Commentary: The European Parliament’s Draft Directive on Corporate Due Diligence and Corporate Accountability

Please note that the article reflects solely the views and opinions of the undersigned scholars. The ECGI does not, consistent with its constitutional purpose, have a view or opinion

 The European Parliament’s Draft Directive on Corporate Due Diligence and Corporate Accountability

signed by The European Company Law Experts Group*

1. What does the draft Directive require?

1. On March 10 the European Parliament adopted a resolution setting out a draft Directive on Corporate Due Diligence and Corporate Accountability[1] “aimed at ensuring that undertakings . . . fulfil their duty to respect human rights, the environment and good governance” (Art.1(1)). In order to achieve this aim, the Directive imposes on those undertakings within its scope (ie principally but not only companies) a due diligence duty to “take all proportionate and commensurate measures and make efforts within their means to prevent adverse impacts on human rights, the environment and good governance from occurring in their value chains, and to properly address such adverse impacts when they occur.” (Art.1(2)). The central output of this duty is to be the production by the company, in consultation with stakeholders, of a due diligence strategy, its implementation and its periodic revision, also in consultation with stakeholders (Arts.4 and 8). The due diligence exercise has to be carried out not only in relation to the undertaking’s own operations nor even just along its supply chain but along its whole “value” chain. This means “entities with which the undertaking has a direct or indirect business relationship upstream and downstream, and which either: (a) supply products, parts of products or services that contribute to the undertaking’s own products or services, or (b) receive products or services from the undertaking . . .” (Art.3(5)). So, customers as well as suppliers.

2. It is likely that substantial parts of the value chain (both in terms of suppliers and customers) will be located outside the EU. Indeed, it is clear that a major aim of the Directive is to change the human rights, environmental and good governance practices of businesses operating outside the EU. This is to be achieved by imposing legal obligations on companies within the EU to “leverage” their relations with customers and suppliers to this end. The Directive expects to maximise the external impact of the Directive by applying its requirements not only to companies incorporated within the EU, but also to non-EU companies when they operate in the internal market selling goods or providing services (Recital 12 and Art.1(1)). An important, but unresolved, issue in the draft is the identification of the national authority which will be expected to supervise the non-EU company (as discussed below) under the Directive. The exclusive allocation of such a function to a national authority is particularly problematic in the case of non-EU companies operating in a large part of the EU territory.

3. The strategy generation process is subject to various forms of supervision and breaches of the Directive’s obligations in relation to the strategy are sanctioned or remedied in various ways, discussed below. It is conceivable, of course, that a company might be involved in breaches of the standards, despite having drawn up and implemented a due diligence strategy. The strategy is essentially an exercise in risk management. Despite precautions, risks may still eventuate or the strategy may have failed to identify all the risks. So, in a second step, the Directive imposes on the company liability for actual breaches of its standards independently of the due diligence strategy (subject in the case of civil liability to a due care defence). Thus, one can categorise the draft’s approach to the human rights etc standards as being in part indirect (the standards are made binding on the company by shaping the legally enforceable strategy which results from the due diligence exercise and which the company is obliged to implement) and partly direct (liability for beaches of the standards independent of the strategy).

4. (a) Supervision and sanctions in relation to the strategy. In the production, implementation and revision of the strategy companies are to be supervised by the appropriate national authority (Art.12). The national authorities have the power to investigate whether the company has complied with its obligations under the Directive (Art.13). Besides acting of its own motion it may act on information provided by third parties. To that end, the supervisory authority “shall facilitate the submission by third parties of substantiated and reasonable concerns”, including submissions made on a confidential or anonymous basis. The authority may order the company to take remedial measures and fine those which do not take them. Where the authority concludes that failure to comply with the Directive “could directly lead to irreparable harm”, it may impose remedial measures itself or order a temporary cessation of business of the company. In the case of companies incorporated outside the EU but operating in the internal market, the temporary suspension of activities may imply a ban on operating in the internal market. This provision clearly shows that it is contemplated that the authority will have power to investigate the adequacy of the company’s strategy, for example, where it is argued that the strategy has failed to identify a certain category of risk or that it has dealt with it ineffectively.

5. Independently of the investigation power, Member States must provide dissuasive but proportionate sanctions for breaches of the Directive’s obligations, including, therefore, those related to drawing up, revising and implementing the strategy (Art.18). It is clear that a powerful set of sanctions is in contemplation, for the national authorities must be given the power “in particular [to] impose proportionate fines calculated on the basis of an undertaking’s turnover, temporarily or indefinitely exclude undertakings from public procurement, from state aid, from public support schemes including schemes relying on Export Credit Agencies and loans, resort to the seizure of commodities and other appropriate administrative sanctions.” However, a complete cessation of activities is a power available to the public authorities only after an investigation.

6. In addition, companies must themselves provide a grievance mechanism “allowing any stakeholder to voice reasonable concerns regarding the existence of a potential or actual adverse impact on human rights, the environment or good governance [Art. 9].” The grievance mechanism is drawn from United Nations Guiding Principles on Business and Human Rights, where, however, it operates as a substitute for state supervision or court redress. Here, it is imposed in addition to these other mechanisms. The terminology of Art.9 would seem wide enough to cover not only complaints about the construction, implementation and revision of the strategy, but also allegations about the company’s involvement in breaches of the standards which are not dealt with in the strategy. The sanctions on the company under the grievance mechanism are mainly reputational, ie publicity for the grievance and the company’s response. However, Art.9 provides in addition that the “grievance mechanisms shall be entitled to make proposals to the undertaking on how potential or actual adverse impacts may be addressed.” Since, as a matter of language and logic, a mechanism cannot make a proposal, but a proposal may result from a mechanism, this provision may suggest that the grievance mechanism must contain an independent element.

7. (b) Sanctions for “adverse impacts” There are two provisions sanctioning “adverse impacts” whether or not they arise out of a failure by the company to abide by its strategy. First, Art.19 requires Member States to have in place a regime of civil liability under which companies are to “be held liable and provide remediation for any harm arising out of potential or actual adverse impacts” on the human rights etc standards where they have “caused or contributed to” those harms. Second, under Art. 10 Member States are to require companies to provide or cooperate with an extra-judicial remediation process (whose contours are not defined) where the company itself identifies that it has caused or contributed to an (actual) adverse impact or to cooperate with that process where the company identifies that it is “directly linked” to such adverse impact. “Adverse impact” is defined for each of the areas covered by the Directive in Art.3(6) to (8) by reference to a list of international and European standards to be set out in Annexes to the Directive (and so not specified in the draft). Whatever ends up in the Annexes thus becomes directly binding on companies, by virtue of these two articles. For example, in relation to environmental standards Art.3(7) states that “‘potential or actual adverse impact on the environment’ means any violation of internationally recognised and Union environmental standards.” This approach is in contrast with the due diligence obligation where the standards become binding on the company indirectly, ie after the company and stakeholders have considered how they apply to the company’s operations and embodied them in a strategy.

2. Critique

8. It goes without saying that we think companies should respect international standards in the areas covered by the proposed Directive when these are relevant to their business operations. We take the view that most businesses think so as well. We are also in favour of an appropriate mechanism which puts a legally binding obligation on companies to do so. We think that this step might have the advantage of pushing human rights, environmental and good governance standards up the corporate agenda when complex decisions have to be taken involving many competing considerations. However, we think the Directive as envisaged by the European Parliament is not an appropriate instrument to this end and, in fact, is likely to be counter-productive. We reach this conclusion on the basis of the interaction between two principal features of the draft: the wide range of imprecise standards which are to be applied to companies and the highly constrained context in which companies decide how those standards are relevant to their businesses.

9. As we explain below, although the standards to which the Directive refers are still to be specified in Annexes which are empty in the draft, the Recitals to the draft make it clear that what its authors have in mind are international and regional standards in the areas of human rights, the environment and good governance which were originally negotiated between states. We think that far too little thought has been given to the question of how these open-ended inter-state standards are to be made workable within a business context, if they are to be made legally binding, directly or indirectly, on companies. In fact, the difficulties which are generated by this simple transposition of standards from one context to another have been exacerbated by two further features of the Directive. The first is the sheer range of international instruments which are to be made legally binding on companies. The second is the highly supervised context in which companies will have to adapt to these standards to their particular business operations. As noted above, oversight of managements’ responses to the standards is envisaged in the draft in three different ways, by the public authorities, a wide range of stakeholder groups and, ex post, the courts.

10. Pressure on companies to respond positively to international standards has been growing for some time. The OECD Guidelines for Multinational Enterprises were first adopted in 1976 and have been up-dated subsequently on a number of occasions; the UN’s Guidelines on Business and Human Rights date from 2011. However, once this proposal for a Directive enters into force a Rubicon will be crossed. The OECD and UN instruments are guidelines, as is made clear in their titles. Companies, therefore, have significant flexibility in interpreting the standards contained in them and in applying them to their own particular circumstances. Moreover, there are no direct legal sanctions attached to breaches of the guidelines, though the OECD Guidelines contain a dispute resolution machinery (via “National Contact Points”) of a consensual and non-binding nature, which might nevertheless result in reputational damage being incurred by the company if its case appeared weak. By contrast, the proposed EP Directive changes the legal position of companies in relation to the international standards by transposing them, directly or indirectly (via the required strategy) into hard law. This step imposes a responsibility on the law-maker to produce rules which those subject to them (principally corporate management but not only them) can make operational in relation to their specific activities at reasonable cost and with a high degree of certainty that the changes in corporate policies and behaviour which management introduces will satisfy the requirements of the new legal rules.

11. Our view is that the EP draft fails to meet this elementary requirement for good legislation. We focus below on three main points.

1. The failure of the draft to particularise the standards with which companies will be expected to comply, which are, as the draft stands, multifarious and highly imprecise in some cases.
2. The ambiguity in the draft about what a company has to do to comply with the standards in its business. The range of actions required of the company is potentially very wide and this creates further uncertainty for the company and others as to how it should conduct itself.
3. The wide range of challenges which may be mounted against the company where non-compliance is alleged. Uncertainty (for everyone) about what the draft requires by way of corporate behaviour coupled with a wide range of challenges is likely to result in the company devoting considerable time and resources to dealing with these challenges, which may not, individually or cumulatively, in the end generate clarity about what the rules require. Even worse, some of these challenges may be mounted by groups whose principal aim is to advance reform agendas rather than to secure corporate compliance with existing standards. We deal with point (c) in the course of discussing points (a) and (b) below.

12. In Part IV we propose a way of rescuing the Directive from these faults. If it is considered desirable to derive the standards to which companies must adhere from international standards in the area of human rights etc, we think companies must be granted much greater autonomy when making decisions about how these standards apply to the company. Applying standards, even open-ended ones, is not in itself an impossible task for management. The difficulty with the draft is that it essentially requires that the application of the standards to the company is something which the company should negotiate with (a wide range of) stakeholders and with the public authorities. We think this is likely to be a chronically contentious and counter-productive process. We think that better results are likely to flow if, instead of a negotiated result, corporate management took ownership of the process of translating the standards into corporate policies. In Part IV we propose to this end an approach based on corporate good faith, which would be supported by a reduced, but not completely eliminated, role for the public authorities and the courts. Only in this way is it likely that legal certainty will be obtained, which, as the draft recognises in its Recital 13, is the main gain business expects to obtain from this instrument. We think that the current draft will not create certainty and is more likely to generate confusion and conflict about what is required. At this stage, our proposals in Part IV are simply indications, not a fully worked out draft, but we hope it may prove useful to those who have to consider the EP’s version in the future.

III. Detailed critique of the draft

13. We flesh out below the summary critique developed above. There are many other issues that could be raised in relation to the draft (for example, the range of undertakings to which it should be applied) but we concentrate on the elements central to our critique that the core exercise of transferring standards from the international arena and making them binding in the commercial context has not be thought through. There are two central elements in the draft: the first is a requirement on companies to formulate a due diligence strategy (Art.4); the second is exposure of companies to liability to provide “remediation for any harm arising out of potential or actual adverse impacts” arising out of breaches of the international standards (Art.19).  

(a) Multifarious and open-ended standards

14. The draft Directive covers the three areas of human rights, the environment and “good” governance, three large and disparate areas. Human rights is a term given a broad interpretation in the draft so as to include “social, worker and trade union rights” (Art.3(6)) – an extension likely to be of particular significance to companies. “Good governance” refers, apparently, not to governance of the corporation, but to the governance of states in which a company may invest and carry on operations. Thus, Recital 24 of the Draft says that the good governance standards to which companies should adhere should include Chapter VII (Combatting Bribery, Bribe Solicitation and Extortion) of the OECD Guidelines for Multinational Enterprises and the principles of the OECD Convention on Combating Bribery of Foreign Public Officials in International Business Transactions. The reference to the OECD instruments in particular indicates that the Directive, like those instruments, aims at corporate activities in states with weak laws in the three areas covered by the Directive and/or weak enforcement of those laws. Although not without application within the EU itself (for example, in relation to “modern slavery”), it seems likely that the principal impact of the Directive, if adopted, will be in countries outside the EU which have low standards in the areas covered by it.

15. Unhelpfully, the draft does not specify in its text which international or regional instruments in the three fields are the applicable ones, but the recitals indicate that it will be a large number (Recitals 21, 23 and 24). This is particularly true in the human rights area. Recital 21 refers specifically or generally to some 20 international standards, some of which incorporate further sets of standards. For example, the Recital refers to the International Bill of Human Rights which in turn states that it “consists of the Universal Declaration of Human Rights, the International Covenant on Economic, Social and Cultural Rights, and the International Covenant on Civil and Political Rights and its two Optional Protocols.”  As to the environmental harms, Recitals 22 and 23 specify UN standards where these harms are connected to human rights, but otherwise list a long range of topics which the company must evaluate when setting its strategy: “impacts should include, but should not be limited to, production of waste, diffuse pollution and greenhouse emissions that lead to a global warming of more than 1,5°C above pre-industrial levels, deforestation, and any other impact on the climate, air, soil and water quality, the sustainable use of natural resources, biodiversity and ecosystems.” [2]

16. Recital 4 concludes: “Undertakings thus currently have at their disposal an important number of international due diligence instruments that can help them fulfil their responsibility to respect human rights.” This is a somewhat disingenuous statement. Companies and other interested parties will have to pore over a long list of documents in order to work out the overall requirements to which they are subject and to deal with potentially conflicting obligations. The “important number” is likely to generate confusion and controversy rather than act as an aid to the setting of a strategy or the conduct of corporate affairs.  

17. It is therefore surprising that the operative text of the draft does not specify which international instruments are the relevant ones in relation to the legal obligations it creates. It simply refers to Annexes to the Directive, currently empty, to be filled in, apparently, by the Commission. Given the centrality of the standards in the structure of the Directive, it needs to be made clear in the EU legislative process (ie before the Directive is adopted) which standards are to be made binding and the list of applicable standards needs to be reduced to manageable proportions. Since this is an EU Directive, it might be sensible to apply, where available, European standards, for example, in the human rights area the European Convention on Human Rights, not the long list of human rights standards currently mentioned.

18. Even if the number of relevant standards is reduced, a more significant problem will remain. This is the open-endedness of many of the standards set out in the international instruments. This is particularly true of human rights standards. In the “governance” area there exist workable definitions of bribery and corruption, but it is in this area the EU initiative has least value. Member States’ laws and the laws of third countries to a significant extent have already put in place rules on these matters, and normally those rules apply to bribery and corruption by companies outside the enacting jurisdiction. In the human rights area, however, the standards are often open-ended. These instruments were originally negotiated at inter-state level and so their imprecision is perhaps understandable: the standards needed to be acceptable to the widest array of states. But imprecision becomes a defect, once hard-law obligations are attached to the standards. For example, both the International Covenant on Economic Social and Cultural Rights and the European Social Charter require “fair remuneration” and “just conditions of work” (Arts 7 of the former and 2 and 4 of the latter). The same precepts are contained in Art. 23 of the Universal Declaration of Human Rights. What this means within the EU is probably clear enough, but its meaning across non-EU parts of the “value chain” is likely to be highly contentious. Equally, with other open-ended obligations in international standards.

19. Open-ended provisions will cause serious difficulties for companies when they engage in the process of drawing up a due diligence strategy and implementing it thereafter, even if the company were free to determine by itself what the standards require. But, as we have seen, it is not. Companies must “carry out in good faith effective, meaningful and informed discussions with relevant stakeholders when establishing and implementing their due diligence strategy” (Art 5.1). Stakeholders are a widely defined group extending to all those who may be affected by actual or potential breaches of the international standards and so “can include workers and their representatives, local communities, children, indigenous peoples, citizens’ associations, trade unions, civil society organisations and the undertakings’ shareholders” (Art 3(1)). It is unlikely that these groups will all agree on the meaning for the company of open-ended standards. They are likely to be to some degree in conflict with each other and with the management of the company. Although it is presumably up to the company to decide ultimately on the contents of the strategy, the required consultation is likely to be a long-winded and fractious business, in which the company loses the goodwill of at least some of the groups involved and fails to obtain a solid platform on which it can conduct its business.

20. Moreover, the conflict of views will not be put to an end when the strategy is adopted. The consultation obligations noted above continue to apply as the strategy is implemented, so that continuing disagreement and conflict among interest groups may be anticipated. In addition, since the strategy needs to be reviewed at least annually, again in consultation with stakeholders, and possibly revised (Art 8), a formal opportunity for interaction with stakeholders is created. Dissension and disagreement may therefore become continuing elements in this area of the company’s operations.

21. For those dissatisfied with the results of the consultation process, whether initially, during implementation or during revision, the draft provides two ways in which they may pursue their concerns about the company’s strategy further, ie through the grievance mechanism noted in para 6 or by seeking to persuade the competent national authority to carry out an investigation (para 4).  In addition, the national competent authority may take a different view of what is required by way of effective strategy from that arrived at by the management and require the company to make alterations to it. Finally, outside the strategy, the company may be subject to civil litigation for acting in breach of the standards or be required to engage in a remediation process (see para 7). This will provide further opportunities for friction and disagreement over the meaning and applicability of the standards.

22. There is one general provision in the draft which aims to mitigate the frictions likely to arise when setting the strategy. This is contained in Art.14: “In order to create clarity and certainty for undertakings, as well as to ensure consistency among their practices, the Commission . . .  shall publish general non-binding guidelines for undertakings on how best to fulfil the due diligence obligations set out in this Directive.” However, this provision is likely to be ineffective in producing clarity and certainty. First, when producing the guidelines the Commission must take “due account” of eleven sets of international standards set out in Art. 14.3. Consequently, the guidelines are unlikely to simplify the task facing those interested in drawing up the corporate strategy, since the guidelines, presumably, will have to cover the principal obligations contained in each of the sets of standards. Second, even if the Commission were to decide to take a bold tack and confine the guidelines to a small number of core provisions, the non-binding nature of the guidelines would defeat this strategy. Being non-binding the company is not obliged to comply with the guidelines, but equally those seeking to challenge the company’s approach to its strategy would not be bound by them either. They could still seek to argue, in the company’s consultation, in the grievance machinery, before the national authority or in litigation that the applicable international standards required more of companies than the guidelines recognised. In other words, compliance with the guidelines provides no safe harbour for the company. The guidelines are thus likely to exacerbate the difficulties of reaching agreement on a strategy, because yet another text has to be considered, whilst if the guidelines simply repeat what is in the international standards, no increase in “clarity and certainty” is to be expected.

23. (b) The scope of the compliance obligation

Assuming for the sake of argument that the company has extracted from the many international instruments potentially applicable the obligations which are relevant to its operations and has come to a view about the proper interpretation of those obligations, the next question is to establish what the company has to do to comply with those obligations as identified. This involves three separable issues: (i) the range of business relationships which are relevant; (ii) the extent of the risks which must be addressed; and (iii) assessing the available reactions.

24. (i) The range of relevant business relationships As noted above, the due diligence exercise has to be carried out, not only in relation to the company’s or group’s own operations but across the whole “value chain”, of which it is a part. The “value chain” is defined (Art. 3(5)) in the following terms:

‘value chain’ means all activities, operations, business relationships and investment chains of an undertaking and includes entities with which the undertaking has a direct or indirect business relationship, upstream and downstream, and which either: (a) supply products, parts of products or services that contribute to the undertaking’s own products or services, or (b) receive products or services from the undertaking . . .”

Thus, the company’s strategy must take into account the potential for infringements by those who supply inputs to it or who are consumers of its outputs. The terms “indirect business relationship” indicates, moreover, that the value chain extends beyond a contractual relationship with a group company: suppliers to suppliers to the company or customers of customers of the company are included as well – but it is not at all clear when a business relationship passes beyond being indirect and ceases to be a relationship at all. The definition of a “business relationship” in Art 3(2) does require that it be “directly linked to the undertaking’s business operations, products or services”, but how this definition squares with the inclusion of an “indirect” business relationships in Art.3(5) is wholly unclear. In any event, “directness” is a term which carries with it a considerable penumbra of uncertainty.

25. (ii) The linkage between the value chain and the risks to be addressed Without the extensions mentioned in the previous paragraph, it is clear that the draft would be open to evasion by the simple technique of companies contracting out work to non-group companies and, perhaps, by further contracting out by those non-group companies. However, the extensions create a need for clear rules about the linkages between the extended chain of business relationships and the risks of harms which the company needs to cover in its strategy and for which it is potentially liable. This matter is dealt with in Art.4 which lacks specification and potentially has an excessively broad scope.

26. Art 4.2 deals with the linkages between the company’s operations and/or those of its related businesses, on the one hand, and the harms to human rights, the environment and good governance, on the other, in broad terms. It and Art.4.4 are poorly drafted.[3]  It appears that the strategy must cover situations where the operations of businesses within the value chain “cause or contribute to or are directly linked to” potential or actual adverse impacts arising out of breaches of the international standards (Art.4.2). Causation is an established legal notion, though often difficult to apply in practice. “Contribution” is defined (Art.3(10)) as including situations where “the activities of the undertaking cause, facilitate or incentivise another entity to cause an adverse impact.” So, it is necessary for the company to address the risk of harms in fact caused by some entity outside the value chain, but where those harms are facilitated or incentivised by an entity within the value chain. This formulation is very broad. For example, it appears to create risks for any supplier of goods or services to any non-democratic government, on the basis that such supplies “facilitate” breaches of human rights standards on the part of the government (see, for example, Art 21 of the UN Universal Declaration of Human Rights which states that “everyone has the right to take part in the government of his country, directly or through freely chosen representatives.") Would supplying fast cars to that country’s police force be included, even when at the time of supply there was no civil unrest in the country? In an economically small, non-democratic country, even a company which does not supply to the country but exports from it (for example, oil or minerals) might be regarded as contributing to non-compliance with Art.21 if the royalties and other taxes payable constitute a major part of the state’s revenues. The coda to Art 3(10)) provides the comfort that creating the “general conditions” in which the adverse impact may occur does not fall within the notion of contribution and the contribution to the risk of the harm occurring must be substantial. Once again, there is scope for much contention about where the line lies between “general conditions” and facilitating breaches of the standards, a matter on which stakeholder groups are unlikely to agree with management or among themselves. Pro-democracy civil society groups are likely to take a strong view on not dealing with such a country; management and employee groups probably a less strong view.

27. The notion of “direct linking” – the third connecting factor mentioned in Art.4(2) - is not defined but logically must go beyond “causing” and “contributing”, for otherwise the words “direct linking” would need not to be included. “Direct linking” appears to open up the possibility that the strategy must cover situations where the company does not cause the harm, whether directly or indirectly via another entity in the value chain. In other words, the strategy will have to cover situations where there is no causal link between the operations of the company or its contractors and the harm caused, but there is a “direct linkage”. The language here is opaque and the comfort contained in Art.3(10), mentioned in relation to “contributing”, does not apply to direct linking. The scope of this connecting factor needs to be made much clearer or this connecting factor should be eliminated.

28. Art.4(2) goes on to state that companies “shall in an ongoing manner make all efforts within their means to identify and assess by means of a risk based monitoring methodology that takes into account the likelihood, severity and urgency of potential or actual impacts” the risks which the value chain causes, contributes to or is directly linked to. The qualification that identification and assessment efforts need to be related to the company’s means is important, though, again, not precise. However, the standard for assessment of an identified risk is not specified in Art. 4(2). This is a major omission and a further issue on which those involved in setting the strategy are likely to take very different views. Except for small companies and those operating wholly within the EU, Art.4(4)(i) requires specification in the strategy (which is a public document) of the risks that “are likely to be present in its operations and business relationships, and the level of their severity, likelihood and urgency and the relevant data, information and methodology that led to these conclusions.” If this requires the specification of all risks which have been identified, including those assessed as being low-level, this is a burdensome obligation which has little legitimate purpose.

29. (iii) Required responses The risks having been identified and assessed, the strategy has to set out the company’s proposed response to them. Art.4.4(iii) requires that the undertakings through the strategy “adopt and indicate all proportionate and commensurate policies and measures with a view to ceasing, preventing or mitigating potential or actual adverse impacts.” Art.4.6 requires that the company ensures that its own policies are in line with the strategy and Arts.4.8 and 4.9 that they ensure that their business partners do likewise. It is clear that not all risks are expected to generate the same level of response by the company. Art.4.4(iii) limits itself to “proportionate and commensurate” measures. Art.4.4(iv) recognises that the company may need to prioritise some risks over others “in the event that they are not in a position to deal with all the potential or actual adverse impacts at the same time”, though the cross-reference to Principle 17 of the UN Guiding Principles on Business and Human Rights in fact gives little guidance on how the prioritisation is to be carried out. As to other entities in the value chain, companies “shall carry out value chain due diligence which is proportionate and commensurate to the likelihood and severity of their potential or actual adverse impacts and their specific circumstances.”

30. Thus, the draft rightly recognises that the appropriate responses to the identified risks will depend heavily on the context in which the company and its business partners operate and cannot be specified in advance in the legislation. At the same time, this approach creates a further area for dispute and disagreement among the managers and stakeholders involved in setting the strategy.

IV        Making the draft workable

31. The conclusion from the above analysis is that the draft imposes a set of obligations on companies which would be regarded as burdensome in any context. It might be sensible to impose them if the conduct required of companies were clear and the benefit to society substantial. An example of this type might be the rules against money-laundering imposed on banks and other financial institutions. Here, the mischief the rules are aimed at (money-laundering) is fairly precisely defined, even though the rules have significant consequences for how banks are expected to operate in practice. In the case of the areas covered by the draft, the harms the international standards are aimed at are, in many cases, of a very open-ended character. The difficulties which flow from the imprecision of the legal obligations which companies are expected to observe are compounded by the range of parties to be involved in the process of transposing the standards into corporate policies and by the large number of supervision and complaint processes available to those who think their views have not been properly reflected in those policies international instruments companies are expected to observe.

32. It is true that making the international standards workable for companies was not an insurmountable problem so long as they were used only as guidelines. If they are to be embodied in hard law, however, with significant sanctions attached to them, companies and those affected by their operations deserve and need to be presented with a set of obligations which are capable of being implemented with some confidence that the corporate responses will be in compliance with the law. We do not believe that the current draft gives that assurance to companies and others involved in the due diligence process. Indeed, the draft maximises the potential for companies to be in continuing conflict with civil society groups and worker representatives about what the new provisions require.

33. We might stop at this point. We do not do so because we believe that making companies sensitive to the human rights, environmental and good governance implications of their activities is a worthwhile enterprise. We believe that the majority of companies, both in the EU and the wider OECD, already share this view. The surveys show that business is supportive of initiatives such as that embodied in the draft. But it is clear that the main benefit business hopes to obtain from the extra regulation is clarity about what is required by way of compliance. For the reasons given above, we do not believe that an acceptable level of legal certainty, whether for companies or anyone else, will emerge from this draft. So, we turn to sketch out a programme for maintaining, but at the same time making workable, a law requiring corporate compliance with international standards in the areas of human rights, environmental standards and “good governance”.

34. We have suggested above that the range of international standards referred to in the Directive might be slimmed down, by concentrating on European instruments. Within the instruments then remaining, there might be scope for excluding provisions which should be regarded as aimed mainly at the conduct of states rather than at commercial enterprises. We think this is an important first step, to avoid contradictory obligations and to facilitate identification of the standards companies should focus on. However, we do not believe the underlying problem is solvable completely through this technique. International standards in the human rights area in particular, but also elsewhere are often expressed in general language, and for good reasons, given, for example, the range of ways in which human rights may be abused. The same is probably true, though to a lesser degree, in relation to environmental and “good governance” standards.

35. We therefore turn to our proposals for amendments to the draft. Our central proposal is that the company’s obligation to assess the harms which the international standards refer to and to draw up a strategy for eliminating or reducing them should be a good faith obligation. In addition, we propose that the provisions imposing direct liability on companies for breaches of the international standards (ie not via the strategy the company adopts) should be removed from the Directive.

36.  (a) A good faith requirement to produce a strategy and revise it

Under our first amendment, the company would be obliged to consider in good faith how the applicable international standards relate to its business (along the whole of the value chain) and then to produce and revise from time to time, again in good faith, a plan for addressing the actual or potential harms identified. Once produced, however, hard law obligations would apply to the strategy. Thus, the company would be subject to sanctions for failing to abide by its own strategy, but not for failing to abide by the international standards themselves. This would put the burden on the company to fashion a workable strategy whereby, in its particular business context, it and its business partners could comply with the international standards contained in the Directive.

37.  The obvious criticism which could be advanced in relation to this suggestion is that a mere good faith obligation will induce companies to produce unambitious and limited strategies, especially as hard law sanctions will attach only to the strategy the company formulates. To address this issue, we suggest that the public authorities be given a role in supervising the discharge by companies of their good faith obligation. If a lack of good faith was found by the supervisory body, it should have at its disposal an appropriate range of administrative tools. We stress that this supervisory role would not put the public authorities in the position of being required to assess the adequacy of the company’s strategy, still less would they have to say whether they considered the strategy was the best available one. The supervisory role would be limited to assessing whether the company’s strategy was the result of good faith efforts on its part to apply the international standards to its value chain. Of course, failure by a company to consider the relevance to its operations of one or more international standards would be evidence of a failure to approach the assessment in good faith, but the company’s conclusion that a particular standard was of limited relevance to its business or that there was little the company could do to ensure compliance by other persons with a particular standard would not amount in themselves to bad faith. If, as we suggest, the obligation to publish the strategy were normally retained as part of the Directive, there would clearly be reputational sanctions on the company as well, if it did a poor job when devising its strategy.

38. A notable feature of the draft is the obligation upon companies to consult a wide range of stakeholders, on a continuing basis, with regard to the formulation, implementation and revision of the strategy. We think that the concept of company “ownership” of the strategy is inconsistent with a legal obligation to consult a wide range of stakeholders over the strategy. It would turn the good faith assessment by the company into a negotiation. Even if a legal obligation to consult stakeholders were removed, there would still be an incentive for companies to carry out consultation, because it might help to demonstrate good faith. However, a legal obligation to consult would provide stakeholders with a platform to pursue their own agendas rather than incentivise them to work out a feasible strategy for the company. However, it would be possible to provide for stakeholder to be represented in the supervisory body, which would provide them with an important channel of influence at a macro level the discharge by companies of their good faith obligations. And stakeholders would still be entitled to complain to the supervisory authority that the strategy had not be adopted in good faith.

39.  For similar reasons, we do not propose that the company should have to provide a grievance procedure (which is unnecessary if the strategy is legally binding on the company). The supervisory authority might develop remediation machinery which it could make available to complainants and companies. In order to encourage its use, it would be important that the supervisory authority have the power to determine in appropriate circumstances that those choosing to use the remediation process would give up their entitlement to bring civil proceedings, either before entering into the process or before the result of the remediation process was accepted by the complainant.

40. Where the company has acted in breach of the strategy it has adopted, we think that civil remedies should be available, but under the existing and developing principles of Member States’ existing legal systems. In other words, there should be no requirement in the Directive for Member States to amend their national regimes to comply with some EU-level principles, other than the application of national standards in this area. While this proposal accepts that the rules on civil liability are likely to vary from Member State to Member State, this will be the case to a significant degree under the existing draft. This is because national rules on vital procedural matter such as costs, the burden of proof and class actions are not harmonised in the draft (or elsewhere). More important, we think the application of civil sanctions (especially damages) raises many delicate issues about the formulation of civil law duties, causation and the assessment of damages which, for the moment at least, are best left to national courts to experiment with. That national legal systems are engaged in this process already can be seen from the fact that courts, both within and without the EU, have accepted the capacity of groups from outside the jurisdiction of litigation to bring cases against multinational companies incorporated within the litigating jurisdiction in respect of harms inflicted elsewhere.[4] This approach is not inconsistent with conferring on the supervisory authority power to impose administrative penalties where it is clear that the company has ignored its strategy but no harm to any group of persons appears to have been caused.

41. Finally, we think there might be a role for certification by an independent third party, both in relation to drawing up the strategy and in relation to its implementation, at least as far as concerns some elements of the strategy or its implementation. What we have in mind here are analogies with the “Fair Trade” certification provided by the Fairtrade Foundation. We have not investigated in depth the availability or reliability of such certifications in the areas covered by the draft. What we have in mind is the general idea that an appropriate certification would insulate the company, in the relevant area, from supervisory authority action on the grounds that the strategy was not adopted in good faith or, if the certification related to areas of implementation, from civil or administrative liability in relation to the alleged breach of the strategy.  

(b) No direct applicability of the international standards

42. Above, we have proposed a scheme for incorporating the international standards in a workable way into a set of corporate policies which would then be legally binding on the company. For the reasons given above concerning the imprecision and contestable meaning of many of these standards, we think the Directive should reject the principle of direct liability, ie liability for breach of the international standards outside their incorporation in a corporate strategy. Thus, we proposed that Art.10 should be removed from the Directive and Art.19 (as amended in accordance with our proposals in para 40) clearly confined to breaches by the company of its adopted strategy. If there is any doubt about this under the current draft, it should be made clear that the powers of the supervisory authority are limited to the company’s adoption and implementation of its strategy.

V. The policy underlying the Directive

43. We believe that, with the above amendments, the draft would operate so as to meet the standards of clarity, precision and workability normally required of legislation. We also think that the draft, revised in this way, would deal in a more sophisticated way with the difficult policy issues underlying the draft, which naturally do not surface in the draft itself. The crucial point here is that the draft is not confined to the regulation of those parts of the company’s value chain which are located within the EU. Once a company falls within the draft because it “operate[s] in the internal market selling goods or providing services” (Art.1.3) and meets the required size thresholds, then the whole of its value chain, wherever it is located, is subject to the provisions of the draft.

44. The draft assumes that the company will wish to maintain its access to the internal market and so bring its value chain into compliance with the international standards, thus generating a positive externality provided by EU regulation to persons located outside its territory. Many, perhaps most, companies with make this choice. But there are alternatives open to them. The first is to withdraw from business relations with the problematic supplier or customer, perhaps from relations with all businesses in a particular jurisdiction. Assuming the customer or supplier is outside the EU, this will lead to no overall increase in global welfare if the company operating in the EU is replaced by a company not so operating and which is not subject to equivalent restrictions to those in the Directive and which continues business on the same terms (or perhaps even worse ones) than those applied by the EU company. Even worse, global welfare may be reduced if the now-abandoned contractor ceases to trade entirely.

45. Second, a company with a limited presence in the EU may withdraw from the internal market or decide not to enter it, in order to continue its business without regard to the Directive. This is particularly likely where the company supplies the good or service predominantly outside the EU, and so the potentially high costs of complying with the Directive would be attributable to only a small part of the value (to the company) of that chain. Again, there would be no gain to global welfare, since its business relationships outside the EU would continue unaffected. Although this may seem a small loss to the EU, considered narrowly, it may be bad for competition, actual and potential, in the internal market, if the currently small operator (in EU terms) might have developed a larger EU business which would have been a challenge to incumbents. A possible solution would be to develop a test for the level of economic presence in the EU that is required of a non-EU company in order for it to be subject to the obligations of the Directive. Given the global scope of these obligations, the requirements of international trade law and considerations of opportunity for European trade, it would be appropriate to limit the application of the Directive to non-EU companies with substantial presence, measured mainly by their European turnover.

46. Third, without either of these withdrawal decisions by companies, trade may simply be re-routed as a result of market forces, to which the company is forced to respond. This is important because a mere shift in demand for goods from problematic regions can lead to the regulation not achieving its intended effect; after all, the aim is to make the entire global economy more environmentally compatible, more compatible with human rights, etc. If beef from the Amazon is no longer consumed in the EU because the Directive’s requirements have made it more expensive, but instead beef from Argentina, but the Argentines no longer eat their own beef, but instead beef from Brazil, the effect of the new regulation is negligible due to the mere substitution of demand (or even negative, because additional transport is required). Provided only that the exporter of beef from Argentina has no business relationship with the exporter of beef from Brazil, this result could occur without the Directive being engaged.

47. The above risks are inherent in the draft Directive, of course, but we think it is important to minimise them. We think the good faith test mentioned above will enable companies to handle with a higher degree of flexibility pressures on them to choose any of the above courses of action. The good faith test may also make it somewhat easier for companies to minimise a third problem which is likely to be a significant one in some important non-EU markets. This is adverse reactions from states where EU companies have taken action in line with the Directive. This is clearly not a fanciful risk. Thus, the Swedish company, H&M, was recently frozen out of electronic access in the Chinese market, possibly only temporarily, because of a statement made a year earlier in a corporate transparency statement about its concerns relating to forced labour in a Chinese province. See Financial Times, “China forces brands to make a cotton choice” and “H&M and Nike face China backlash over Xinjiang stance” (March 31, 2021). It is to be noted that in the British Institute of International and Comparative Law Report, carried out for the EU  Commission on supply chain due diligence in 2020,[5] H&M was praised on the grounds that the “new transparency policy of H&M is an example of how a company with a complex supply chain can achieve traceability”, though the Report also recorded criticism from the pressure group, Labour behind the Label, that the transparency statement did not give information “about wages paid at suppliers and comparing that to the living wage benchmarks or their promises on living wages.”

48. The more general point is that, across countries, views on human rights, environment protection and good governance are mixed. Referring to national sovereignty, a growing number of states insist on their own standards, which differ from European or “Western” views. Cutting ties with all these economies would not enhance global welfare and might undermine existing human rights dialogues. Complying with both local laws and standards should therefore be the minimum standard required by the Directive. Insistence on worldwide application of European standards risks counterproductive consequences. Human rights standards might even be seen by some as an instrument for hegemony or even European neo-imperialism. Complying with both local standards and practice in most countries meets basic human rights standards. Just as European states are cautious in their responses to what they perceive as the unacceptable aspects of social and political arrangements in some foreign countries, it seems to us that companies should have that freedom as well. A good faith approach would help companies balance the difficult and complex competing considerations in these demanding situations.

VI. Conclusion

49. Overall, we think the European Parliament draft is unworkable as it stands, ie that it cannot be implemented by companies in a rational and straightforward way, essentially because it lacks legal certainty. We propose two changes to remove this defect. These are a shift to good faith as the basis for the due diligence obligation for companies and the removal of those articles which generate liability for direct breaches of the standards (ie other than via their incorporation into the company’s individual strategy).