Conceptualizing Corporate Accountability in International Law: Models for a Business and Human Rights Treaty

This article conceptualizes corporate accountability under international law and introduces an analytical framework translating corporate accountability into seven core elements. Using this analytical framework, it then systematically assesses four models that could be used in a future business and human rights (BHR) treaty: the United Nations Guiding Principles on Business and Human Rights model, the Universal Declaration of Human Rights model, the progressive model, and the transformative model. It aims to contribute to the BHR treaty negotiation process by clarifying different options and possible trade-offs between them, while taking into account political realities. Ultimately, the article argues in favour of the BHR treaty embracing a progressive model of corporate accountability, which combines ambitious development of international law with realistic prospects of state support.


Introduction
In 2014, following a proposal by Ecuador and South Africa, the UN Human Rights Council established an open-ended intergovernmental working group with the mandate to "elaborate an international legally binding instrument to regulate, in international human rights law, the activities of transnational corporations and other business enterprises". The intergovernmental working group held its first session in Geneva in 2015 and its latest session in October 2019. At the time of writing, four draft documents have been published: Elements for the draft legally binding instrument on transnational corporations and other business enterprises with respect to human rights in 2017, the Zero Draft in 2018, the Draft Treaty in 2019, and the Second Revised Draft Treaty in 2020. The very idea of a business and human rights (BHR) treaty has led to numerous debates about whether it is feasible (Bilchitz 2016;De Schutter 2016;McConnell 2017), timely (Ruggie 2014;Leader 2017), or even a positive development from the perspective of victims of corporate human rights abuse (Ruggie 2014;Simons 2017).
The treaty should strengthen corporate accountability for human rights to the benefit of rights-holders, but how this will be done remains uncertain. The complexity of accountability as a concept adds to the confusion. This article aims to answer the following research question: how can corporate accountability be conceptualized in the future BHR treaty so as to maximize the treaty's impact both in terms of contents and state support? To address this question, this article first conceptualizes corporate accountability under international law and introduces an analytical framework translating corporate accountability into seven core elements. Second, it introduces four models of corporate accountability that could be used in the future treaty. It then systematically unpacks those models by means of the analytical framework. It aims to contribute to the BHR treaty negotiation process by clarifying different options and possible trade-offs between them, while taking into account political realities. For example, the BHR treaty may embrace the notion of direct human rights obligations for corporations, but limit those obligations to certain types of corporations. By contrast, the BHR treaty may create direct obligations for states only but require states to regulate the activities of all corporations. Ultimately, the article argues in favour of the BHR treaty embracing a progressive model of corporate accountability, which combines ambitious development of international law with realistic prospects of state support.
The four models of corporate accountability I examine in this article are (1) the UN Guiding Principles on Business and Human Rights (UNGPs) model, with corporate accountability framed on the basis of a do-no-harm requirement, defined itself as a social expectation; (2) the Universal Declaration of Human Rights (UDHR) model, with a wider type of accountability, grounded in international law but no specific enforcement mechanism; (3) the progressive model, with corporate accountability grounded in international law, and relying on specific domestic enforcement mechanisms, including extra-territorial mechanisms; and (4) the transformative model, with corporate accountability also grounded in international law, and relying both on domestic (including extraterritorial) and international enforcement mechanisms.
The term model is understood here as "a system that can be used as an example for other people to copy" (Oxford Learners' dictionary 2020). Throughout the article, "model" is used to refer to types of corporate accountability as already articulated in existing texts (UNGPs, UDHR) or to novel forms of corporate accountability (progressive and transformative models). These models are hypothetical, developed for the purpose of this article, and derived from academic and grey literature. Broadly speaking, as the article shows, every proposal for a BHR treaty put forward by academics, civil society organizations, and public actors fits within one of these models. Some features are critical in shaping a particular model while others are only marginally important. Although not set in stone, the models make it possible to organize the discussion in a systematic manner, and to assess the merits and drawbacks of the different positions behind them.
The article is structured as follows. "Conceptualizing Corporate Accountability: Analytical Framework" conceptualizes corporate accountability and introduces the analytical framework. "UN Guiding Principles on Business and Human Rights Model", "Universal Declaration of Human Rights Model", "Progressive Model", and "Transformative Model" present models for a BHR treaty and unpack them by means of the analytical framework. "Corporate Accountability in the Future Business and Human Rights Treaty: from Theory to Practice" elaborates and endorses the progressive model of corporate accountability for the future BHR treaty with the view to reconcile human rights ideals and political realism. "Conclusion" provides a brief conclusion and suggestions for future research.

Conceptualizing Corporate Accountability: Analytical Framework
In the introduction to the chapter on "Accountability and Remedy" in their book on business and human rights, Baumann-Pauly and Nolan noted that "the term 'accountability' means dramatically different things to different people" (Baumann-Pauly and Nolan 2016, p. 239). The academic literature on accountability reflects this diversity of views. This section briefly investigates how accountability, particularly corporate accountability, is defined in political science and international relations, business and management studies, and law. Drawing on these definitions, the section explains how corporate accountability is conceptualized for the purpose of this article. It then introduces the analytical framework used in the following sections to assess proposed models for a BHR treaty.
Accountability is the subject of an abundant literature in political science and international relations. A recurrent idea is that in a representative democracy "those who govern are held accountable to the governed" (Grant and Keohane 2005, p. 29) However, this is not possible in global governance, where democratic institutions do not exist, which calls for different forms of accountability (Grant and Keohane 2005;Keohane 2006;Bovens 2007). In a nutshell, "the concept of accountability implies that the actors being held accountable have obligations to act in ways that are consistent with accepted standards of behavior and that they will be sanctioned for failures to do so" (Grant and Keohane 2005, pp. 29-30).
Bovens distinguishes between accountability as a virtue and accountability as a mechanism. As a virtue, accountability is a "normative concept" or "a set of standards" against which behaviour may be assessed. As a mechanism, accountability is "an institutional relation or arrangement in which an agent can be held to account by another agent or institution" (Bovens 2010, pp. 947-948). Perhaps a more classic way to conceptualize accountability is to break it down into two concepts, answerability and enforceability (Newell 2008, p. 124). In an article focusing on multilateral climate politics, Gupta and van Asselt propose to further disaggregate answerability and enforceability into five components: "relations, standards, judgments, sanctions, and redress". Relations is about "who is to be held to account (and to whom)"; standards refers to "standards of performance against which to be held to account (i.e. (...) accountability for what)"; and judgments denotes "a process by which to assess if standards are being met (i.e. the how of accountability)" (Gupta and van Asselt 2019, p. 20). Together, those three elements form the answerability segment of accountability.
Sanctions means "(legal, reputational, financial) penalties if standards are not met" and redress focuses on "the scope and modalities of liability and compensation for harm inflicted as a result of standards not being met". Those two elements constitute the enforceability element (Gupta and van Asselt 2019, p. 20).
In business and management studies, the literature naturally focuses on corporate accountability but the exact meaning of holding to account in this context is not well defined. Instead, the literature focuses on the type of actor who can hold corporations to account, whatever this exactly means. Under a traditional approach, corporations are accountable mostly to their shareholders and should prioritize profit-maximization (Benston 1982). Other authors insist that corporations are accountable to all their stakeholders, and not just shareholders (Luo 2005;Cooper and Owen 2007;Mohammed 2013).
By contrast, there is little discussion about the meaning of the term accountability in the legal field. In public international law, responsibility and liability have specific meanings while accountability does not. The Draft Articles on Responsibility of States for Internationally Wrongful Acts provide a clear and authoritative definition of the term "responsibility". Under Draft Article 1, "Every internationally wrongful act of a State entails the international responsibility of that State." Hence, "the term "international responsibility" covers the new legal relations which arise under international law by reason of the internationally wrongful act of a State" (United Nations International Law Commission 2001). Since it is debatable that corporations have international obligations under international law, using the term "responsibility" to describe the legal relations which arise when a corporation violates human rights shows that one has chosen a side in this debate. Under the UNGPs, corporations have a responsibility to respect human rights but in this context, responsibility is not a legal concept. Moreover, the term liability is not fitting either because it implies that there was a legal responsibility to begin with (Bedjaoui 2000, p. 212). Using the term corporate accountability allows staying clear of the debates and uncertainty associated with the notion of responsibility in the context of corporate activity. Unlike responsibility and liability, accountability does not have a clear definition in international law. Therefore, it is fitting to use the term accountability in this exploratory article to describe processes by which corporations can be held to account under international (soft) law. The term is used here in a non-specific, neutral way, encompassing legal, quasi-legal, and non-legal relations.
My proposed framework to assess the different models for a BHR treaty is based on Gupta and van Asselt's framework and builds on it to better suit the issues at stake. In their framework, the questions of who is to be held to account, and to whom, are merged into one component called relations. The BHR treaty negotiation, and BHR as a field in general, raises specific questions related to relations. For example, who is to be held to account leads to questions about the type of corporations the treaty might cover (small/large, transnational/domestic, etc.). Those significantly complicate matters. This is why in my framework I break this component in two and ask two distinct questions: (1) who is to be held to account; (2) to whom?
In Gupta and van Asselt's framework, the next question is "accountability for what?", which they frame as standards of performance. This component is crucial in a legal context, but needs to be further broken down. The content of the standard is important, but so is its legal force (binding/non-binding/grey area). In my framework, I ask two questions: (1) what is the expected behaviour?; (2) what is the standard's legal force?
Gupta and van Asselt then look at the how of accountability, which is the "process by which to assess if standards are being met". They describe the process of state reporting and peer review in place in climate governance. In a proposed BHR treaty, self-reporting by states is one option. One could also see the adoption of provisions that have a certain legal force and contents, and aim to hold certain corporations to account, but without a formal process to assess if standards are being met. This process may well be decentralized, with civil society organizations assessing whether the standards are being met and then relying on different types of sanctions (see below). Hence, in my framework, the word "process" is to be understood loosely. I ask the question: "what is the process by which to assess if standards are being met?" Finally, Gupta and van Asselt focus on sanctions and redress. Their framework can here be used almost as such with regard to sanctions. It translates into the following question: what sanctions (legal, reputational, financial) will happen if standards are not met? Concerning redress, human rights law calls for more nuance. Of relevance here is the distinction between reparations and remedy. In 2005, the UN General Assembly defined the right to remedy as encompassing: "(a) Equal and effective access to justice; (b) Adequate, effective and prompt reparation for harm suffered; (c) Access to relevant information concerning violations and reparation mechanisms" (UN General Assembly 2005). In other words, reparations are part of the right to remedy. The framework needs to reflect this complexity. The final questions are what remedial processes are in place (judicial/non-judicial)? and what forms of reparations are available to victims?
The table below summarizes the framework, which consists of seven questions designed to facilitate discussion of key issues ( Table 1).
The following sections ("UN Guiding Principles on Business and Human Rights Model", "Universal Declaration of Human Rights Model", "Progressive Model", and "Transformative Model") are structured in the same way. First, each model of corporate accountability is briefly introduced. Second, the model is broken down into its seven components. To avoid repetition, this part describes each element but also includes analysis where warranted. Third, in a final part called "Evaluation", the model as a whole is discussed against the framework, which is therefore non-normative.

UN Guiding Principles on Business and Human Rights Model
The UNGPs, endorsed in 2011 by the UN Human Rights Council, are the most authoritative international standard in business and human rights. They were developed by Professor John Ruggie, then Special Representative of the UN Secretary-General on Human Rights and Transnational Corporations and Other Business Enterprises during his mandate. The UNGPs rest on three pillars: (a) the state duty to protect human rights against abuses by businesses; (b) the corporate responsibility to respect human rights, which is a social expectation not grounded in international law; and (c) enhanced access to remedy.
1. Who is to be held to account? Under the UNGPs model, the corporate responsibility to respect human rights applies "to all enterprises regardless of their size, sector, operational context, ownership and structure". (United Nations 2011). The UNGPs also single out certain individuals within the business. GP 16 asks that the company's policy commitment to meet their responsibility to respect human rights be "approved at the most senior level of the business enterprise". Under this model, senior executives such as CEOs and board members can therefore be held accountable if they do not follow these recommendations. 2. Accountable to whom? The corporate responsibility to respect human rights is "a global standard of expected conduct for all business enterprises." In this sense, under the UNGPs corporations are accountable to society as a whole, an undefined group of individuals and organizations. GP 29, as part of Pillar III on access to remedy, singles out "individuals and communities who may be adversely impacted by a business enterprise" and for whom the company must set up operational-level grievance mechanisms. 3. What is the expected behaviour? Pillar 2 of the UNGPs establishes the corporate responsibility to respect human rights. There is no responsibility to protect, let alone fulfil human rights. What is more, the UNGPs' approach to respecting human rights is that it entails doing no harm, which is the narrowest possible interpretation of this notion. Some have even questioned whether such a limited definition warrants the use of the term responsibility (Karp 2014(Karp , 2020. Guiding Principle 12 elucidates that the corporate responsibility to respect human rights "refers to internationally recognized human rightsunderstood, at a minimum, as those expressed in the International Bill of Human Rights and the principles concerning fundamental rights set out in the International Labour Organization (ILO)'s Declaration on Fundamental Principles and Rights at Work." Although the list is open-ended ("at a minimum"), the commentary to this principle makes clear that the rights of, for example, indigenous peoples and women are "additional standards" that "business enterprises may need to consider". By presenting the rights of the most vulnerable as add-ons, the UNGPs are not anchored strongly enough to the international legal framework in its entirety. This à la carte approach undermines "the widely accepted doctrine of the indivisibility and interrelatedness of all human rights articulated in the Vienna Declaration and Programme of Action 1993." (Simons 2017, p. 60). 4. What is the standard's legal force? The UNGPs differentiate between state obligations under international human rights and labour law, articulated as the state duty to protect human rights (pillar 1), and the corporate responsibility to respect human rights (pillar 2), which is a mere social expectation. By separating the two, the UNGPs bring conceptual clarity. Whether or not a state fulfils its obligations, companies should respect human rights at all times and in all circumstances because society expects them to do this (Ruggie 2013, p. 84). Pillar 2 rests on the idea that companies that do not respect human rights may lose their "social license to operate" (Ruggie 2008, p. 17). To avoid reputational damage, being judged in "courts of public opinion" and lose their social licence to operate, companies should respect human rights (Wheeler 2015, p. 764). The social licence to operate rests on the assumption that companies fear, and therefore seek to avoid, reputational damage. (Ruggie 2013, pp. 93-94 (United Nations ECOSOC 1976). Whether the code was going to impose direct obligations on corporations was envisaged but never solved (United Nations ECOSOC 1976). In the 1980s, discussions on the Code of Conduct became embroiled in wider debates on foreign direct investment and economic sovereignty of host states and had to be abandoned (Bernaz 2016, pp. 174-175;Hamdani and Ruffing 2017). The 2003 UN Draft Norms went even further than the Code on the issue of corporate accountability. The Draft Norms purported to assign human rights obligations to corporations under international law, albeit in a limited number of areas. The Norms proved unacceptable to most states (Bernaz 2016, p. 188). In this context, it is perhaps unsurprising that when given the opportunity a few years later, states rallied around the non-legal corporate accountability model of the UNGPs, seemingly with little difficulty. 5. What is the process by which to assess if standards are being met? There is no formalized process to assess if corporate accountability standards are being met under the UNGPs. In practice, civil society organizations, academic researchers, and journalists are assessing corporate practice, on behalf of society as a whole.
Other types of stakeholders, such as investors and business partners, also assess corporate practices but for their own needs. 6. What remedial processes are in place? The UNGPs do not set up any remedial process but make recommendations for a better use of state-based and non-statebased mechanisms and outline key characteristics processes should have.
7. What forms of reparation are available to victims? The UNGPs themselves do not provide for reparations but the commentary of GP 25 does list possible reparations.
Evaluation From a legal perspective, the overall picture that emerges is that the UNGPs model is a weak form of corporate accountability. It covers all companies and it is an important tool for advocacy in the area of business and human rights. Those are its main strengths. By contrast, it is institutionally weak, with no specific enforceability mechanism. As a response to the limitations of the UNGPs model for corporate accountability, a strong civil society movement for a BHR treaty emerged that views the treaty as a way to address the perceived weaknesses of the UNGPs (Bernaz and Pietropaoli 2017; Garrido Alves 2019). Admittedly, the very adoption of a BHR treaty, irrespective of contents, would strengthen corporate accountability and would be a step forward compared with the UNGPs, which are only recommendatory in nature. An explicit mention of the UNGPs in the treaty would further strengthen corporate accountability by bringing coherence to the business and human rights regulatory landscape. However, if the treaty was to adopt the same model of corporate accountability as the UNGPs, it would bring only marginal change in terms of corporate accountability. Specifically, the treaty would incorporate the following three elements. First, it would define corporate accountability on the basis of a do no harm requirement, with no role in the protection, let alone fulfilment of human rights. Second, the treaty would adopt a limited definition of human rights that undermines the ideals of indivisibility and interrelatedness of human rights. Third, the treaty would define corporate responsibility as a vague social expectation relying, like the UNGPs do, on the concept of the social licence to operate. It would mention the concept in the preamble of the treaty, and not in binding, operative provisions. Adopting this model in the treaty is tempting for treaty supporters because of the consensus surrounding the UNGPs (Methven O'Brien 2020, p. 190). It would fit within a state-centric vision of international human rights law (Van Ho 2018). However, it would be the weakest outcome from the perspective of advancing corporate accountability under international law.

Universal Declaration of Human Rights Model
The UDHR is a resolution adopted in 1948 by the newly established UN General Assembly. As such, it is formally non-binding but aims to provide broad standards of conduct in the area of human rights.
1. Who is to be held to account? Unlike the UNGPs, which clearly address the private sector, the UDHR does not explicitly mention corporations. However, two provisions are relevant from the perspective of corporate accountability. First, the preamble of the UDHR presents the Declaration as a common standard of achievement for all peoples and all nations, to the end that every individual and every organ of society, keeping this Declaration constantly in mind, shall strive by teaching and education to promote respect for these rights and freedoms and by progressive measures, national and international, to secure their universal and effective recognition and observance. (emphasis added) "Every organ of society" is clear enough language, which "excludes no one, no company, no market, no cyberspace" (Henkin 1999, p. 25). Besides the preamble, Article 30 of the UDHR also sets standards of behaviour for companies. It provides that "nothing in this Declaration may be interpreted as implying for any State, group or person any right to engage in any activity or to perform any act aimed at the destruction of any of the rights and freedoms set forth herein." The travaux préparatoires of the UDHR indicate that the term "group" was meant to include private companies (United Nations General Assembly Third Committee 1948) Thus, as early as 1948, states intended for corporations to fall under the purview of the UDHR. 2. Accountable to whom? The UDHR defines human rights standards meant for humanity as a whole, which similarly to the UNGPs' mention of society as a whole is vague. 3. What is the expected behaviour? In contrast to the UNGPs model of corporate accountability, the UDHR model expects corporations to not only do no harm, or even respect human rights, but also "keeping this Declaration constantly in mind" to engage in "teaching and education to promote respect for these rights and freedoms". This approach to corporate accountability is broader. It views corporations not as entities that need to be kept at bay but as actors for the promotion of human rights for all who are expected, arguably required, to use their agency to this end. In terms of contents, the UDHR lists core human rights that every organ of society, including corporations, should respect with little detail about what that entails in practice. 4. What is the standard's legal force? Adopted as a UN General Assembly Resolution, the UDHR is formally a non-binding declaration creating no obligations under international law. However, many would argue that at least some provisions of the UDHR have become part of customary international law (Schabas 2013, p. cxxi.). 5. What is the process by which to assess if standards are being met? There is no formalized process to assess if corporate accountability standards are being met under the UDHR. As with the UNGPs model, civil society organizations, academic researchers, journalists, investors, business partners, and other stakeholders may assess corporate practice against standards set in the UDHR. However, because the UDHR is not an instrument specifically designed for corporate accountability, stakeholders are more likely to use other instruments. 6. What remedial processes are in place? The UDHR does not set up any remedial process but Article 8 states that "everyone has the right to an effective remedy by the competent national tribunals for acts violating the fundamental rights granted him by the constitution or by law." 7. What forms of reparation are available to victims? The UDHR does not provide for specific reparations for victims of corporate human rights abuse.
Evaluation In the same way as with the UNGPs model, the adoption of a BHR treaty embracing the UDHR model, however weak and irrespective of the exact contents, would strengthen corporate accountability and, on this point, it would be a step forward from the formally non-binding UDHR. Under the UDHR model, corporate accountability is grounded in international law, though it does not include specific enforcement mechanisms and is therefore very weak on this point. As with including the UNGPs model in the treaty, I must distinguish between two possibilities. First, the treaty could embrace the UDHR model simply by adopting the same type of corporate accountability as exists under the UDHR. This option is the weakest. Second, the treaty could embrace the UDHR model by making an explicit reference to the UDHR, thus presenting the treaty as the next step in the process of strengthening corporate accountability under international law. I focus here on the second option. Anchoring corporate accountability to this symbolic instrument is appealing for treaty supporters. Substantively, the UDHR model is broader than the UNGPs model because it implies that corporations have a role to play in the protection and promotion of human rights, beyond the mere do no harm approach of the UNGPs. The UDHR model also rests on relatively solid ground because it benefits from the status of the UDHR at least as a symbolic instrument in the international political landscape, and arguably as customary international law. Finally, the UDHR model is interesting because although the UDHR was adopted over 70 years ago, and it does not explicitly mention the private sector, companies themselves seem familiar with this instrument. It is frequently mentioned in corporate reports. In a survey conducted in 2007 in the context of his mandate as UN Secretary-General Special Representative, John Ruggie asked the Fortune Global 500 companies "what if any international human rights instruments their policies and practices draw upon" (Ruggie 2007, p. 24). The results indicate that "approximately one-fourth of the respondents skipped this question, presumably indicating that they reference no international instrument. Among the other 75 per cent, ILO declarations and conventions top the list, referenced by seven out of ten. The Universal Declaration on Human Rights (UDHR) is the next highest" (Ruggie 2007, p. 24). Moreover, "in the extractive sector, (...) every single respondent cites the UDHR" (Ruggie 2007, p. 24

Progressive Model
Under the progressive model, which I have designed drawing on various proposals stakeholders have put forward over the years, corporations have legal obligations. Corporate accountability is grounded in international law but enforcement mechanisms are domestic and include the possibility of extraterritorial enforcement. Unlike the previous two models, this model is not based on any specific international human rights instrument. However, it borrows from existing forms of corporate accountability under international law in areas other than human rights, for example under environmental treaties (Deva 2015) and Part XI of the UN Convention on the Law of the Sea (Bernaz and Pietropaoli 2020).
1. Who is to be held to account? Under this model, corporations are explicit dutybearers and two options are possible: either the treaty covers all corporations; or it covers only transnational corporations. Both options could work but because the treaty would create binding obligations, the type of corporations covered by the future BHR treaty needs to be established without ambiguity. The mandate of the open-ended intergovernmental working group is to "elaborate an international legally binding instrument to regulate, in international human rights law, the activities of transnational corporations and other business enterprises". As discussed above, the UNGPs model encompasses "all enterprises regardless of their size, sector, operational context, ownership and structure". (United Nations 2011). Arguably, this approach is possible because under the UNGPs model businesses are not subjected to legally binding obligations. Under the progressive model, however, corporations would be subject to such obligations so the broadbrush approach appears more difficult. 2. Accountable to whom? Under the progressive model, duty-bearers (corporations) are accountable to rights-holders, that is anyone whose human rights are potentially impacted by the activities of the duty-bearers. Corporations may also be accountable to society as a whole. 3. What is the expected behaviour? Under this model, different options are possible.
First, similarly to the UNGPs model, the progressive model can require corporations to respect human rights, and not necessarily to protect or fulfil them. It can also encompass, at a minimum, the instruments listed in Guiding Principle 12, namely the International Bill of Rights and the ILO Declaration. Second, the progressive model can also go further on this and encompass more human rights standards as well as the full range of human rights obligations from respecting to fulfilling human rights. 4. What is the standard's legal force? Under the progressive model, corporations have direct legally binding obligations. This is a critical feature of the model. 5. What is the process by which to assess if standards are being met? Different processes are possible and fall under two main categories: monitoring and enforcement. First, as with the UNGPs model, civil society organizations, academic researchers, journalists, investors, business partners, and other stakeholders may monitor how corporations meet the standards set in the treaty. Those processes are unsystematic and decentralized; how standards are interpreted may vary and are likely to produce different results. Under this model, states have the obligation to ensure, through domestic mechanisms, that corporations abide by their international obligations, including when they operate abroad. The state responsibility to protect human rights against corporate abuse can be monitored via an international mechanism, such as a treaty body or a human rights court. Indirectly, then, those bodies may assess whether corporations meet the standards. Second, under the progressive model, corporate accountability enforcement mechanisms remain at the domestic level, through judicial and non-judicial mechanisms that also include extraterritorial enforcement. Each mechanism has its own process to assess corporate conduct, following local standards of proof and rules of procedure and evidence. Prior to engaging with such formal processes, victims, supporting NGOs and their lawyers conduct a legal assessment of corporate conduct in order to ascertain their chances of success in a given forum, with special attention given to jurisdictional issues in extraterritorial cases.
6. What remedial processes are in place? The progressive model relies almost entirely on domestic mechanisms for remedies. Quality and accessibility of the justice system vary considerably from state to state, especially because in many cases victims seek justice in the country where the parent company is incorporated or has its main place of business, thus raising complex extraterritoriality-related issues. Access to justice for victims of corporate abuse is therefore as good as access to justice for victims of human rights abuse generally in the state in question. Legal remedies may be under civil or criminal law, and under this model, states are free to choose the most appropriate legal remedies suitable for their legal systems and traditions. To complement judicial mechanisms, the progressive model may also rely on non-judicial mechanisms, following the example of the Organization for Economic Cooperation and Development (OECD) National Contact Points (NCPs) mediation procedure. 7. What forms of reparation are available to victims? The progressive model mandates states to provide reparation for victims but leaves states free to choose the forms it may take. Reparation varies greatly from state to state. For example, legal systems may include punitive damages on top of compensatory damages, allow for compensation in nature, and rely on civil litigation to deal with issues that would fall under the criminal law in other systems (Scheffer and Kaeb 2011).
Evaluation The progressive model is a significant step forward for corporate accountability compared with the UNGPs and UDHR models. Including this model in the treaty would result in a text that sets up clear obligations for companies to respect human rights, however broad, and for states to prevent and facilitate the redress of corporate human rights abuse, including through exercising extraterritorial adjudicative jurisdiction (Bernaz 2013). To ensure that states actually live up to their obligations, a treaty body would monitor whether and how states meet their obligations. A treaty embracing this model would be an important addition to the existing UN human rights treaty system and would not radically deviate from the other UN human rights treaties in terms of approach and practical features. The key innovation would be the recognition of direct binding human rights obligations for corporations. This model's weaker point is that it relies on domestic systems of enforcement, making it almost entirely reliant on state performance in this area as well as particular rules of jurisdiction (Roorda 2019).

Transformative Model
The transformative model of corporate accountability for human rights is grounded in international law. Unlike the progressive model, the transformative model also includes international enforcement mechanisms. At a minimum, the transformative model includes a treaty body competent to receive petitions against corporations, even if it cannot issue binding decisions.
1. Who is to be held to account? The transformative model is the same as the progressive model on this point in that corporations are explicit duty-bearers.
2. Accountable to whom? The transformative model is the same as the progressive model on this point. Duty-bearers (corporations) are accountable to rights-holders, that is anyone whose human rights are potentially impacted by the activities of the duty-bearers. Corporations may also be accountable to society as a whole. 3. What is the expected behaviour? As with the progressive model, several options are possible. The transformative model may require corporations not only to respect human rights but also to protect, and even to fulfil them. It can also only require corporations to respect human rights. This model encompasses at a minimum the instruments listed in Guiding Principle 12, namely the International Bill of Rights and the ILO Declaration, but could also include other instruments, particularly other UN human rights treaties. 4. What is the standard's legal force? Under the transformative model, corporations have direct legally binding obligations. This is a critical feature of the model. 5. What is the process by which to assess if standards are being met? The transformative model comprises all the monitoring and enforcement mechanisms of the progressive model, including extraterritorial mechanisms, and incorporates international enforcement mechanisms as well. International mechanisms can take various forms. One possibility is a treaty body competent to receive petitions against corporations as well as states, even if it cannot issue binding decisions. Fasciglione has elaborated on the other possible international enforcement mechanisms in business and human rights. In a piece published in 2019, he draws attention to the deficiencies of relying on host states to provide remedies and defends a point of principle about rights needing to be enforceable; otherwise, they risk being "meaningless" (Fasciglione 2019, p.186). In the same piece, he presents three options for enforcement. First, he discusses the creation of a corporate human rights chamber at the International Court of Justice. This is quickly dismissed as unfeasible as it would require amending the UN Charter (Fasciglione 2019, p. 190). Second, he considers reorganizing the three regional human rights systems to grant them competence ratione personae over corporations. At the moment, the European Court of Human Rights, the African Court of Human and People's Rights and the Inter-American Court of Human Rights only have jurisdiction over states. All three systems have dealt with cases on state responsibility for failing to protect human rights against corporate abuse. Moreover, the political bodies in the European and Inter-American systems have endorsed the UN Guiding Principles on Business and Human Rights. In the African system, the Malabo Protocol adopts corporate responsibility alongside state responsibility for human rights violations (Fasciglione 2019, p. 191). Third, he suggests the adoption of the so-called ICC model for enforcing corporate obligations under international law, recognizing that this would mean limiting enforcement to violations of human rights that amount to international crimes. He proposes either the creation of a tailor made criminal court or amending the ICC Statute. Each of these proposed international enforcement mechanisms would have its own process to assess whether the standard has been met. 6. What remedial processes are in place? The same points made above regarding remedial processes under the progressive model apply. Access to justice before the international mechanisms also needs to be facilitated. 7. What forms of reparation are available to victims? Under the transformative model, reparations can take as many forms as under the progressive model. Corporate Accountability in the Future Business and Human Rights Treaty: from Theory to Practice Previous sections have conceptualized corporate accountability using hypothetical models for a BHR treaty. This section moves from theory to practice and examines treaty proposals, particularly the 2019 Draft Treaty and 2020 Second Revised Draft Treaty, against those models. Ultimately, I advocate for the progressive model of corporate accountability in light of the current negotiation process.
1. Who is to be held to account under the future treaty has prompted many discussions. Under Resolution 26/9 of the Human Rights Council, the mandate of the open-ended intergovernmental working group is to "elaborate an international legally binding instrument to regulate, in international human rights law, the activities of transnational corporations and other business enterprises". The resolution, however, contains a footnote that states that ""other business enterprises" denotes all business enterprises that have a transnational character in their operational activities, and does not apply to local businesses registered in terms of relevant domestic law". In October 2015, at the first session of the working group, the European Union demanded that text in the session's programme of work be changed from "TNCs and other business enterprises" to "TNCs and all other business enterprises" (Lopez and Shea 2015, p 112). The EU failed to secure this change and consequently left the session that year. While many stakeholders agreed with the EU's proposal, most states present at the session argued that the change would alter the mandate of the working group in an unacceptable manner. Lopez and Shea noted in 2015 that "this is likely to remain one of the stickiest issues in the process" (Lopez and Shea 2015, p. 114). The 2018 Zero Draft was limited to business activities of a transnational character. Article 3(1) of the 2019 Draft Treaty on Scope, however, indicates that the treaty "shall apply, except as stated otherwise, to all business activities including particularly but not limited to those of a transnational character." As Cassel noted after the publication of the 2019 Draft Treaty, "this clearly improves the treaty's chances of success and is a welcome development" (Cassel 2019). The 2020 Second Revised Draft Treaty is unchanged on this point though Draft Article 3(2) does introduce flexibility for states to adapt their domestic law to different types of businesses (Deva 2020). 2. Accountable to whom? The 2020 Second Revised Draft Treaty focuses on "victims" which Article 1(1) defines as "any persons or groups of persons who individually or collectively have suffered harm, including physical or mental injury, emotional suffering, or economic loss, or substantial impairment of their human rights, through acts or omissions in the context of business activities, that constitute human rights abuse" by a business enterprise, as well as direct victims' family and dependents and "persons who have suffered harm in intervening to assist victims in distress or to prevent victimization." As Deva noted, the text is now "defined in line with the [UN] Basic Principles and Guidelines on the Right to a Remedy and Reparation for Victims" (Deva 2020). 3. What is the expected behaviour? With regard to scope, the Second Revised Draft Treaty provision sits between the UNGPs and the UDHR models. Draft Article 3(3) states that the treaty "shall cover all internationally recognized human rights and fundamental freedoms emanating from the Universal Declaration of Human Rights, any core international human rights treaty and fundamental ILO convention to which a state is a party, and customary international law". This contrasts with the undefined approach adopted in the 2019 Draft Treaty, which focused on "all human rights", while keeping a wide scope (Deva 2020 (Hulme 2016). The formulation used in the Draft Treaty draws heavily on the UNGPs, as seen in Table 2. This single-sentenced paragraph is the only provision in the Second Revised Draft Treaty that hints at some form of corporate accountability at the international level. The rest of the Draft is concerned with state obligations regarding protecting human rights against corporate abuse, and how they should discharge this obligation in their domestic legal systems, including by setting up corporate due diligence mechanisms to be monitored domestically. Admittedly, the very fact this provision is put in a treaty, albeit in the preamble, creates a form of corporate accountability that is stronger than the mere social expectation outlined in the UNGPs. From the strict perspective of legal force, however, corporate accountability in the 2020 Second Revised Draft Treaty is weak. Arguably, it is even weaker than corporate accountability under the UDHR model because it is articulated in the preamble and not in the operative part of the treaty. 5. What is the process by which to assess if standards are being met? The 2020 Revised Draft Treaty relies exclusively on domestic processes of enforcement, including extraterritorial mechanisms. Under Article 6(1), states "shall ensure that their domestic law provides for a transformative and adequate system for legal liability for human rights violations or abuses in the context of business activities including those of transnational character". 6. What remedial processes are in place? and 7. What forms of reparation are available to victims? The 2019 Draft Treaty requires states to have "effective, proportionate and dissuasive sanctions and reparations to the benefit of the victims" (Article 6(4)). No further detail is provided. Article 13 provides for the creation of a treaty body who can receive regular state reports and issue concluding observations and general comments (Kletzel G et al 2018).
Thus, with regard to enforcement, the Second Revised Draft Treaty is close to the progressive model and provides some guidance about how states should go about enforcing corporate accountability domestically, including through extraterritorial jurisdiction where relevant. The treaty provides for a treaty body with jurisdiction over states only. Over the course of the negotiation, an international forum to receive complaints against corporations was mentioned, but for now, this GP11: Business enterprises should respect human rights. This means that they should avoid infringing on the human rights of others and should address adverse human rights impacts with which they are involved. GP14: The responsibility of business enterprises to respect human rights applies to all enterprises regardless of their size, sector, operational context, ownership and structure. GP17(a): [Human rights due diligence] Should cover adverse human rights impacts that the business enterprise may cause or contribute to through its own activities, or which may be directly linked to its operations, products or services by its business relationships; Underlining that all business enterprises, regardless of their size, sector, operational context, ownership and structure have the responsibility to respect all human rights, including by avoiding causing or contributing to adverse human rights impacts through their own activities and addressing such impacts when they occur (emphasis added).
idea seems to have been abandoned (Lopez and Shea 2015, p. 115;Lopez 2017, p. 363;Macchi 2018). On the whole, the 2020 Second Revised Draft Treaty embodies a halfhearted form of corporate accountability, far from the expectations of many stakeholders (Bernaz and Pietropaoli 2017;Garrido Alves 2019). If the final text of the treaty were to keep this wording, the outcome would be disappointing to many civil society organizations. Arguably, this language is the price to pay to bring more states on board, particularly from the Western European and Others Group at the United Nations, which is crucial for the success of the new instrument. This might well be the case, but I submit here that the progressive model offers a balanced, moderate alternative which ought to be reconsidered. The crucial feature of this model is the inclusion of direct human rights obligations for corporations. This issue has polarized the discussion, unnecessarily in my view. In reality, direct obligations for corporations exist (López Latorre 2020; Bilchitz 2013) in fields of international law other than human rights, for example in environmental treaties and the UN Convention on the Law of the Sea, which count many states parties (Deva 2015;Arnold 2016;Bernaz and Pietropaoli 2020). The new treaty including direct human rights obligations would be an important step forward and would certainly constitute progressive development of international law, but not drastic change.
By contrast, the new treaty including the transformative model of corporate accountability would be a substantial change. All the options for international enforcement that were discussed in "Transformative Model" involve significant changes to existing legal regimes. Technical complexity and cost would likely trigger political resistance.

Conclusion
This article has conceptualized corporate accountability under international law and introduced an analytical framework translating corporate accountability into seven core elements (see Table 1). The article has also introduced four possible models of corporate accountability that could be included in the future BHR treaty, based on existing systems (UNGPs and UDHR) or proposals (progressive and transformative models). Using the analytical framework, I have systematically assessed the four models to get a full picture of possible options, and how they relate to each other. I then have looked at the current proposals for a BHR treaty, particularly the 2020 Second Revised Draft Treaty, to see how they fit within the framework and what model(s) they follow.
Putting the proposals in context has highlighted the current approach is weak from the perspective of corporate accountability, and imposing clear, direct human rights obligations upon corporations needs to be considered. In this respect, the progressive model of corporate accountability combines ambitious development of international law while still maintaining flexibility regarding the definition of human rights, the scope of obligations, and the creation of a treaty body competent to receive state reports and individual petitions against states. All this is likely to make it a more attractive proposition for states and to ensure a healthy ratification rate.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article's Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.