSR that allows for a degree of realignment between the management-oriented approach to CSR as being anything but applicable, enforced positive law (lex) and the much wider understanding of law as societal regulatory response and normativity (ius) found among branches of legal theory, especially but not exclusively the sociology of law. Accordingly, CSR is understood in this article as action by private or public institutions to contribute to the sustainable use or development of human or environmental resources and which is not required by binding (‘hard’) law directly applicable to that entity or enforced as a result of such law. This leaves open the possibility that CSR practices may be guided by soft regulation or by hard law not directly applicable to the acting entity (e.g. international treaty law in relation to private non-state actors). The case for this understanding is further elaborated in the context of the following section, which specifically considers the links between human rights and CSR normativity in an international and EU law, as well as EU Member States, context.

4. CSR normativity and human rights: linking objectives and action through sources of law

International human rights law presents a particular potential for the inter-governmental regulation of social responsibility. The Special Representative on business and human rights, SRSG John Ruggie, has done much to connect international and national law, especially through emphasis on state obligations under international law to protect against human rights violations by individuals (including legal persons) against other individuals, and international law as a normative source for company due diligence and the corporate responsibility to respect human rights (Ruggie, Reference Ruggie2008; Reference Ruggie2010). Recent developments, such as the BP oil spill in the Gulf of Mexico in 2010 and increased desertification, for example in China, have drawn attention to the linkages between environmental degradation and climate change and human rights, such as access to food and the health impacts of fighting environmental damage. With the potential to expand the European and Northern hemisphere attention to human rights beyond civil and political rights, which in many ways tend to be the primary area of human rights focus in those parts of the world, such developments have also contributed to strengthening insight into the links between social and economic human rights, and environment and climate. Recent years' climate change concerns have also spurred some interest in exploring how a human rights approach can feed into addressing environmental and climate concerns.Footnote ¹⁶

The relevance of addressing the implications of CSR on normative conceptions and forms of regulation is underscored by the fact that legal regimes must and do adapt to normative changes in society. A dynamic and adaptive character of law has been emphasised as a feature of international human rights law and has also evidenced itself clearly in the course of ECJ case-law on fundamental rights. As CSR is increasingly addressed not only by companies but also by (inter-) governmental agencies, regulatory initiatives may gradually take on new forms to allow them to be contained within the constraints of traditional views on duty-bearing legal personality and formal regulatory powers at inter-governmental level.

CSR normativity reflects an internationalisation of societal norms and law on human and labour rights and the environment, informing the demands on business from both without and within the business itself. As indicated, instruments of international human rights (used here in a broad sense encompassing also international labour rights/law) are often relied on by NGOs, companies, auditors and (inter-) governmental bodies to explain the substantive contents of social issues of CSR. In terms of substantive contents, CSR is often related to societal expectations of good corporate behaviour, i.e. expectations of a certain level of consideration of environmental sustainability and of morals in observing minimum standards in relation to labour or human rights, even if these minimum standards do not apply to the company in a legally binding way. For example, many consumers and investors expect a responsible company not to pollute or deplete drinking water sources for the local community, not to employ forced labour and not to exploit the labour force in poor countries by offering substandard working conditions. Such societal expectations become normative as they drive and motivate corporate practice. Their role as possible precursors to legal regulation have been acknowledged by the Special Representative of the UN Secretary General on Human Rights and Business in his February 2007 report to the UN Human Rights Council (Ruggie, Reference Ruggie2007, para. 62).

Consumers' and other stakeholders' expectations of the moral conduct of companies are often shaped by sources of international law, especially international labour and human rights law, e.g. relating to child labour/minimum age for employment, working hours and overtime pay. This applies even though these sources of law do not apply directly to companies, but to states. International treaties on human rights are binding (hard law) on states that have ratified them. Normative directions contained in CSR are increasingly seen by companies and some types of stakeholders as binding, even though they are not formally so. Corporate CSR action, for example value statements, codes of conduct and non-financial reports, increasingly refer to international law on human and labour rights and the environment, although private companies are not duty-bearers under international law. Evidenced inter alia by the human and labour rights principles of the UN Global Compact (especially Principles 1–6), the GRI reporting guidelines and the process standard SA8000,Footnote ¹⁷ the social or ‘people’ dimension increasingly appears to be perceived or explicitly defined as containing aspects of human and labour rights, including rights to non-discrimination and equal opportunities, education, vocational training, work and decent working conditions. The UN Global Compact and the draft UN Norms both specifically define corporate responsibilities on the basis of international law. Through such references, international law comes to inform CSR in terms of its normative substance in ways comparable to the role that sources of law play in conventional legal practice and analysis.

The preamble to the Universal Declaration on Human Rights, in which the UN General Assembly proclaimed the Declaration “a common standard of achievement for all peoples and all nations, to the end that every individual and every organ of society […] shall strive […] to promote respect for these rights and freedoms”, has come to serve for many companies as a basis for a commitment to human rights. The revisions in 2000 of the 1976 OECD Guidelines for Multinational Enterprises, which were originally formulated in the context of transnational finance, were made to include labour standards and human rights and to extend the applicability of the Guidelines to worldwide activities of OECD-based multinational enterprises, and a recommendation to encourage compliance with the Guidelines through the supply chain. Also carried out in 2000, revisions to the 1977 ILO Tripartite Declaration were made to incorporate fundamental principles and rights at work. The Universal Declaration on Human Rights, the OECD Guidelines, and the ILO Tripartite Declaration are among soft law instruments relating to human rights that are often referenced in corporate CSR policies, strategies, Codes of Conduct and CSR reporting.

In developing countries, CSR action based on international law as a source may contribute to human rights implementation. The EU and several Member States have policies of promoting human and labour rights in Third World countries. Traditionally, development co-operation and bilateral policy dialogue have been the means for the implementation of foreign policy goals. If EU-based companies undertake activities that go beyond the requirements of national law or beyond the extent to which national law is enforced, they may effectively contribute to the implementation of those as foreign policy goals. In an EU context, CSR has recently been seen as a means for contributing to ‘responsible competitiveness’, i.e. competitiveness with respect to human and core labour rights.

Commission statements on CSR already emphasised in the 2001 Green Paper that CSR has a strong human rights dimension (Commission, 2001a, p. 14), and that “this is recognised in international instruments such as the ILO Tripartite Declaration […] and the OECD Guidelines” (p. 14). In its 2002 Communication on CSR, the Commission stated that EU based companies should “demonstrate and publicise their worldwide adherence” to the OECD Guidelines, and that their Codes of Conduct should integrate the ILO fundamental Conventions and draw on these instruments in defining CSR (Commission, 2002, pp. 6–7, 13). Overall, the message was that EU-based companies should observe the OECD Guidelines and the ILO Tripartite Declaration in their actions outside the EU. Thus, at least with regard to activities outside of the EU, the Commission expressed a clear intention that EU-based companies should let CSR be informed by international human rights instruments of a hard as well as soft law character.

The Commission also tied CSR to the EU's ‘Lisbon process’. The Lisbon Strategy was agreed to in March 2000 by EU heads of state and government. The overall objective was to make the EU “the most competitive and dynamic knowledge-based economy in the world, capable of sustainable economic growth with more and better jobs and greater social cohesion”. Setting a time goal of 2010, a number of specific goals were set out, often referred to as the ‘Lisbon goals’, with the process to gradually implement them as the ‘Lisbon process’. The Lisbon Strategy was revised by the European Council of 22–23 March 2005, which decided to place renewed focus on growth, innovation and employment and to encourage the strengthening of social cohesion and the mobilisation of national and community resources in the Strategy's economic, social and environmental dimensions (Lisbon European Council, 2000; Council of the European Union, 2005). Some of the Lisbon goals, such as the goals of equal opportunities, social inclusion, raised employment and employability and lifelong learning, have much in common with legal claims or rights to non-discrimination in the labour market and beyond, the right to work and to labour market education. Such rights are provided by constitutions or statutory law in some Member States, and their implementation is a legal obligation of Member States under international human rights treaties such as the European Social Charter and the International Covenant on Economic, Social and Cultural Rights, which Member States have ratified independent of their EU membership.

CSR topics related to human rights have entered the EU Lisbon goals indirectly through a focus on economic, social and environmental dimensions, with calls for business to develop and engage in CSR to promote inclusiveness, growth and employment and to cater for European competitiveness. Each of those social issues is directly linked to human rights (especially rights to non-discrimination, and social rights to work, education, an adequate standard of living, social services and the highest possible standard of health), and consequently to a Member State's obligations under international human rights law.

Inclusiveness has obvious parallels with the overarching human rights principle of non-discrimination. The non-discrimination principle, and its detailed corollary that discrimination on the grounds of race, colour, sex, language, religion, political or other opinion, national or social origin, property, birth or other status are assumed prima facie to be arbitrary unless solid reasons are demonstrated for the opposite, run as a general undercurrent of most international human rights law instruments and are embodied in the European Convention on Human Rights, the International Covenant on Economic, Social and Cultural Rights and the European Social Charter. The principle of non-discrimination is also established in the Charter of Fundamental Rights of the European Union (EU Charter). Among discriminatory grounds, the Charter includes national minority, disability and age. Inclusiveness informed by the human rights principle of non-discrimination would therefore mean, in relation to the labour market, a labour market without discrimination on any of the above grounds. The connection to national human rights obligations, however, connects EU policy goals within employment and social policy. From this perspective, the interconnectedness between CSR and human rights could be made pragmatically useful to both Member States and the EU both before and after the entry into force of the Treaty of Lisbon.

CSR is not only about acting responsibly in developing countries or in countries other than one's own. Increasingly, CSR is approached, particularly by public institutions in welfare states, as a response to challenges resulting from aging, migratory patterns and employment needs. Perhaps because the Lisbon process is discussed academically mainly by political scientists and because the ‘Lisbon goals’ are pursued from political or economic perspectives, the Lisbon goals of inclusiveness and employment are generally not addressed in terms of human rights in intra-EU contexts. Nevertheless, there are important points of complementarity between specific human rights and the Lisbon goals, in particular related to non-discrimination, the right to work, certain rights in work within all of the EU, and to addressing labour and human rights concerns related to production in or sourcing from developing or transitional countries. Among the latter, child labour, forced labour, excessive working hours and low standards of working conditions, occupational health and safety and pay insufficient for an adequate standard of living have made the headlines in European news media recurrently in recent years, and increasingly inform the investment decisions of institutional investors as well as the buying patterns of consumers and political pressure exerted by politicians, NGOs and others.

Although social or economic development does not equate to social or economic rights, there is ample possibility that CSR may contribute to improved implementation of relevant international law or related standards not only in states outside the EU, but also within the EU. While the EU has certain powers to regulate companies as well as Member States, the implementation by Member States of their international human rights obligations is not a direct Community concern. However, as all Member States have acceded to key international human rights treaties dealing, inter alia, with aspects related to employment rights and non-discrimination, and consequently have obligations under international law to implement those treaties, Member States' progressive implementation of their commitments under international law is potentially instrumental in achieving the EU objectives of employment and inclusiveness, and vice versa. Given that records of social and non-discrimination rights are among the most problematic in several EU Member States and that – as indicated by Philip Alston and Joseph Weiler (Reference Alston, Weiler and Alston2004) – human rights are increasingly becoming a challenge for the EU as such, the potential synergy would seem to be considerable.

The Commission has specifically tied CSR to intra-EU welfare state challenges through links to the implementation of the Lisbon goals. The EU Commission's 2006 Communication on CSR (Commission, 2006a) links CSR explicitly with the Lisbon goals of promoting employment and inclusiveness within the EU. Although not stated and perhaps not realised in the political science and economics oriented context of the Lisbon process, corporate action for increased inclusiveness in the labour market, or for the provision of vocational training, may result in improved implementation of the human rights of non-discrimination and of the right to work. Inclusive employment practices are therefore instrumental for implementing states' obligations under international or EU law to combat discrimination in employment. Affirmative action oriented recruitment policies and efforts to establish on-the-job training programmes may also contribute to further implementation of human rights related to rights to work and vocational training, as well as related human rights, such as work related rights, education, adequate standard of living, social services and the highest possible standard of health. Similarly, with regard to the ‘responsible competitiveness’ of EU-based companies acting outside the EU, if such companies have recruitment and employment practices which respect international standards on minimum working age, forced or involuntary labour, working hours and leisure, occupational health and safety, pay and related concerns, this would in effect contribute to the improved implementation of these standards in host states.

With the introduction in the EU Treaty of the commitment to defend human rights and basic freedoms (Art. 6) and the powers to take legislative action to combat racial discrimination (Art. 13), the EU has involved itself more directly in anti-discrimination beyond gender discrimination. Still, the national-level implementation of human rights as well as the social objectives of employment and inclusiveness in the labour market do not fall within the ordinary legislative powers of the EU. Seen in this light, CSR and co- and self-regulatory processes involving business and civil society present themselves as alternative options to regulate human rights issues as they effectuate themselves or may be implemented at the level of EU-based companies' interaction with society. Based on these considerations, and particularly the potential that CSR and human rights have for mutually strengthened processes of implementation, the analysis below will consider, inter alia, the way in which human rights have been linked with emerging EU regulation of CSR.

5. The EU multi-stakeholder forum and CSR alliance

6. The relevance of reflexive law for analysing emerging EU regulation of CSR

For the legal scholar, the Commission's insistence on CSR being voluntary, combined with its efforts to promote support for the public–private regulation of CSR and to provide the substantive normativity of human rights, raises a question of how to analyse and categorise the process. On the one hand, the Commission's efforts to promote CSR and the two public–private initiatives of the MSF and the CSR Alliance do not fit ordinary patterns of parliamentary law-making. On the other hand, the normative and regulatory aim is clear, as is the objective of inducing certain behavioural regulation in companies beyond simply setting out public policies whose implementation requires the active action of the private sector, and hoping that companies will follow by themselves. The MSF and the CSR Alliance demonstrate an intention to procedurally initiate and substantively influence the process that will ultimately lead companies to self-regulate. The MSF and the CSR Alliance somewhat resemble the definition of co-regulation provided by the EU Commission in its 2001 White Paper on Governance as an alternative to legislation and non-legislation, with the reservation that the possibilities for EU authorities to intervene “by establishing the specific rules needed” is limited (Commission 2001c, p. 21, compare p. 33).

From the traditional perspective of the legal scholar and legal regulation, the MSF and the CSR Alliance are not law, neither are their objectives to create law in the ordinary positivist sense. Nevertheless, the above overview and preliminary analysis of the case-studies indicate that the two initiatives are not irrelevant from a legal perspective, particularly one concerned with the process of making law including soft law, and one concerned with the interrelationship between international human rights law and normativity and human rights implementation in other legal systems. In this case, this interrelationship plays out in relation to at least two other legal systems: the system of EU law (which in the case of human rights necessarily involves the interaction between the system of EU law and national legal systems), and the system of private law as it relates to companies. In addition, we are able to draw some preliminary findings on the MSF and the CSR Alliance: both initiatives are primarily procedural, but with considerable substantive intentions to shape normativity and the self-regulation of companies on CSR. The MSF 2002–2004 was set up as a co-regulatory initiative, prepared within the framework of the 2002 Communication serving as a preparatory instrument. In the absence of clear and forward-moving results of the kind envisaged by the 2002 Communication, the Alliance was established and presented as an instrument for self-regulation. However, given the amount of directions contained in the 2006 Communication, the Alliance in combination with the MSF 2006 makes a guiding claim, giving effect to Community policy objectives which, in combination with the outcome of the MSF 2002–2004 and its reference to international law, may be intended to have a considerable normatively guiding effect on the outcome of the Alliance.

This combination of procedural fora established by an authority with limited regulatory capacity in the pertinent field (in casu the Commission and the powers of the Commission and the EU as such to regulate the human rights behaviour of EU-based companies and their limited powers to regulate labour law), and the clear objective of promoting self-regulation by providing possibilities for various stakeholders representing various societal interests to share views, expectations and experience makes a case for considering the MSF and the CSR Alliance from the perspective of reflexive law. As we recall from the brief introduction in Section 2, reflexive law is characterised by dialogue and learning, and exchange between social subsystems of their expectations and demands of each other, as well as best practice and other modalities which will support reflection within and between social subsystems. Arguably, the possibilities created by the MSF and the CSR Alliance for interaction and learning between different social subsystems, including not only companies, NGOs and political and regulatory institutions but also the different types of legal systems with regard to human rights normativity, further strengthen the relevance of taking a reflexive law approach.

Moreover, the background to the EU Commission's CSR initiatives resembles that which led to the formulation of the theory of reflexive law: societal needs translate into policy objectives and needs to be addressed through effective regulatory action; the inability or inefficiency of formal legislation or law-making to deliver; and the potential role of the private sector to contribute to solving the problems.

The theory of reflexive law was formulated as an alternative to traditional instrumental or substantive law-making and its opposite: complete deregulation. It recognised that the regulatory strategies of welfare states in the 1970s and 1980s were ineffective for addressing environmental degradation, unemployment and social inequalities and other societal concerns, which required the co-operation of the economic subsystem for their solution (Teubner, Reference Teubner1983; Reference Teubner and Teubner1986; Reference Teubner1993, esp. p. 66). The theory argued that substantive law (aiming at particular, often welfare-oriented, goals through state-made formal regulation and specific standards) was unable to meet the demands that result from an increasingly differentiated society. Reflexivity was argued as a middle path: it would lead to more adequate legal regulation produced in internal reflection within the pertinent organisations or subsystems, through a sort of regulated autonomy (Teubner, Reference Teubner1983; compare Scheuerman, Reference Scheuerman2001; Lobel, Reference Lobel2005). To promote reflexive regulation, governmental bodies may establish procedures and organisational fora and frameworks that facilitate internal discourse among other organisations and their adaptation to external demands, concerns and expectations through self-regulation. Authorities may guide the process by providing normative guidance, but normative outputs are ideally made by those who will be subjected to the output and who represent the interests at stake.

Reflexive law assumes three types of reflexivity: reflection within public authorities on their own regulatory capacities; authorities setting up procedures for self-regulation or co-regulation by those to be subjected to the resulting normative output; and reflection by the latter upon their actions and their impact and needs for adopting normative guidelines or correctives.

As a regulatory strategy, reflexive law builds on the assumption that regulation is more effective and more easily accepted if created by those directly affected by it. The focus is on establishing procedures that allow for a range of those involved or affected (or, in CSR language, stakeholders) to have access to influence decisions or rules that are made and the way that conflicts are solved. Reflexive regulation does not directly regulate behaviour to realise particular, predetermined goals. Instead, it aims at encouraging processes of reflection leading to self-regulation, based on the particular organisation's reflections on its impact on society, and the sharing of views, expectations and experience. Through the procedures of reflexive regulation, authorities retain control, whereas other participants are offered a learning process that enables them to reflect on the consequences of their actions in relation to the community and to integrate societal needs and demands through self-regulation. To limit the abuse of power by stronger participants in the process, reflexive law assumes a degree of neutralisation of power disparities by institutional procedural mechanisms (Teubner, Reference Teubner and Teubner1986, pp. 316–19; 1993, p. 94). This establishes a connection to the common usage of ordinary procedural law to ensure the rights and opportunities of individuals and groups, especially in relation to more powerful ones, such as the relationship between citizens and authorities in human rights law and national administrative law.

As examples, social reporting schemes such as those introduced by a number of European states contain reflexive law elements by defining procedures (such as for stakeholder dialogues) which allow companies to learn about the expectations and demands of stakeholders and to reflect these in their self-regulation. Certain labelling schemes, such as governmentally authorized and controlled eco- or sustainability labelling schemes, contain reflexive law features by providing for a process of the reflection of societal expectations of the product, producer and production process, and internal reflection leading to the decision to adopt the label and self-regulate so as to honour the conditions for the use of the label. The establishment of social reporting schemes, as well labelling schemes, assumes authorities' reflection on their regulatory capacities with regard to achieving the desired end result. Indeed, the theory of reflexive law has been applied to such schemes, primarily at national level but also at EU and transnational level. Touching on environmental management and labelling as self-regulation and auditing (Orts, Reference Orts1995a, pp. 1311–13; 1995b, pp. 788–89), non-financial reporting (Hess, Reference Hess2008; Reference Hess1999), globalisation of law and transnational private regulation (Scheuerman, Reference Scheuerman2001), and labour law (Arthurs, Reference Arthurs, Bercusson and Estlund2008; Deakin and Hobbs, Reference Deakin and Hobbs2007; Deakin, Reference Deakin and Alston2005; Rogowski, Reference Rogowski, Priban and Nelken2001; Reference Rogowski, Ralf and Ton1994; Rogowski and Wilthagen, Reference Rogowski, Wilthagen, Ralf and Ton1994; Wilthagen, Reference Wilthagen, Rogowski and Wilthagen1994), several of these are closely related to substantive issues that are often considered to come under the umbrella of CSR.

While reflexive law has often been seen to be primarily procedural, Deakin and Hobbs (Reference Deakin and Hobbs2007) demonstrate that reflexive regulation may be oriented towards a specific substantive output. Also of interest in the current context, in his study of the EU's Eco-Management and Audit Scheme (EMAS) and other environmental regulatory schemes, Orts (Reference Orts1995a, p. 1287) found that many environmental laws are reflexive in character but appear to be so “almost unconsciously […] almost randomly, without a theory informing them” and therefore less effective than intended.

Arguably, reflexive law is also normative in its own right. This normative quality comes out through the theory's emphasis on two interrelated aspects: the participation of the economic subsystem and civil society in the development of norms, and the establishment of a balance of power between participants in this process. The latter reflects back on procedure as a power balance should be achieved through the procedures set up for the reflexive regulatory process. In establishing reflexive regulatory processes, authorities therefore need to be aware of the normative requirement that procedures take power disparities into account to ensure that the procedures cater for equal possibilities for all affected to inform the substantive normativity. However significant this is for the legitimacy of the reflexive regulatory process and its substantive outcome, the technicalities of how to provide the process with the procedural set-up to balance power disparities is one of the points about reflexive law theory that Teubner has not developed in detail. In fact, the omission of detailing this important part of the theory has led authors who have otherwise welcomed the theory of reflexive law to express a degree of hesitation towards its applicability (Sand, Reference Sand1996, p. 94, Scheuerman, Reference Scheuerman2001, p. 86; Neves, Reference Neves2001, pp. 263–64; compare the discussion by Koudiadaki, Reference Koudiadaki2009, esp. pp. 411–12). The significance of the institutional and procedural set-up to counter power disparities will be further explored below.

The EU Commission's approach to regulating CSR in Europe has been to involve affected partners while indicating from the outset that CSR was not to be an issue of mandatory requirements on the basis of directions issued by (EU) authorities. This view is reflected in the Commission's definition of CSR and also in the announcement by Commissioner Spidla at the relaunch of the MSF in 2006 of CSR. As we saw above, on this occasion the Commissioner stressed both that CSR should be developed in a public–private partnership drawing on dialogue, and that international human rights law be a relevant normative source to inform CSR. The former part of the statement especially indicates a view that for CSR regulation to be effective, in the sense of being accepted by those for whom it creates (societal) obligations, it should not be based on directives by authorities. Indeed, this might be a sort of a compromise between the results of the 2002–2004 MSF that CSR was not to be subject to formal regulation, and a view of the Commission that there is a social interest in CSR being subject to some form of regulation, even if through co- or self-regulation. Accordingly, the Commission has established channels through which stakeholders have been able to meet and co-operate to create substantive regulation on CSR. With the 2002 Communication and the 2002–2004 MSF, the Commission offered invited stakeholders the opportunity to set the framework for CSR regulation in the EU. The MSF launched in 2002 involved not only businesses as entities directly affected by CSR expectations, but also other stakeholders, such as civil society organisations representing employees and NGOs representing communities. The Commission documents indicate that the process, which was launched with the MSF, recognises that societal interests demand that companies take on certain responsibilities for their actions in society and for the role they may play in promoting welfare policy objectives. This approach to regulation has the key characteristics of reflexive law. The strategy is at once one of co-regulation and one aimed at corporate self-regulation. The CSR Alliance was established as a learning forum for private companies, aimed at exchanging best practices. The learning forum approach is a key feature of reflexive law. The use of a learning forum approach indicated that that the regulatory procedures initiated by the Commission work along the lines of ideas like those of reflexive law.

Approached along these lines, the Commission's efforts to promote the regulation of CSR in EU through the establishment of public–private fora to provide for dialogue, learning and ultimately the internalisation of societal concerns in companies, may be characterised as an example of reflexive law. Like the EMAS in the example analysed by Orts (above, this section), the MSF and CSR Alliance were reflexive in character but seem to be so almost unconsciously or randomly, without a theory informing them. Indeed, as further discussed in the following section in relation to power disparities among stakeholders, this may be why the MSF was less effective than intended, both with regard to the establishment of a guiding framework for CSR and particularly, and as further discussed in the following section, with regard to integration of human rights normativity.

The fact that the regulatory approach was not fully effective with regard to parts of the original substantive objectives, precisely because it was not informed by relevant regulatory theory, does not negate the relevance of theoretically approaching the MSF or other EU CSR initiatives from the perspective of reflexive law. On the contrary, doing so provides a framework for analysis which assists in understanding the processes from a legal perspective and in explaining some of the failures of the MSF in terms of norm-setting, particularly because of the lack of fulfilment of the normative requirement of balance of power. This also helps to explain civil society's reservations about the CSR Alliance due to lack of participation.

Although the theory of reflexive law was originally developed for application in a national law context, the observations above indicate that it may also be relevantly applied in a context of inter-governmental regulation. The following section draws on the theory to gain a more detailed understanding of the discursive process and outcome of the 2002–2004 MSF and the launch of the CSR Alliance with regard to the role played by international human rights law for CSR normativity in the context of EU efforts to promote European companies' self-regulation.

7. The MSF and the CSR Alliance discourses on the form of regulation and the role of international human rights law

The analysis in this section combines the theoretical approach of reflexive law, including its normative focus on procedures to neutralise power disparities among stakeholders, with the discourse analysis inspired readings of the MSF and Alliance texts in sections 5.2 and 5.3. The objective is to analyse what those texts and readings tell us about the way the arguments and positions promoted by participants worked to influence the result of the MSF and the establishment of the CSR Alliance. We want to see how these arguments, positions and participants influenced the construction of CSR with particular regard to the form of regulation of CSR and the role of international human rights law as a normative source. Based on that, we will assess how the reflexive law procedure dealt with power disparities between participants (to the extent that such discursive power was not simply based on the more convincing argument).

We have established above that reflexive law is a useful theoretical approach for understanding the MSF and the CSR Alliance. Accordingly, this section will focus on what actually happened within the procedural framework established by the Commission for the MSF, what the results were and why they arose, and how this influenced the establishment of the Alliance. The subdiscourse on whether CSR should be ‘voluntary’ or ‘mandatory’ is material on its own, but also for the role of international human rights law to inform CSR normatively. References to international human rights law or specific references to human rights standards can be made explicitly in Member States' statutory regulation, establishing binding obligations for companies.

Recall a few key points of the theory of reflexive law: through establishing procedures for the exchange of views and concerns, reflexive law is based on a learning process building on the exchange of views and experiences of those subjected to regulation, and in some cases also those affected by actions of the former. Reflexive law theory recognises that power differences between participants in the reflexive regulatory process may exist and should be minimised through procedures.

At the launch of the 2002–2004 MSF, the Commission clearly indicated that it intended CSR to be informed by certain normative sources, in particular international human rights law and related ‘internationally agreed’ instruments. Thus, the Commission did not limit itself to establishing the procedure for regulation, but also indicated its view on social interests to be considered and of substantive norms. Although the Commission insists that CSR is a voluntary action and not law, law – albeit not law that directly binds companies, but international law which addresses itself directly to states – is awarded a major role as a normative source of CSR.

Initially, the MSF in its mandate and debate expressed clear links between CSR, European and international law, and the aims of the Lisbon strategy, including inclusiveness and responsible competitiveness. From this perspective, law, including international human and labour rights law, is a clear and obvious source for CSR, and CSR-informed business practice an obvious avenue towards implementation of the Lisbon goals. In later meetings, the links to law, including international law, were downplayed in favour of a debate on CSR as voluntary rather than mandatory. The links between business action and the implementation of the Lisbon goals were also played down. The Final Report, however, reiterates the basic links, albeit in a way that does not integrate human rights law with operative aspects of CSR.

As the objective of the 2002–2004 MSF was not met with regard to establishing a framework or, arguably, with regard to the role that international law on human rights and particular human rights instruments was given by the 2004 Final Report, the Commission launched a new initiative in 2006, the CSR Alliance, and also organised a relaunch of the MSF. In this case, the regulatory strategy was more clearly aimed at self-regulation, but in a two-pronged way: the MSF is a channel for keeping the process going, and as we saw in Commissioner Spidla's speech, and also serves as input on normative issues and normative sources.

The Communication which announced the Alliance refers to those same instruments of international law as were mentioned in the MSF Final Report, and the launch of the MSF 2006 reintroduced them. Although the references are limited to the instruments agreed to by the MSF, these steps indicate that the Commission is serious about getting companies to consider human rights, but also that it is struggling with the approach it can take to make companies internalise this expectation.

The Commission's response to the outcome of the MSF through the subsequent establishment of the Alliance indicates that the Commission considers CSR a sine qua non for the implementation of paramount societal challenges of the EU and its Member States in relation to employment and inclusiveness and in relation to the future of European business in a world of increasing global competition. The Commission places a considerable part of the responsibility for meeting these challenges on business by tying together the relevance of CSR to meeting the goals of the relaunched Lisbon strategy and sustainable development, with companies as the main players in CSR, at the same time reiterating the role of instruments of international law, agreed to in the MSF Final Report, as key informing sources of business action for CSR.

The debates and intentions surrounding the MSF and the Alliance indicate that in the Commission's approach to CSR as a modality for meeting societal challenges on employment, inclusiveness and sustainable development, law and legal institutions not only have a considerable role, but the role of international law as a source of the substance of CSR is also significant. International law takes on relevance for CSR in an EU context because the obligations of Member States to implement human rights, e.g. of non-discrimination and labour rights, may support EU objectives of inclusiveness and employment, as well as their regard to responsible competitiveness and the implementation of EU human rights policy objectives outside the EU's borders. The analysis shows that law and law-based rules in the view of the Commission and several participants from within civil society, but also some business organisations, play an important role in defining the substance of CSR. In this connection, the Commission and civil society, as well as some business organisations, particularly emphasised international law and soft regulation on human and labour rights (the 1998 ILO Declaration, the OECD Guidelines, the UN Global Compact, and a range of UN and European hard law instruments on human and labour rights). The analysis also shows that interest groups (stakeholders) understand this and co-operate in networks comprised of business, trade unions and NGOs to influence the substance and form of CSR regulation to counter or promote corporate responsibilities to lift tasks which are state obligations not only in traditional welfare state thinking (e.g. promoting employment opportunities) but also according to international human rights law.

As we saw above, industry argued mainly against mandatory regulation and against basing CSR regulation on hard law instruments of international law. NGOs and trade unions argued in favour of binding regulation based on hard international law. The Commission initially referred to international human and labour rights law, but in relation to the construction of CSR in a European context through the MSF, the Commission was not able to bring forward the international law point to have much bearing on the outcome. This raises issues concerning the ability of the Commission's choice of regulatory strategy to bring about certain substantive outputs, and its ability to balance inputs as part of this process.

As a regulatory strategy, the MSF 2002–2004 established procedures for input from stakeholders with the aim of producing a substantive normative output. The actual output, as well as the role played by business for the outcome of the 2002–2004 MSF, indicates that the procedure did not effectively deal with power imbalances between stakeholders. This is supported by the observation that the Commission – as reflected in the 2006 Communication and in Commissioner Spidla's speech – still considers human rights an important normative issue in CSR and still considers international human rights law an important informing source. The regulatory strategy of the 2002–2004 MSF did not seem to take power differences much into account. If it had, we could have expected to find as part of the documents or instructions presented as part of the launch of the MSF in 2002 a list of procedural rules reflecting a procedural set-up intended to neutralise power disparities between stakeholders. It seems that the multi-stakeholder strategy was chosen primarily as a means of bringing stakeholders together, to exchange views and subtly attempt to direct the substance of resulting co- and self-regulation. In accordance with the finding above, this indicates that the Commission's objectives and its perception of regulatory difficulties and social needs to be addressed resembled those which led to the development of the theory of reflexive law, but that the Commission just happened to apply the reflexive regulatory strategy without being informed by the theory.

The fact that power differences were not dealt with through the procedure strengthens the assumption that the application of regulatory strategy had not been informed by the theory of reflexive law. The outcome is mainly in accordance with the discursive interests and positions of industry. The MSF outcome relegated international human rights law to a somewhat isolated role with only a few instruments mentioned as key instruments for CSR in a European context. It also demonstrates the power of strategic discourse to shape normativity and achieve results that favour the interests of particular participating stakeholders, especially if such stakeholders understand the positions of others and the formal rules which these are ultimately subjected to, such as in this case doctrine on international law subjectivity. Those not interested in CSR becoming ‘mandatory’ were not only successful in achieving that no references were made to CSR to be regulated by law. They also obtained that references to law were made in such a way as to encourage the Commission and Member States to use law to promote a favourable environment for companies with no specific references between this and CSR action or commitments. In addition, industry successfully ensured that references to human rights obligations mentioned obligations of states and authorities, not companies. Correct as this is according to international law doctrine, it does little to promote human rights normativity among European companies.

The process comprising an ongoing series of meetings and the possibility for stakeholders to present statements is closely interwoven with the outcome both in terms of the form of regulation of CSR recommended by the Final Report and the role awarded to international human rights law. Both in terms of process and outcome, in particular the limited role awarded to international human rights law, the MSF experience confirms the need to procedurally deal with power differences when reflexive regulatory strategies are employed.

The establishment of the CSR Alliance appears to represent a lesson learned by the Commission through the MSF, but it does not represent a more inclusive or balanced approach to dealing with power disparities between economic actors and civil society. The decision to exclude civil society from the Alliance may work in the sense of establishing a ‘safe haven’ for the exchange of views and experience for companies, but it almost killed civil society support for the Commission's initiative and belief in its neutrality and sincerity with regard to promoting CSR.

The process and outcome of the MSF indicate that reflexive law in principle works as a regulatory strategy to encourage co- and self-regulation on CSR and the more limited subdiscourse of business responsibilities for human rights, but that achieving particular substantive results may be difficult. Successfully achieving such results through a process which builds on reflexive law's paradigm of effective implementation as a result of ‘ownership’ created through participation in norm-creation may require very careful design of the reflexive regulatory process to respect its autonomy and still provide substantive normative guidance.