How to Incorporate Human Rights Obligations in Bilateral Investment Treaties?

Bilateral investment treaties (BITs) are, at least in their present form, asymmetrical. Foreign investors are being accorded substantive rights under these treaties without being subject to any specific obligations. In this context, one question that has been increasingly debated in academia and in civil society is whether there is a need for a greater degree of balance in BITs between the legitimate interests of investors and host countries. This question is part of a boarder debate on how human rights violations committed by corporations doing business abroad can best be addressed. Some international instruments, such as international human rights treaties, are specifically directed at the activities of corporations. However, the obligations contained in these instruments are binding on the contracting states and not on corporations themselves.[1] International law (as it now stands) does not impose any direct legal obligations on corporations.[2] However, nothing in international law prevents countries from signing treaties (such as BITs) that would impose human rights obligations upon corporations.

This short essay examines one promising option: to impose human rights and other non-investment obligations directly upon corporations in BITs.[3] Very few BITs refer directly to human rights issues. However, when they do, they clearly do not impose any binding obligations on foreign corporations. As a result, human rights concerns can only be raised in a very limited number of circumstances before arbitral tribunals in the context of BIT arbitration proceedings.[4] The following paragraphs will provide a concrete analysis of how BITs could be drafted (or redrafted) to incorporate non-investment obligations.

The first question is where human rights obligations on investors could be placed in BITs? Referring to corporations’ responsibilities in the preamble of a BIT would undoubtedly have a positive impact. The preamble is a contextually important part of a treaty and could serve to indicate and colour the treaty’s object and purpose. However, a simple reference to human rights in the preamble would not create any substantive obligations for the investors. A more promising avenue is for human rights obligations to be expressly referred to in the main text of the BIT. The type of language used is another related and equally important issue. Merely encouraging investors to do something has not worked in the past and is unlikely to be an effective remedy in the future. It is therefore paramount that a treaty provision creates mandatory legal obligations that would force corporations to adopt a certain behavior. The provision must also establish a mechanism whereby non-compliance is efficiently sanctioned by an arbitral tribunal.

A pragmatic approach calls for limiting the scope of obligations imposed upon corporations in BITs to those found in the following areas of law: human rights, labour rights, protection of the environment and anti-corruption. But how should these obligations be incorporated into BITs? One option would be for parties to determine for themselves, during treaty negotiations, which of the many fundamental human rights and other non-investment obligations they want to include in the BIT. In our view, this is not the most suitable approach as such negotiations would likely take a considerable amount of time and raise numerous controversial issues. A more straightforward solution would be for BITs to refer directly to the following five well-recognized international instruments that corporations must comply with: Universal Declaration of Human Rights (1948), United Nations International Covenant on Civil and Political Rights (1966), ILO Declaration on Fundamental Principles and Rights at Work (1998), United Nations Convention Against Corruption (2003),and the Rio Declaration on Environment and Development (1992).

The first reason for choosing these particular instruments is because they have been ratified or endorsed by a significant number of countries. It is easier to convince countries to incorporate human rights obligations when the principles contained in these few instruments are not controversial and are supported by the vast majority of them. In fact, the content of some of these instruments is considered as customary international law. The second reason why these five instruments should be selected is simply because they are already accepted by a large number of corporations as guiding principles of conduct for their business activities abroad. This is, for instance, the conclusion reached by John Ruggie, the former UN Special Representative for Business and Human Rights, in a survey on representatives of multinational corporations.[5] Similarly, in the context of the “United Nations Global Compact,” a non-binding initiative, no less than 8,700 corporate participants and other stakeholders from over 130 countries have committed themselves to respect ten universal principles that are drawn from the above-mentioned five instruments.[6] These ten principles have also been accepted by a considerable number of countries via a UN General Assembly Resolution in 2010.[7] It is therefore submitted that countries will be increasingly more open to the imposition of international legal obligations on corporations knowing that there already exists widespread support for them in the business community.

The above-mentioned five international instruments are certainly not the only ones that could be referred to in BITs. Reference could also be made, for instance, to soft law instruments that have been adopted by countries; instruments such as the ILO Tripartite Declaration of Principles Concerning Multinational Enterprises or the OECD Guidelines for Multinational Enterprises. However, the main problem with this proposal is the non-binding nature of these instruments. While it is true that these soft law instruments include some principles that are themselves contained in other international treaties that are binding on countries, the other principles do not impose any obligations on either party. Countries are unlikely to be willing to ‘transform’ these soft law instruments into hard law ones by simply incorporating them in their BITs.

Referencing specific international treaties in a BIT is only the first step to be considered when seeking to improve the protection of human rights in the context of BITs. The investor-state dispute resolution clause of the treaty must also contain a provision indicating specifically how human rights obligations imposed upon corporations can actually be enforced before an arbitral tribunal. The provision must make it clear that an arbitral tribunal has jurisdiction over allegations of human rights violations committed by corporations. There are at least three different enforcement possibilities that can be envisaged in a BIT’s investor-state dispute resolution clause. We have proposed elsewhere specific drafting examples for each option.[8]

As a first option, an investor’s protection under a BIT could be conditioned upon its respect for human rights (and other non-investment) obligations. This is the ‘clean hands’ doctrine.[9] Contracting parties are free to limit consent to arbitration to disputes that satisfy specific characteristics. Thus, nothing prevents them from conditioning the availability of substantive protection for investors on their compliance with fundamental human rights obligations. If a tribunal comes to the conclusion that a corporation has committed human rights violations contrary to its obligations under a BIT, it could find the investor’s claim inadmissible. In fact, several arbitral tribunals have already, to some extent, made use of the clean hands doctrine and held that they either lacked jurisdiction or that a claim was inadmissible when faced with the illegal conduct of an investor, such as misrepresentations made by the claimant, fraud, or bribery/corruption. The solution that prevailed so far for bribery, should, a fortiori, be applicable when a tribunal comes to the conclusion that a corporation has committed human rights violations contrary to its obligations under a BIT.

A second available option would be to permit an investor’s claim even in the face of human rights violations, but to allow the respondent state to raise any such allegations during the arbitral proceedings. This is the ‘offsetting of damages’ (or ‘mitigation’) option. A tribunal would thus take into account such allegations when making its determination on the merits of the dispute. Such allegations should also have an impact on a tribunal’s assessment of compensation for damages claimed by an investor. A third available, ‘counterclaim,’ option is a variant of the ‘mitigation option.’ Under this option, a claimant investor would be permitted to file a claim even in the face of human rights violations, but the host country would be allowed to raise human rights allegations in a counterclaim. The possibility for counterclaims by host countries should be expressly provided for in the BIT’s investor-state dispute resolution clause.

At the moment, the prospect of a new generation of BITs balancing the rights and obligations of corporations is uncertain. There does not seem to be any clear political will amongst countries for such developments. Ultimately, all countries, both developed and developing, would have a great interest in pursuing these changes in future treaties. In our view, emerging markets (which appear as host countries in most cases) will increasingly realize that the proposed changes are for their benefit since they provide additional tools in their defence against claims by foreign investors. Objections to the proposed changes may come from capital-exporting countries whose goal in signing BITs is to provide extensive legal protection to their national investors conducting business abroad. Yet, there is a growing concern about the negative impact that corporate activities may have on local populations with respect to human rights and related issues. In our view, one simple way for capital-exporting States to respond to these grievances from segments of civil society would be to adopt BITs imposing human rights obligations upon corporations.

Authors: Patrick Dumberry is Assistant Professor, Faculty of Law (Civil Law Section), at the University of Ottawa (Patrick.dumberry@uottawa.ca). Gabrielle Dumas-Aubin is a Quebec Bar student.


[1] “Business and Human Rights: Mapping International Standards of Responsibility and Accountability for Corporate Acts,” Report of the Special Representative of the UN Secretary-General, Mr John Ruggie, on the issue of human rights and transnational corporations and other business enterprises, UN Doc. A/HRC/4/035 (February 9, 2007), para. 44.

[2] One notable exception is the peremptory norms of international law (jus cogens norms) for which corporations can be held directly accountable.

[3] Patrick Dumberry and Gabrielle Dumas-Aubin, “How to Impose Human Rights Obligations on Corporations under Investment Treaties?”, 4 Yearbook on International Investment Law and Policy (2011-2012), p. 569-600.

[4] Patrick Dumberry and Gabrielle Dumas-Aubin, “When and How Allegations of Human Rights Violations Can Be Raised in Investor-State Arbitration,” 13(3) Journal of World Investment & Trade (2012), p. 349-372.

[5] “Interim Report of the Special Representative of the Secretary-General of the United Nations on the issue of human rights and transnational corporations and other business enterprises,” John Ruggie, E/CN.4/2006/97 (February 22, 2006), 34.

[6] United Nations, Global Compact – The Ten Principles (available at http://www.unglobalcompact.org/aboutthegc/thetenprinciples/index.html (last visited April 21, 2012).

[7] United Nations, Global Compact, UN GA Res. 64/223, March 25, 2010.

[8] Dumberry & Dumas-Aubin, supra note 3, p. 589 ff.

[9] Patrick Dumberry and Gabrielle Dumas-Aubin, “The Doctrine of “Clean Hands” and the Inadmissibility of Claims by Investors Breaching International Human Rights Law”, Transnational Dispute Management Special Issue: Aligning Human Rights and Investment Protection (2013), 10(1).