Environmental, Labour and Anti-Corruption Texts in International Investment Agreements: A Large Sample Survey

Abstract

This paper surveys the environmental, labour and anti-corruption texts in a sample of 296 international investment agreements, including bilateral treaties and free trade agreements with investment chapters. The paper’s key findings are: 1) Most countries never include language on environmental labour or anti-corruption concerns in their investment treaties. Twenty-four of the 39 countries do not include any language on such issues in any of their investment treaties. Among the 15 countries that have included such language in one or more treaties, the language covers mainly environmental and labour issues. More recently, anti-corruption concerns and, very rarely, human rights have been mentioned in a few treaties. Treatment of these issues varies from language in Preambles (e.g. Finland and the Netherlands) to substantive and procedural language in provisions, annexes and side agreements (e.g. many North American agreements). 2) Over the past two decades, more countries have been including such language in their investment agreements. The first agreement in the sample that covers such issues is the 1990 Polish-US bilateral investment treaty (BIT). Since the mid-1990s, Canada, Mexico and the United States have accumulated a large stock of agreements that include language on environmental and labour issues. More recently, other countries (Belgium, Finland, Japan) and regional organisations (European Union and European Free Trade Area) have included environmental and labour language in agreements. Anti-corruption language is a more recent innovation – it appears in four US agreements in the sample as well as in three cooperation and partnership agreements (Japan-Philippines and EU-Russia and the Cotonou Cooperation Agreement between the EU and the Africa, Caribbean and Pacific (ACP) countries). 3) The treatment of these issues varies considerably from one agreement to the other. Variations in language show both idiosyncratic influences and substantive differences in approach to issues. Some treaties explicitly refer to relevant international instruments (e.g. Universal Declaration of Human Rights). The sample texts also show that innovations in language in one agreement are often adopted by other countries for use in their own agreements and that this process of mutual influence has resulted in partial harmonisation of texts. 4) Investment treaty language does not keep pace with innovations in the labour, environmental and anti-corruption fields. International initiatives in the environmental, labour and anti-corruption fields have produced a rich array of new international norms.In anti-corruption, six major conventions or protocols have been signed since 1996. Several hundred international environmental agreements have been signed since the Stockholm Conference of 1972 and the International Labour Organisation is active in the development and promotion of new labour norms. This survey shows that, presumably because of the high cost of renegotiating treaties, the language of international investment agreements, is not quickly updated to reflect these developments. The “stickiness” of investment treaty language, juxtaposed with the dynamism of norms development in the other fields, means that investment treaty language can quickly become dated or incomplete. This is a consideration for both users and drafters of investment treaties, with the former needing to consider whether and how new norms can be integrated into the interpretation of older investment treaties and the latter facing the challenge of drafting treaty language that will be robust with respect to the rapid evolution in labour, environmental and anti-corruption norms.