1. Introduction
This article discusses empirical research on business and human rights, and networks of transnational governance in political science. It does so by arguing the case for meta-governance. By meta-governance, this article understands the governance of governance networks, that is, ‘governance-governance’.Footnote 1 It consists of ‘secondary rules’Footnote 2 that define how conflicts of governance may be mitigated and resolved, and who are the actors to which authority may legitimately be delegated in this respect.Footnote 3 The need for meta-governance arises as a result of numerous norm and power conflicts with which transnational governance networks are confronted in practice. These conflicts have the potential to undermine the effectiveness of transnational governance networks and put their legitimacy and very existence into question – hence the need for meta-governance, or mechanisms for conflict resolution. Arguing the case for meta-governance, the article also bridges the gap between legal scholarship interested in normative orders beyond the state, soft lawFootnote 4 and transnational governance networks on the one hand, and empirical debates and findings in political science on the other. Investigating meta-governance will involve empirical analyses of examples of successful and failed approaches to the governance of governance networks. At the same time, it requires us to think theoretically: what are the ‘secondary rules’ that can and should define meta-governance – for example, what is, and what should be, the nature, form and legitimacy of meta-governance? It is thus a topic that is located right at the intersection between normative-legal and empirical social science research.
Before I argue the case for meta-governance in detail, I shall briefly clarify some of the concepts I use here. In particular, as I report from the different disciplinary context of political science, misunderstandings may otherwise arise over the meaning of the terms I use. By governance I understand the making and implementation of rules, norms, standards and decision-making procedures for the production or provision of collective goods.Footnote 5 Hence, governance can consist of normative stipulations; it may, however, also consist of the delivery of material or quasi-material goods such as health care services. Rules, norms, decision-making procedures and standards have in common that they all define standards of appropriate behaviour around which actor expectations converge.Footnote 6 Usually, in political science rules are also considered to be more strictly defined and supported by some kind of compliance mechanism, whereas norms are often understood as being more general and loosely formulated.Footnote 7 Standards and decision-making procedures, in turn, involve some technical aspects of implementation: for example, ISO 14001 is an environmental management standard that effectively stipulates a system of rules to guide the implementation of substantive environmental standards.Footnote 8
By networks, I understand any set of interconnected nodes (individuals, groups, formal organisations) whose relationships and patterns of interaction shape the behaviour of any single node; any single node, in turn, also contributes to the shaping of the relationships and behaviour of others and of the network as a whole.Footnote 9Transnational refers to interactions that involve a cross-border element and at least one non-state actor.Footnote 10 Networks of transnational governance thus consist of mutually influencing, interconnected nodes, which include non-state actors and a cross-border element in the making or implementation of norms, rules and standards for the production or provision of collective goods and services. I will also sometimes refer to limited statehood in the analysis of the relationship between businesses and transnational governance networks. By ‘limited statehood’ I understand a policy space in which the state is unable or unwilling to set or enforce strict regulation effectively.Footnote 11 Policy space refers here to an area that is defined functionally (for example, ‘the environment’ or ‘human rights’), historically, socially (that is, in relation to certain groups or sub-segments of society) or territorially, and which can exist within a state at the local level, across states at the transnational level, or globally. Another important concept in the analysis of business and governance networks in relation to human rights is corporate social responsibility, which here refers to the idea that firms contribute to the provision of collective goods, or the reduction of negative externalities of production, by adhering voluntarily to international or transnational norms and standards.Footnote 12 Firms that implement the core labour standards of the International Labour Organization (ILO), for example, may abolish child labour in their factories and thereby, potentially, reduce the negative externalities of production. They thus act in a socially responsible way. This article shows, however, that even when firms adhere to standards of corporate social responsibility, the situation often remains normatively complex, and contested. For example, when firms implement international standards, they frequently run into conflict with local communities which have an idea of the common good different from that envisioned by international standards. It is to this type of conflict that this article draws our attention, making the case for meta-governance.
In what follows, I will introduce the empirical research in political science referred to, which points to the need for meta-governance. Political scientists are often, if not primarily, concerned with questions related to the behavioural consequences of, and causes for governance. Hence, they ask themselves why and under what conditions firms would adhere to standards of corporate social responsibility – human rights, labour standards, and environmental standards – as laid down by transnational governance networks. Section 2 of this article discusses the empirical approaches that have been suggested in response to this question, and the findings that these approaches have generated. Two such approaches stand out. The first focuses on external pressure factors that make firms adhere to corporate social responsibility norms and standards; the other emphasises intra-firm drivers. The former is the more established approach in the literature, whereas the latter (which emphasizes internal drivers) has emerged only recently. Since the internal driver-based approach is less familiar – and because I have contributed to it in my own work – I will discuss it in more detail than the external pressure factor-oriented approach, and in so doing will also present some empirical examples.
Political scientists have recently become aware that firms are not only the addressees of transnational governance networks (that is, the ‘rule takers’). They are also norm makers and, in this context, become constitutive nodes and drivers of transnational governance networks. This awareness has opened up new research agendas and avenues, introduced in Section 3. They concern mainly the (interaction) effects of transnational networks of business governance, particularly in areas of limited statehood. Again, two approaches stand out. The first asks why, and under what conditions collective goods provision of transnational governance networks involving businesses is efficacious in areas of limited statehood, and when it is not. The second analyses the interaction effects of transnational governance networks. The plethora of governance networks in existence today creates a situation of normative fragmentation and complexity. What explains this fragmentation, and what are its consequences for the legitimacy, effectiveness and viability of governance as an approach to social ordering?
The discussion of empirical findings that have been brought to light in response to these questions shows that, in practice, networks of transnational business governance confront three important problems. First, such networks confront norm and power conflicts with local communities and governments, which often make them ineffective. Second, collective goods provision through networks of business governance may result in a clustering of governance services in the industrial centres of emerging markets – rather than where the need for collective goods is greatest; this raises questions of distributional justice. Third, opportunistic behaviour of firms may undermine, and even destroy, the legitimacy and effectiveness of transnational governance networks. To be able to deal with these problems, transnational governance networks are in need of meta-governance, and meta-governors to provide this meta-governance. Meta-governance, however, is still entirely under-explored, in particular in areas of limited statehood – hence, my call for an inter-disciplinary research agenda on the empirical reality and potential normative sources of meta-governance, involving political scientists and legal scholars.
2. Business Behaviour and Regulatory Standards
Since the early 1990s with the end of the Cold War, a new wave of transnationalist thinking evolved in political science.Footnote 13 This thinking soon became closely related to, and merged with new approaches to global governance that had also developed at that time.Footnote 14 These approaches deviate from earlier approaches ontologically in that they go beyond the discipline's traditional ‘methodological nationalism’:Footnote 15 they conceptualise non-state entities as actors of crucial relevance for global governance. It is in the context of these intra-disciplinary developments that political scientists became interested in transnational networks of governance and businesses as actors of political relevance in relation to such networks, and to social ordering in general.Footnote 16
This interest, of course, also reflected some significant ‘real’ world phenomena: since the early 1990s the world economy has become more and more integrated as a result of drastic trade liberalisation, negotiated in the context of the World Trade Organization (WTO)/General Agreement on Tariffs and Trade (GATT) regime, regional economic organisations and bilateral trade agreements. With trade liberalisation, globalisation and limited statehood became major issues of concern for approaches to social ordering beyond the state. Massive relocation of production from the ‘Global North’ (that is, the ‘West’) to the so-called emerging markets such as the BRICS (Brazil, Russia, India, China and South Africa) was a consequence of this trade liberalisation.Footnote 17 While these ‘newly industrialising states’ are strong vis-à-vis their societies and, increasingly so, in the international system, they often lack the capacities for industry regulation.Footnote 18 Today, therefore, most production takes place under conditions of limited statehood in these emerging market states, so that firms can wield at their leisure and externalise the costs of production. A recent example illustrates this. In Dhaka (Bangladesh), a factory building collapsed on 24 April 2013, killing more than 1,000 workers.Footnote 19 The investigation into this tragedy brought to light that, in order to keep costs at a minimum, the textile firms that operated in the factory building systematically ignored security and safety standards. This blatant neglect of basic standards of safety caused the collapse. The factories were overcrowded beyond the carrying capacity of the building. The investigation also revealed that the workers, mostly young women, were severely mistreated in the factories before the collapse. Regulations that prohibit such business misconduct do exist in Bangladesh,Footnote 20 yet they are not being enforced so that firms are able to ignore them in order to maximise profits.
While this example illustrates the problem of negative externalities of business activities in areas of limited statehood, it does not mean that human rights are generally of no concern to business managers. Since the early 1990s and concurrently with increasing trade liberalisation, calls for corporate social responsibility have emerged in reaction to reports on ‘sweatshop’ conditions in areas of emerging market production.Footnote 21 These calls were fuelled by non-governmental organisations, consumer and media campaigns in the United States (US) and Europe against firms that showed corporate misconduct in emerging markets. The activists demanded that these firms show respect for human rights and the environment. In 1999, at a World Economic Forum meeting in Davos (Switzerland), Kofi Annan, then UN Secretary-General, therefore suggested the establishment of the Global CompactFootnote 22 – a public–private partnership between the UN and participating corporations. The Global Compact addresses firms directly, encouraging them to respect basic labour and environmental concerns. By the late 1990s a whole plethora of networks of transnational governance such as the Global Compact had emerged as a consequence of activism by the corporate social responsibility movement,Footnote 23 demanding that firms integrate human rights and environmental standards in their daily business operations around the globe and in their supply chains. The demand for supply chain regulation raises important questions regarding the limits (or the absence thereof) of corporate social responsibility: can, or should (buyer) companies be held responsible for the behaviour of legally and organisationally different (supplier) businesses in countries far away from their own place of incorporation?
Generally, the question has been answered in the affirmative.Footnote 24 However, do companies really have the means to effectively regulate supply chains? Are they technically at all capable of doing so? Empirically, this seems to be the case: global value chain analyses in business studies show that ‘lead firms’Footnote 25 – which are often large multinational corporations from the Global North – can and do organise and upgrade the supplier networks in which much of the added value element of the end products they sell is created.Footnote 26 What is more, when buyer–supplier relationships are marked by mutual dependency, analyses have found that buyers actively regulate suppliers in order to mitigate the risks inherent in such dependencies.Footnote 27 The resulting supply chain regulation may not simply relate to product aspects such as quality, but also to the production process, and detail environmental, health, social and labour standards. Thus, the problem of supply chain regulation in relation to human rights and corporate social responsibility is not that it is not feasible technically or organisationally, but rather that in many industry sectors lead firms try to avoid the substantial costs it involves. This is particularly so when the mutual dependency between buyers and suppliers is low (such as in the textile industry).
In light of these considerations it seems highly justified that legal scholars now discuss the possibility of normative ordering beyond the state and beyond the framework of conventional international law.Footnote 28 In this vein, they have investigated the nature, relevance and potential sources of legitimacy of transnational soft law and of networks of transnational governance.Footnote 29 In political science, research into these phenomena and developments was for a long time interested primarily in the exploration and explanation of the relationship between firms' behaviour and regulatory standards under conditions of limited statehood. Initially this relationship was defined mainly in negative terms. A common argument was, and still is, that economic globalisation allows firms to relocate to where production costs are lowest – that is, to where regulatory standards are lax – which in turn triggers a regulatory ‘race to the bottom’ among states:Footnote 30 competitive downsizing of regulatory standards to obtain comparative cost advantages. From this perspective, networks of transnational governance such as the Global Compact are incompatible with the interests of businesses in maximising profits, and can do little more than conceal the true, exploitative nature of their capitalist interests.Footnote 31 In fact, from this perspective, they produce instances of ‘blue’ and ‘green washing’:Footnote 32 firms assuming a ‘green’ (eco-friendly) or a ‘blue’ image in the context of the UN Global Compact to portray themselves as good corporate citizens without, however, changing any of their exploitative business practices. Indeed, there are many cases in which corporations continue to violate workers' rights in the context of ‘sweatshops’, such as the example in Dhaka described above. However, more recent studies have shown that the relationship between firms' behaviour and regulatory standards is not as straightforwardly negative as presumed by the ‘race to the bottom’ argument.
In fact, since the early 2000s, the ‘race to the bottom’ argument has been successfully challenged. Scholars point to many examples of corporate social responsibility, suggesting that transnational governance networks and private regulation can align profit maximisation with human rights and other governance concerns.Footnote 33 Some have even argued that human rights and business activities are positively related, and that international trade and foreign direct investment diffuse labour and social standards.Footnote 34 Rather than a ‘race to the bottom’, the globalisation of business activities and value chains may result in a ‘race to the top’.Footnote 35 The debate today is no longer so much about whether businesses drive either a regulatory ‘race to the bottom’ or a ‘race to the top’, but rather about why and under which conditions which dynamic predominates.Footnote 36 There are some intriguing answers to this question.Footnote 37 I suggest differentiating generally between answers that concentrate on external pressure factors working on firms in their strategic environment, on the one hand, and internal drivers of business social conduct on the other.
2.1. External Pressure Factors
External pressure factor-oriented explanations for the adoption of labour, social, health and other human rights by businesses, as well as environmental standards, assume – not unlike the advocates of the ‘race to the bottom’ argument – generally hostile preferences of firms towards regulatory standards. Firms are profit-seeking entities; the adoption of voluntary standards such as a minimum wage or the abolition of child labour is costly. Therefore, if not coerced into it, firms choose according to external pressure factor-oriented approaches to ignore labour and social standards – and standards of corporate social responsibility in general. In ‘strong’ consolidated states, self-regulation by business is therefore usually brought about by ‘legislative threats’: the so-called ‘shadow of hierarchy’.Footnote 38 State agencies threaten to regulate, for example, an industry sector – which, as a reaction to this threat and in order to avoid regulation, engages in self-regulation. However, in emerging markets, which are often areas of limited statehood as far as industry regulation is concerned, the ‘shadow of hierarchy’ is not available. The state here is either unwilling or incapable (because of the absence of regulatory capacity) to set and/or enforce regulation. In addition, the shadow of hierarchy cast by state agencies over large buyer firms in the Global North, which incentivises them to engage in self-regulation at ‘home’, does not usually reach supply chains in the Global South ‘abroad’ – or if it does it is only under exceptional circumstances.Footnote 39 Norms and standards of corporate social responsibility concern so-called ‘process regulation’.Footnote 40 Process regulation formulates requirements for the creation of a product such as a minimum wage, abolition of child and slave labour, energy consumption reduction in the production process, or recycling. Process regulation is usually distinguished from ‘product regulation’, which defines the physical attributes of the production output, that is, of a service or product. Examples of product regulation include measures of quantity and size as well as quality-related aspects. Unlike product regulation, process regulation cannot be used as an import barrier under the WTO/GATT regime. That is to say, ‘strong’ states of the EU and the US or Japan cannot prevent imported goods from the Global South from entering their markets based on, for example, concerns about the labour conditions in the producing country. In consequence, states in the Global North have no means of forcing (or threatening to force) importing firms in the Global North to regulate their supply chains in the Global South under the current world trade regime.Footnote 41
Why then would businesses in the Global South nonetheless voluntarily adopt human rights-related standards and corporate social responsibility policies?Footnote 42 Self-regulation can also emerge as a reaction to the regulatory vacuum in areas of limited statehood, as this may expose business activities to extreme uncertainties and risks.Footnote 43 More precisely, under conditions of ‘functional equivalents’Footnote 44 to the shadow of hierarchy, firms may feel particularly exposed to the uncertainties and risks involved in areas of limited statehood. These functional equivalents can incentivise firms to comply with international or transnational standards of corporate social responsibility. For example, some firms rely heavily on their brand name reputation, and invest significantly in marketing campaigns to support their brand name image. In this situation, honouring human rights and other voluntary standards may serve their aim to create a positive image of themselves in the eye of the public and consumers. Adhering to high standards may also provide firms with protection against watchdog NGOs and media campaigns, and consumer protests against alleged corporate misconduct. NGOs and consumers are thus potential pressure factors in the strategic environment of firms that may motivate them to adopt costly standards, even in the Global South and under conditions of limited statehood.Footnote 45
This argument has been extended to whole value and supply chains, and in this context has been used for a general redefining of the contested relationship between labour, social and human rights on the one hand, and international trade on the other. This redefinition goes as follows:Footnote 46 NGOs and consumers in countries with strict regulation and a strong emphasis on labour and human rights (the ‘Western’ world) put pressure on firms to require their entire business operations – including those in offshore production locations and supply chains – to comply with basic human rights and labour standards. This pressure from NGOs and consumers makes importing firms in these ‘Western’ countries source from exporting offshore production locations and from firms in these locations with a good record on human rights and corporate social responsibility.Footnote 47 This preference of importing firms, in turn, encourages exporting firms in emerging markets to adopt high labour standards and lobby their governments to improve human and employment rights.
This redefinition, which essentially suggests a social-normative entrenchment of global markets,Footnote 48 has been invoked to explain why, empirically, an increasing number of macro-quantitative econometric studies find that labour standards and human rights spread with international trade and foreign direct investment.Footnote 49 It should be mentioned here, however, that this assertion of a positive relationship between global business operations and human rights is also contested. Empirically, events such as the recent collapse of a factory in Dhaka referred to above, or reports about slave labour on Thai prawn vessels,Footnote 50 among many others, point to a negative rather than a positive relationship between global business operations and human rights. This anecdotal evidence is also backed up by econometric studies that reach different conclusions from those cited above. These other quantitative studies suggest a neutral or even an outright negative relationship between international trade and regulatory standards, rather than a positive relationship.Footnote 51 The same can therefore be said regarding the external pressure-factor oriented explanatory model as can be said regarding the ‘race to the bottom’ argument: it seems to depend on further context conditions whether the theorised logic unfolds or not. Rather than striving for a general theory concerning the effects (positive, neutral or negative) of international trade on regulatory standards, we should seek to define why and under which conditions international trade exerts what kind of effects. Recent studies have pointed to the importance of internal drivers in this respect.
2.2. Internal Drivers
Internal driver-oriented studies focus on intra-organisational dynamics, motivations and incentives in order to shed light on why and under which conditions firms show a respect for human rights and take their social responsibilities seriously.Footnote 52 Analytically, the main difference from external pressure-based analyses is that internal driver-oriented analyses take into consideration the possibility of varying preferences of firms in relation to human rights and voluntary standards. External driver-based studies assume – not unlike the ‘race to the bottom’ argument – that profit-maximisation and the establishment of social and environmental standards are generally incompatible (see above). Indeed, there are many examples to illustrate this assumption, such as the textile factories that operated in the factory building that collapsed in Dhaka. In such cases only strong and continuous pressure from NGOs and consumer boycotts can possibly make them respect human rights, if at all. Internal driver-oriented analyses do not deny the importance of such external pressure factors in cases like this; however, they argue that external pressure factor-based analyses do not give us the full picture.
To illustrate the point, I will present a set of cases involving textile and automotive firms from South Africa, and their fight against HIV/AIDS in the workplace.Footnote 53 The cases are based on my own field research – mostly semi-structured personal interviews with managers, factory workers, union representatives, NGOs, government agencies, journalists, activists and researchers. The interviews were conducted between 2006 and 2008, during four extended research trips to the country. In addition, the information revealed in the interviews was updated during a research trip in summer 2013. The interviews I held were (mostly) recorded and transcribed; where this was not possible, I wrote detailed field notes directly after the interview.Footnote 54
Why choose these firms and their HIV/AIDS programmes? The two industry sectors are particularly apt for an analysis of the drivers of corporate social responsibility as they are very different. The automotive industry illustrates high degrees of fixed capital and high barriers to market entry, high-tech production, high productivity, skilled labour, and a technically complex production output. The textile industry, by contrast, stands out for its relatively low capital investment, low market-entry barriers, low productivity, high labour intensity, and relatively simple products. Hence, if an argument for internal drivers of corporate social responsibility holds for cases in both industries despite their differences, this indicates the validity and explanatory power of this argument. In addition, the two industry sectors are particularly relevant for the economic development of developing countries. Hosting a textile industry is generally regarded as a ‘first step’ to industrialisation and modernisation for developing countries, whereas the automotive industry is considered as the ‘next step’ afterwards, as it is associated with technological upgrading and middle class jobs. South Africa is one of the countries that is most heavily affected by HIV/AIDS,Footnote 55 with an adult prevalence rate of about 18 per cent.Footnote 56 Though the country has always had world-class hospitals and universities, only a small percentage of wealthy citizens – among which ‘whites’ are grossly over-represented – can afford to pay for the private health care insurance that guarantees access to adequate treatment in these top institutions. Until today most South Africans, particularly those in the predominantly black townships, are notoriously under-supplied with medical services.Footnote 57 Many do not have access to the most basic medical assistance. Drug coverage for persons suffering from AIDS is an estimated 20 per cent nationwide, although it is much lower in the townships. Under township conditions, HIV is therefore still often a death sentence in South Africa: untreated, it leads to rapid deterioration of the physical condition, and to an early outbreak of AIDS. Not surprisingly, therefore, the intense spread of HIV/AIDS throughout the country from the early 1990s created an unprecedented health crisis shortly afterwards among township inhabitants, who are also the main blue-collar workforce in the country. This health crisis was soon transformed into a major social and economic crisis, which today severely affects family and community life, gender relations, and the productivity of business entities.Footnote 58
Crossley Carpets is a firm that produces hotel carpets for large chains such as the Hilton and Marriott in Durban, KwaZulu Natal.Footnote 59 In light of the emerging health crisis in the country, the company decided to provide comprehensive HIV/AIDS-related health care services for its employees and their families. It has an on-site clinic that employs specifically trained nurses and arranges visits by doctors. It hands out immune boosters to employees who have contracted HIV and educates them on how to avoid an outbreak of AIDS. What is more, the clinic treats the so-called ‘secondary diseases’ of HIV/AIDS such as tuberculosis, organises sex education training and awareness-raising campaigns, and distributes condoms and information pamphlets among employees.
What is remarkable about this case is that Crossley Carpets was not facing any external pressures when it decided to confront the problem of HIV/AIDS; nor was the firm pushed by consumers, nor forced by government, NGOs, unions or associations. Instead, it decided to fight the disease to protect the rare skills on which its production depends.Footnote 60 It thus did so for intra-organisational, internal reasons. The firm invests significantly in the training of employee skills, which are not directly available on the South African labour market and without which the business could not survive. As the company's Human Resources Director explains, ‘the skills that we require here, most of them cannot be found or recruited from anywhere because we use specialized machines. Hence, the difficulty with respect to a high prevalence rate of HIV/AIDS is that you will be losing some very good, experienced workers that have attained these unique skills. The HIV/AIDS program is one way of trying to retain these skills'.Footnote 61
Crossley Carpets thus has an intrinsic motivation to provide health services for its employees. This intrinsic motivation is based on economic rationale. The firm needs the unique skills of experienced employees; it cannot afford to lose them, and therefore provides high-level health services for them.
On a more general level, what this case illustrates is ‘asset specificity’Footnote 62 in the employment relationship as an internal driver of a business entity's social conduct.Footnote 63 Asset specificity refers to investments that, once made in support of a task, would subsequently lose productive value if redirected to support another task. Applied to labour relations within firms, asset specificity refers to investments by firms in training schemes that teach employees skills that are difficult to come by on the labour market – as in the case of Crossley Carpets.Footnote 64 Such investments motivate firms to adopt high social, labour and health standards and become frontrunners in promoting human rights.
To demonstrate this claim further, consider the following sample of firms, taken from the research I conducted in South Africa referred to above.Footnote 65 The firms consist of four large (Sä, Sh, Fe and AL) and three small to medium-sized (FE, V and CS) suppliers in the automotive industry – and two medium to large textile firms (CC and DBA).Footnote 66Figure 1 maps them according to the degree to which they fight HIV/AIDS in the workplace and provide comprehensive HIV/AIDS-related health services for employees.Footnote 67 It illustrates a stark variation among South African-based firms in respect of their engagement in the fight against HIV/AIDS. Some firms (such as the auto supplier CS) do not take any measures at all, while others (such as Sh) offer comprehensive health services to employees.
Variation of HIV/AIDS programmes among car and textile firms in South Africa
The set of firms illustrated in Figure 1 was chosen according to the principles of a y-centred research design.Footnote 68 Y-centred designs select cases on the dependent variable ‘y’ (here, HIV/AIDS programmes), while at the same time excluding commonly known independent variables as explanations (such as NGO and consumer pressure). They are thus puzzles in the sense that they illustrate variation that seems inexplicable in view of the literature. This invites innovative theoretical and empirical inquiries into the causes of the variation.
What is important concerning the set of cases illustrated in Figure 1, therefore, is that the differences among the firms in terms of their HIV/AIDS workplace programmes cannot be explained easily in light of any of the known explanatory factors mentioned in the literature. More specifically, my empirical investigation of the firms revealed that none of them faced any pressure or demands from NGOs, worker organisations, civil society groups, government agencies or local communities to address the issue of HIV/AIDS in the workplace. Hence, with regard to external pressure factors, no explanation for the differences illustrated in the chart is available. There are, however, other potential explanations: the size of the firms could be a factor, given that the firms in Figure 1 differ to some extent in this respect.Footnote 69 Size determines the relative costs per employee in an HIV/AIDS-related health care scheme. Such programmes require investment in infrastructure, such as a clinic and medical personnel, and larger firms can use this infrastructure more efficiently. Therefore relative costs decrease with an increase in the size of the firm. However, a closer look at the nine cases reveals that firm size cannot account for the overall variation. For example, AL is a relatively large auto supplier, whereas V is a much smaller car firm.Footnote 70 According to the factor of firm size, AL should have the more demanding HIV/AIDS programme. However, V's scheme is more sophisticated and comprehensive than that of AL. Another example that demonstrates that firm size does not explain the variation in Figure 1 is the textile firm Crossley Carpets (CC). The firm is smaller in comparison with the auto firms Fe and AL,Footnote 71 but it operates a workplace programme on a higher level than those of Fe and AL. Firm size also cannot explain why firms of similar size engage differently in the fight against HIV/AIDS: for example, CS is roughly as large as FE and V and all three are car suppliers.Footnote 72 However, while FE has a strong programme, CS has no programme at all – and V is somewhere in between both firms in this respect.
A further factor that potentially accounts for the firms' HIV/AIDS programmes concerns the ‘problem pressure’ they experience. Based on the difference in prevalence rates of HIV/AIDS in their factories, the firms may not be similarly inclined to see the need to fight the disease. FE, CS and V confront prevalence rates of about 15–20 per cent among their workers, Crossley Carpets and DBA more than 20 per cent, and Sä, Sh, Fe and AL an estimated 10 per cent.Footnote 73 However, the variance in the HIV/AIDS programmes within these three groups turns out to be almost as stark as that between firms that differ with respect to ‘problem pressure’. FE, for example, runs a relatively strong workplace programme, whereas CS – which is similarly affected by the disease – does not have a programme at all. Similarly, Crossley Carpets is a firm that demonstrates a strong approach in response to the disease. By contrast, DBA shows a weaker approach, despite a similarly high prevalence rate. Hence, it is not prevalence rates per se which explain why and under which conditions firms fight HIV/AIDS. While a minimum level of problem pressure is certainly a precondition for such health schemes to emerge, it cannot explain the variation among such programmes illustrated in Figure 1.
The study from which these cases are taken considers many more potential factors. It shows that ultimately the firms' HIV/AIDS workplace programmes bore down to asset specificity. Those firms that depend on their employees' rare skills, and which have invested in training schemes to create these skills, are the firms that fight HIV/AIDS in the workplace. More precisely, we find the level of investment in such training programmes reflected in the degree to which the firms actively fight HIV/AIDS among employees, and offer workers comprehensive health services. Figure 2 illustrates this relationship between asset specificity and HIV/AIDS workplace programmes. Taken together, the nine cases display, more or less, a linear relationship between asset-specific investments in employee skills and HIV/AIDS workplace programmes.Footnote 74
Asset specificity in relation to HIV/AIDS programmes
To summarise, business behaviour in relation to regulatory standards varies, and is driven by a whole set of motivational factors, external and internal to the organisation of the firm. We can broadly differentiate between firms that lack intrinsic motivation to respect human rights and their corporate social responsibilities and those that have such an intrinsic motivation on account of internal drivers. The former have to be coerced into compliance, whereas for the latter corporate social responsibility is part of their profit-making strategy. Keeping in mind these considerations about firm behaviour in relation to regulatory standards, I will now review literature and research in the field of political science that is directly concerned with transnational governance networks.
3. Transnational Governance Networks
The evolution of explanations of business behaviour in relation to regulatory standards, described above, is important for an understanding of the discipline's current research into and literature on transnational governance networks. Historically, transnational governance networks have evolved since the 1990s as a result of concerns over corporate misconduct. The founding of the Global Compact – the aim of which was to give global business operations a ‘human face’ and to align globalisation with societal concerns for human rights and the environment – illustrates this original motivation.Footnote 75 Transnational governance networks thus were initially institutions that defined norms, rules and standards for globalised business. For a long time political scientists concentrated primarily on firms as the addressees and rule takers of the norms suggested by these networks in order to explain the conditions under which these networks exerted effects on firm behaviour, as referred to earlier.
Recently, however, there has been a growing awareness that the relationship between firms and transnational governance networks is, in reality, more complex. Businesses are not just norm takers and addressees of governance initiatives and networks. They are also norm makers and ‘norm entrepreneurs’,Footnote 76 and in this context are constitutive nodes and drivers of transnational governance networks. The example of Mercedes Benz in East London, in the Eastern Cape province of South Africa, is illustrative in this respect. On account of its dependence on specialised skills, the firm has taken on the fight against HIV/AIDS – and not only in the workplace but in the whole locality in which it is set.Footnote 77 The company cooperates with various development agencies, local government, associations, NGOs and suppliers in a large-scale project that, as a first step, developed a model of an HIV/AIDS workplace programme. The firm then worked with its suppliers and small to medium-sized businesses in East London in order to diffuse this model of the programme in the local value chain. It also ‘uploaded’ the model to the global level in the context of the Global Business Coalition on HIV/AIDS and the World Economic Forum's Global Health Initiative, participating in defining a globally applicable standard for HIV/AIDS workplace programmes.Footnote 78 Clearly, Mercedes Benz here is not the addressee of a transnational governance network; the company is instead a constitutive part of this governance network, and participates in norm making and acts as norm entrepreneur. It is this latter role of business as norm maker and entrepreneur that has inspired much of the new important research and literature on transnational governance networks. I will focus here on two particular strands of research, which have recently led to the publication of special issues in relevant journals.
The first strand of research concerns the effectiveness of collective goods provision by transnational governance networks in areas of limited statehood.Footnote 79 This research is interested in the conditions under which material or quasi-material goods and services, such as health care, can be provided by transnational networks of governance in contexts in which state institutions fail to do so. For example, when are governance networks such as that of Mercedes Benz in East London successful, and when do they fail?Footnote 80 The second strand investigates the complexity, fragmentation and interaction of transnational governance networks.Footnote 81 We are confronted today with a plethora of such networks in the area of human rights and labour and social policies.Footnote 82 What explains this fragmentation and what are its consequences for the legitimacy, effectiveness and viability of governance as an approach to social ordering? I argue that both strands of research show that in practice transnational governance networks confront existential norm and power conflicts that potentially undermine their standard setting and problem solving capacities. It is for this reason that I argue that transnational governance networks are in need of meta-governance.
3.1. Collective Goods Provision in Areas of Limited Statehood
When businesses take over governance functions in an area of limited statehood, does this mean that they will also succeed in fulfilling these functions? A main factor for the likelihood of success or failure is, according to a recent special issue in Governance 27(4), the level of complexity involved in the governance task.Footnote 83 Complexity refers to the degree to which collective goods provision requires continuous, reiterated interaction, and the number of actors that need to be coordinated in order to successfully govern. The classical example of a ‘simple’ governance task is a one-shot vaccination campaign, which contributes to the collective good of ‘public health’, and can be relatively easy to organise. The less complex the governance task, the more likely it is to succeed.
An example of a complex governance task is the fight against HIV/AIDS,Footnote 84 which requires continuous medical intervention and coordination efforts with a whole host of actors. Patients have to be convinced to test for HIV/AIDS and, following a positive test, be willing to be treated. This not only requires the cooperation of patients but also of their families and communities. Moreover, medical staff in clinics must be trained so that they are able to diagnose the disease and effectively treat it. Fighting the disease also involves prevention measures, such as sex education programmes in schools, at the workplace, and in public. The efficacy of governance networks such as that of Mercedes Benz in East London is therefore not at all self-evidently positive. The higher the level of complexity, the greater the number of actors that have to be continuously coordinated – as in the case of Mercedes' attempt to fight HIV/AIDS – with the result that conflicts of interest and, consequently, collective action problems are likely to emerge.Footnote 85 Indeed, on at least two levels transnational governance networks confront norm and power clashes that can make coordination for collective goods provision ineffective, particularly when the governance task is complex. One concerns the interaction with local communities; the other with government agencies of the host country of the governance network.
With regard to local community interaction,Footnote 86 governance networks are usually oriented towards international norms and standards. However, these may clash with local norms, the culture, tradition, routines and power structures of those local communities, which are often the designated beneficiaries of international norms and standards. Consider the example of one of Nike's supplier factories in China in 2008.Footnote 87 The sportswear brand had effectively imposed strict labour standards on this supplier; these standards regulated and limited working hours. Hence, the workers in the supplier factories did not normally work more than eight hours per day, in exceptionally busy times ten hours (in return for compensation), and for not more than five days per week. In the context of China, this (self-) regulation is considered to be rather exceptional: it is not uncommon for employees in the textile industry to work 15-hour days, six or seven days per week. According to many observers, workers in the Nike supplier factories often secretly took on a second part-time job in another factory – so that they still effectively worked for 15 or more hours per day, seven days per week. In so doing they undermined Nike's efforts to encourage workers to regenerate after their day's work in Nike's supplier factory, and to impress the core labour standards set by the ILO. The case thus illustrates a situation in which the designated beneficiaries of workers' rights actually had a different perception of what is beneficial to them, and what is their right to do. From their perspective, economic development is the main priority, and their right to work is more important than the welfare concerns held by Western countries.Footnote 88
HIV/AIDS workplace programmes in South Africa represent another example that illustrates conflicts between business and the local community. Many firms that offer sophisticated workplace programmes to their employees are frustrated by the low participation rates in these schemes. Employees frequently resist the programmes because of the stigmatisation associated with the disease in their community.Footnote 89 Prevention and treatment regimes for HIV/AIDS have put established gender relations, sexual identities and traditional power structures (such as the status of traditional healers) into question. They also clash with local power structures. In townships, traditional healers have important positions; they see their social status and income threatened by the company schemes and sometimes actively advocate against the health services being offered. As a result, those who have contracted HIV/AIDS often remain untreated, thus rendering the company workplace programmes ineffective.
However, when firms take over governance functions, norm and power conflicts may not only emerge in relation to local communities, but also with the local, provincial or national government.Footnote 90 South Africa's HIV/AIDS crisis is again illustrative here – or, more precisely, two cases taken from this crisis. One is the case of Mercedes Benz in East London; the other is of a multinational car firm in Durban. The cases illustrate the varying intensities of conflict between the firms' initiatives and the national government in the firms' attempts to provide HIV/AIDS-related services in the communities of East London and Durban. In Durban, a large Japanese multinational car manufacturer attempted to organise a collective response to the HIV/AIDS pandemic between 2001 and early 2003.Footnote 91 As in the case of Mercedes Benz, the firm's business model is based on skills that are not readily available on the local labour market. The firm therefore offers training to its employees. At the time of the governance network, however, HIV/AIDS prevention and treatment not only had capacity problems on the part of the state; the attempt by the Japanese multinational was also resisted by leading figures of the early Thabo Mbeki government, which was in power during that time and opposed the so-called ‘international scientific consensus’ on HIV/AIDS.Footnote 92 This consensus is based on the fact that HIV causes AIDS and medical treatment prolongs the time span between the infection with HIV and the outbreak of AIDS. The South African government publicly denied the connection between HIV and AIDS, and was openly hostile towards any medical approach to the disease.Footnote 93 A key figure in the official denial of the international scientific consensus was the health minister, Manto Tshabalala-Msimang, who proposed garlic, lemon juice and beetroot as remedies for AIDS. President Mbeki himself also attacked anyone in South Africa who questioned his ‘denialist’ position. In this context of denial the Japanese multinational initiated a multi-stakeholder partnership for HIV/AIDS prevention and comprehensive treatment, composed of the local municipality, the University of KwaZulu Natal and the local chamber of commerce. The partners successfully applied for funding in 2002 from the Global Fund to fight HIV/AIDS, tuberculosis and malaria. The project was to be rolled out throughout the local business world and from there to the local community in 2004, and would have provided full medical services for HIV/AIDS, including antiretroviral treatment. The project was based on the norms for public health programmes for HIV/AIDS of the World Health Organization and the Global Fund itself. It was therefore inevitably in conflict with the position held by the Mbeki government. When central government realised that Durban would implement this project, it insisted – at the last minute before funds were released by the Global Fund – on taking full control of the budget and the content of the programme. As a result, the Japanese multinational and its partners withdrew for fear of becoming involved in a project dominated by the Mbeki government, and of consequently being held accountable for the foreseeable misuse of Global Fund money. This transnational governance network had therefore run into norm and power conflicts with the national government of South Africa, which resulted in the failure of the project and the entire network. In the end, no governance services were provided. Hence, it seems that transnational governance networks, in order to be efficacious, need to be legitimate in the eyes of the government. A clash of norms between the governance network and the government, such as occurred in this case, will otherwise become a power conflict and in the process destroy the network.
The second case – Mercedes Benz in East London – further illustrates the assertion that norm and power clashes with government agencies are decisive for the efficacy of governance networks. By 2004, the South African government had come under international and domestic pressure over its ignorance towards the disease and its failure to fight it effectively.Footnote 94 In particular, the civil society pressure group Treatment Action Campaign (TAC) organised broad public resistance against the Mbeki government's policy. The TAC and increasing international isolation forced Mbeki into an agreement imposed on him by his own party, under which he was required to abstain from any public debate on HIV/AIDS. The TAC also put pressure on the cabinet to draft a new government programme, which resulted in the Operational Plan for Comprehensive HIV and AIDS Care, Management and Treatment for South Africa. In addition, the Constitutional Court ruled that the government had to provide antiretroviral drugs to prevent the infection of newborns.Footnote 95 This ruling established the international scientific consensus towards the disease as the only acceptable approach in South Africa; therefore, when then US President Clinton offered a team of experts to help the country put together a national treatment plan, Mr Mbeki agreed, and a team was sent in 2004 to implement the Operational Plan.Footnote 96
Since then the South African government has accepted, and even supported, international help to fight HIV/AIDS. The case of Mercedes Benz in East London reflects this new context. Having rolled out its workplace programme successfully since the early 2000s, the firm decided in 2006 to organise a multi-stakeholder partnership. This partnership made the strategic decision to include the National Ministry of Health in order to achieve the buy-in of the national government. It also included local institutions, the DEG (Deutsche Entwicklungs und Investitionsgesellschaft – a German development agency) and the Border Kai Chamber of Commerce (BKCC).Footnote 97 The aim of the project was to develop and then diffuse HIV/AIDS workplace policies among small and medium-sized businesses, many of which were suppliers for Mercedes. More than 26,000 people have gained access to health care services through this supplier initiative alone, in addition to another 30,000 who have gained access in the context of the company's programme for its own workers and their families.Footnote 98 The project also offers training for medical personnel in local clinics in order to improve the public health services in relation to HIV/AIDS, it coordinates with sex education programmes in schools, and assists local communities affected by the disease in other ways.
This attempt to contribute to the provision of HIV/AIDS-related health care services has been relatively successful. Compared with the previous case, the precondition for this effective roll-out was that the South African government had moved from denial to a position closer to the international scientific consensus on which the Mercedes scheme was built, and that Mercedes made a conscious decision to actively involve the national government from the very start of the project. The two cases show the importance of the interactions of transnational governance networks with governments, and that norm and power conflicts are, positively and negatively, decisive for the efficacy of service delivery.
Governance networks are thus in need of meta-governance. Most networks so far have focused exclusively on their main mission: the provision of a specific service or the promotion of certain rights or standards. However, this sole focus on substantive policy goals may undermine the legitimacy, effectiveness and – as in the case of the network set up by the Japanese multinational in Durban – the very existence of governance networks, given that the norms on which they are based often clash with those held by local communities and governments. This raises the question of whether governance networks could or should be embedded or provided with secondary governance rules that stipulate a requirement for coordination with relevant stakeholders so that conflicts can be dealt with and resolved before they break out.
Apart from conflict resolution, such meta-governance rules could also help in facilitating solutions to problems of redistributional justice. When firms such as Mercedes Benz organise the provision of health services in East London, this is certainly to the advantage of those living there. However, in so doing, the firm is allocating resources – not just its own resources, but also those of the cooperation partners in the governance network. In the case of Mercedes Benz, several development agencies, local government and the national government dedicated organisational and financial support to the company's governance network. Resources of organisations are, however, limited. Firms will naturally plan to dedicate these limited resources to governance projects that help to improve their immediate business environment. If, for example, a firm's production and value chain depends on unusual skills and investment in training programmes which are threatened and undermined by HIV/AIDS, governance schemes will fight the disease in the workplace and the community in which the firm is located. The implication is, however, that locations and communities where firms cluster will receive a disproportionate share of governance services and resources. In more remote areas, far removed from industrial centres, business-driven governance networks are unlikely to emerge; yet, they may be needed. Transnational governance networks driven by business thus potentially create problems of distributional justice, which in turn point to the need for meta-governance rules to coordinate a just and fair distribution of the activities of different governance networks across a policy space. To summarise, transnational governance networks are in need of meta-governance – that is, rules that facilitate coordination and conflict resolution within (that is, between firms, local communities, governments and other stakeholders) and between different governance networks.
3.2. Fragmentation and Complexity
Transnational governance networks have proliferated enormously in the last decade and a half, particularly in the area of human rights. The emerging system of governance – that is, the evolving governance architecture – has rightly been described as ‘polycentric’,Footnote 99 fragmented and complex.Footnote 100 In some of the literature, this complexity is discussed positively, as a feature of the new transnational governance architecture that can enable rule-making and implementation to be more effective and legitimate than in the context of interstate ‘old governance’Footnote 101 regimes. ‘Loosely coupled’Footnote 102 networks, for example, may facilitate some form of ‘experimentalist governance’,Footnote 103 facilitate learning and establish a strong bottom-up, quasi-democratic element. Indeed, there is empirical evidence of experimental governance across fragmented networks of governance. However, this evidence consists of examples from the EU or the US – that is, from contexts of strong, consolidated statehood.Footnote 104 The state here may refrain from direct interference in the market, as in the context of the so-called ‘new modes’ of governance.Footnote 105 Yet it still has the resources, willingness, legitimacy and power to coordinate governance, and thus plays the role of a meta-governor. An example of this development concerns aviation regulation and security.Footnote 106 In the US and Europe, the state has largely delegated security control measures and related regulation to the airline carriers and their associations. At the same time, state agencies engage in extensive meta-governance to ensure that the private sector is really taking things into its own hands in a desired way and with the desired results. In consequence, European and US state agencies review the auditing practices of airlines, and other processes employed by airlines to guarantee air traffic security, in order to exercise some distant control. Beyond areas of consolidated statehood, however – that is, in areas of limited statehood – no such meta-governor exists. Therefore, systems of complex governance often confront a number of norm and power conflicts that undermine their capacity to govern, their legitimacy, and their very existence.
From the business entity's point of view, a complex, fragmented and polycentric governance order increases uncertainty and complexity – with the result that it raises transaction costs. However, it may also provide businesses with ample opportunities to influence the governance architecture so that it better serves their profit-maximising interests. In the previous section, I discussed examples of such business strategies. They involve firms that engage in governance with the aim of providing collective goods, and of firms as norm entrepreneurs. Mercedes, for example, or the Japanese multinational carmaker in Durban attempt to fight HIV/AIDS in the localities of their factories. Their motivation in this respect traces back to their reliance on asset-specific investments in employee skills that are threatened by the disease. However, as discussed earlier in the context of firms' behaviour in relation to regulatory standards,Footnote 107 not all firms can be assumed to have such benevolent preferences. In the absence of asset specificity and of a credible ‘shadow of hierarchy’, firms' preferences are often incompatible with international norms and standards. However, firms may still decide to become norm entrepreneurs – not in order to contribute to governance, but to strategically shape the emerging governance order according to their interests.
Strategies in this respect may involve playing off different governance networks and concerns against each other, forum shopping, regime shifting and the creation of strategic inconsistency.Footnote 108Playing off different governance networks and concerns against each other involves firms supporting activities in one governance task in order to delegitimise or undermine the efforts of another. Forum shopping occurs when firms strategically select governance networks in the hope that their choice will result in decisions or opportunities that further their interests. Regime shifting involves an actor moving a policy agenda from one governance network to another to render this policy agenda in the process. Strategic inconsistency refers to the intentional creation of contradictory rules or exploitation of contradictions between overlapping governance networks in order to weaken the effect of existing disadvantageous rules.
The following example illustrates these strategies.Footnote 109 It involves large multinational car companies in South Africa – BMW, Ford, General Motors, Mercedes Benz, Nissan, Toyota and VW – of which, as has been seen above, many have become actively engaged in the fight against HIV/AIDS. These are thus companies that have made significant contributions to the provision of governance services in the past; they are not the archetypal ‘rent-seekers’. Still, we find strategic norm entrepreneurship among these least likely cases for such behaviour. At one point, these firms decided to set up an environmental governance network, mainly in order to pursue anti-competitive strategies. South Africa is a high-price, high-quality car market, comparable with Europe, Japan and the US. However, the distribution of income is far more disparate in South Africa than it is in these markets.Footnote 110 High rates of unemployment, mass poverty and a relatively small middle class population result in a mismatch between demand and supply.Footnote 111 Since 2006 car manufacturers from countries such as China, India and South America that produce low-price cars have been showing an increasing interest in entering the South African market to supply the demand there for cheap cars.Footnote 112 Chinese companies, Chery and Brilliance, for example, signed contracts with South African car sales dealers to gain access to the South African market. This posed a significant threat to the established South African mass producers, such as Ford, General Motors, Nissan-Renault and VW.Footnote 113 Market entry for cheap car producers originating from emerging markets would make it far more difficult for them to sell their relatively costly cars to the average South African consumer.
The mass-market producers – Ford, General Motors, Nissan and VW – therefore, under the umbrella of the National Association of Auto Manufacturers of South Africa (NAAMSA), actively lobbied the South African government to issue stricter emissions regulations for newly registered vehicles. They did so also by forming an alliance with a number of environmental NGOs in South Africa.Footnote 114 The firms and the NGOs agreed on a division of labour: the firms would lobby the government behind closed doors, but otherwise refrain from making public comments in support of higher emissions standards. The network feared that public industry support of higher emissions standards would raise suspicions of a hidden economic agenda. The NGOs, in turn, would advocate the need for higher emissions standards in public. They would do so by emphasising air quality, which indeed would improve with higher emissions standards. The governance network was successful. In response to pressure from NAAMSA and advocacy by the NGOs, the South African government raised emissions standards to EURO 3 norm emissions control level in 2004Footnote 115 and announced that it would raise the standard to EURO 4 level within the next two years. These stricter regulatory requirements had the effect of preventing the Chinese, Indian and South American competitors from entering the South African market. Their cars observe lower emissions standards and cannot easily be upgraded.Footnote 116
Arguably, the net contribution to environmental governance of this particular governance network is positive, irrespective of the hidden agenda of keeping competitors out. The described activities raised the level of emissions standards in South Africa, although it involved a trade-off. Mobility is a social issue of great importance in South Africa; lack of mobility is one of the main obstacles that prevent township populations from participating in the formal economy. It is thus a cause of poverty and lack of opportunities in the country. Empowering global norms, such as environmental sustainability in this example, thus resulted in the neglect of others: alleviation of poverty and the right to economic development. The problem here is of a normative nature: it seems difficult to justify the imposition of such collectively relevant trade-offs on a society when they are the result of clandestine strategies of business-driven governance networks that represent particular interests. Overdevest and Zeitlin argue that in experimentalist governance in the EU ‘units must report regularly on their performance and participate in a peer review in which their results are compared’Footnote 117 and assessed. Thus, in the European context, the EU institutions deal with such trade-offs, and have the power to tilt them in a direction that is deemed desirable. Hence, here we have a meta-governor who can and does supervise and govern such governance conflicts with some degree of legitimacy. In areas of limited statehood such as South Africa, however, it is unclear from the outset who may legitimately supply this demand for institutional procedures of meta-governance. The example once again illustrates the need for meta-governance. It also shows that, apart from functions relating to coordination and conflict resolution, meta-governance should also fulfil supervisory functions in order to identify clandestine politics, imposed trade-offs and hidden agendas by business or other actors, and to produce legitimate decisions about the sanctioning or rewarding thereof.
4. Conclusion
In this discussion of empirical research on business and human rights in relation to transnational governance networks I have argued the case for meta-governance. Transnational governance networks are in need of meta-governance for at least three reasons. First, their norm conflicts with local communities and governments need to be resolved; otherwise, transnational governance networks cannot deliver governance services efficaciously. This is particularly so when governance tasks are complex, involving the continuous interaction of various actors. Second, the activities of transnational governance networks need to be coordinated by means of meta-governance to create a system of distributional justice in the delivery of governance service; otherwise, the current governance architecture will lead to yet more and deeper divisions between the ‘haves’ and ‘have-nots’ in developing countries. Third, transnational governance networks are in need of meta-governance in order not to fall prey to the opportunistic strategies that business actors often pursue in their context. Playing off, forum shopping, creating inconsistencies, and other strategies by businesses may weaken transnational governance networks and the emergent governance order. Institutions of meta-governance thus need to supervise business strategies, rewarding desired and sanctioning undesired strategies.
In areas of strong, consolidated statehood such as the EU and the US, state institutions can play the role of meta-governor. However, here also their capability as meta-governors is limited in some respects. Many of today's problems, such as climate change, are global in nature and thus need coordinated action on a scale beyond the jurisdiction of these states and entities. In areas of limited statehood, on which I have focused in this article, state institutions, when left alone, are usually incapable of fulfilling meta-governance functions. It is in these areas of limited statehood – and, potentially, at the global levelFootnote 118 – that the need for meta-governance is strongest. Yet, it seems entirely unknown to what extent meta-governance currently exists here, and of what nature, form and substance such meta-governance can and should legitimately take.
This article therefore ends with a call for a new research agenda to explore these questions. This agenda should involve, on the one hand, empirical research that explores and analyses examples of meta-governance. To what extent does meta-governance already exist in areas of limited statehood? Who does it, and why? Under what conditions is it considered legitimate? How effective are these examples of meta-governance, and under what conditions are they more so and less effective? On the other hand, this agenda should consist of legal-normative scholarship. How should meta-governance be organised, and on the basis of which norms and considerations? To whom can, and should the task of meta-governance be delegated? How should meta-governance be considered in relation to conventional international law?
An example may illustrate how this agenda could be achieved. BMW in South Africa, like most multinational car companies operating there, runs an HIV/AIDS programme that offers comprehensive health services to its employees. This programme is more successful than most other HIV/AIDS workplace schemes in the country and, as a result, the company has managed to drastically reduce absenteeism and losses.Footnote 119 One reason, according to the project manager and local NGOs, is that the firm realised in the past that acceptance of the programme in the local communities where its employees live is the key for its efficacy. When the programme began only few employees were willing to enrol in it, which made the programme rather ineffective. It was then that the programme manager recognised the extent to which the firm's health activities clashed with local community norms and power structures. In particular, local healers in the townships saw the company programme as a potential threat to their own authority, power and economic well-being in the township economy and society. Most employees, however, routinely consulted these local healers. The advice they received from the healers when the scheme was first rolled out seemed to have discouraged them from enrolling in it. To rectify this, the manager of the BMW HIV/AIDS programme invited the local healers to the company headquarters to coordinate the firm's programme, which is based on Western medicine, with the advice that traditional healers give to employees and their families. She offered the healers a deal. The manager would in the future routinely refer employees to the healers for nutritional and general lifestyle advice. This offer signalled the acknowledgement of, and respect for the social status and work of the healers, and guaranteed that they would not lose business. In return, the manager asked the healers to encourage employees to enrol in the company programme for medical advice, and to refrain from nutritional and lifestyle advice that could undermine or contradict the medical treatment provided by the firm. Through this move, she managed to secure the buy-in of most traditional healers to the company programme, so that today they recommend that employees enrol in the programme, thus making the company's fight against the disease highly successful.
This example illustrates meta-governance in practice. The manager's efforts to coordinate with local healers helped her to mitigate the conflict between the company's HIV/AIDS programme with local community structures, which initially made the company's efforts to fight the disease unsuccessful. However, it also raises a number of questions. Why did this company engage in the described meta-governance, when others did not? How can interaction between business entities and local communities such as the township healers be organised and structured – and potential conflicts, which are likely to emerge in the course of interaction, be resolved? In a comparative perspective, is BMW really more successful than other firms in its attempt to fight HIV/AIDS as a result of its coordination efforts with traditional healers? Can this outreach and coordination with healers be justified normatively? This is a particularly important question, as coordination may involve the striking of deals and compromise – a potential deviation from a pure Western medicine-informed approach to the disease. Can and should firms deviate from the ‘Western’ protocol regarding HIV/AIDS, even though it is clearly the responsible medical approach? If so, in which areas and when is it acceptable for firms to strike deals and compromise – and in which areas is this not acceptable? These are fascinating questions, involving not only empirically interesting political science research, but also very fundamental normative-ethical legal questions. Addressing them together may allow legal scholars and political scientists to make a real contribution to the future governance architecture.
