Book Review: Commentaries on Selected Model Investment Treaties (Chester Brown, editor, Oxford University Press, 2013)

Kluwer Arbitration Blog
February 6, 2014

Please refer to this post as: , ‘Book Review: Commentaries on Selected Model Investment Treaties (Chester Brown, editor, Oxford University Press, 2013)’, Kluwer Arbitration Blog, February 6 2014,

In recent years, Oxford University Press has devoted considerable effort to enhancing its transnational law list. In doing so, it has added several titles addressing international arbitration and investor-State topics. Among its more recent offerings is Commentaries on Selected Model Investment Treaties (hereinafter “Commentaries”) edited by Dr. Chester Brown.

Commentaries is substantial. It comprises 895 pages and 18 detailed chapters; each chapter is separately authored and the authors follow a common format to provide their observations and analysis. The authors are well-credentialed persons from government, academia, and private practice. With the exception of the chapter on NAFTA, each chapter addresses a specific country and focuses on Model BIT practices of that specific State. Despite its size, the book is readily searchable. It has comprehensive tables listing the book’s contents (chapter-by-chapter), the cases cited, and the treaty instruments examined. It also has a serviceable index. Common chapter features include general commentary on the history of the State’s BIT policy and an article-by-article discussion of the Model BIT’s principal provisions. All but one chapter ends with a “select bibliography.”

It was not that long ago that, when teaching International Business Transactions or a related course, one might reasonably have allocated only modest time to BITs, preferring instead to feature Friendship, Commerce and Navigation (“FCN”) treaties. After all, with the filing in 1987 of the ELSI case,[fn]Elettronica Sicula SpA (ELSI) (United States v. Italy) 1989, ICJ Rep. 15[/fn] there was much that could be explored in connection with FCNs. In that ICJ proceeding, the United States’ claim against Italy generated myriad discussion topics: claim espousal (and diplomatic protection in general), the exhaustion of local remedies doctrine, the ICJ’s chamber system, and substantive protections such as that entitling an investor to compensation in the event of an expropriation and that prohibiting denials of justice.

In the classroom of 2013, ELSI remains important, although it has been somewhat repurposed; today one might use it to demonstrate the inefficiencies and risks associated with claim espousal, and correspondingly to introduce BIT arbitration. Unlike espousal, investor-State arbitration concentrates risk and control in the dispute’s real parties in interest. It is also a system that has generated such richness for classroom purposes that one course on the topic (or, for that matter, one annual conference) is hardly enough.

Although a professor planning a course in the mid-1990s might be have been forgiven for dealing with investor-State arbitration as a NAFTA-related subject, by the end of the decade or so that followed it had become apparent that NAFTA Chapter 11was to be only a small part of the story. As has been much discussed, States from virtually every region have been named respondents, often several times. Much of the investor-State docket, which now includes well over 500 active or concluded arbitrations,[fn]UNCTAD, Reform of Investor-State Dispute Resolution: In Search of a Roadmap, IIA Issue Note No. 3 (June 2013), at 2.[/fn] has resulted from, among other factors, two structural predicates: a high number of ICSID Convention parties (at present, 149 States) and the large number of sovereign offers to arbitrate found, principally, in BITs. Added to these architectural features have been the tendency of States to regulate on a sector-wide basis and the perception among aggrieved investors that BIT arbitration, for all its flaws, is to be preferred to either espousal or local courts. The result has been an accumulation of cases for which the system was not fully prepared.

Although the Commentaries’ ambit is limited to sixteen countries plus NAFTA, the States selected for inclusion in the book remind us that BIT relations are no longer limited to the original pattern in which a capital exporting country, based on its model text, forms a BIT with a country seeking foreign direct investment. While, as one would expect, the book has chapters on the programs of Canada, Italy, France, Germany, the Netherlands, the United Kingdom and the United States, so too are there chapters analyzing the models adopted by inter alia China, Columbia, Korea, Latvia, and Russia.

Despite the potential for bilateral approaches to produce significantly different texts, the broad similarity of the Model BITS in use (to some extent the result of inter-State cross-pollination) has instead promoted relative homogeneity. That broad uniformity has in turn generated and reinforced stakeholder expectations and, some would say, has come to evidence customary international law. The recurrent textual patterns in place allow investors to expect promises of national treatment, most favored nation treatment, fair and equitable treatment, full protection and security, and compensation in the event of measures equivalent to a taking. As the Commentaries also confirms, many investment treaties also preclude certain kinds of performance requirements and require host States to observe the promises they make with respect to an investment (an “umbrella clause” undertaking). Several BITs also establish transparency requirements of the kind that featured, controversially, in the Metalclad case.[fn]See Judgment of the British Columbia Supreme Court on the set aside request of May 2, 2001, para. 70.[/fn]

As one might expect, successive generations of BITs also demonstrate an awareness of the divergent arbitral decisions now in the public domain. The resulting trend is for the texts to contain more detail than earlier models. The substantive impact of the added precision has generally been to restrict theories of recovery. Thus, in many of the more recent texts, fair and equitable treatment is circumscribed by custom (it thus is not merely an autonomous treaty standard informed largely by ordinary meaning of “fair” and “equitable”) and, in some models measures tantamount to expropriation are said (“except in rare circumstances”) not to include regulatory measures aimed at, e.g., health, safety or environmental protection when designed and applied in a non-discriminatory manner.[fn]This clarification appeared in the US Model BIT of 2004, and is found in the current, 2012, US Model and in several models of several other States. Interestingly, some States have embellished the basic formula set forth in the U.S. 2004 model by suggesting what is meant by “except in rare circumstances.” See Rivas, Colombia, reprinted in Commentaries, at 241 (reprinting at 224 article VI of the Colombian model “… such as when (measures) are so severe in light of their purpose that they cannot be reasonably viewed as having been adopted in good faith….”).[/fn]

As a review of investor-State awards indicates, however, differences in BIT texts remain important. Traveling back and forth among the book’s chapters, one encounters some stark line-drawing. The chapter on Colombia’s BIT program contains a section titled “Rejection of the Umbrella Clause”. Dr. Rivas observes: “[a]s a strict policy matter, the Model does not include an ‘umbrella clause’.”[fn]See Rivas, at 241.[/fn] Umbrella clauses, by contrast, are standard in the Chinese and Austrian models. Indeed, the Austrian umbrella provision is distinctive; it adds a helpful detail to the usual “shall observe any obligation…,” explaining: “This means, inter alia, that the breach of a contract between the investor and the host State or one of its entities will amount to a violation of this treaty.” [fn]Id. at 35 (quoting article 11 of the Austrian Model).[/fn]

Some differences among BIT texts are more subtle. For example, the Colombian Model BIT, while having much in common with other BITs, adds a nuance to its treatment of full protection and security. Article III.4, of Colombia’s Model BIT, while promising fair and equitable treatment in accordance with customary international law, circumscribes its undertaking with respect to full protection and security: “The full protection and security standard does not imply…a better treatment to that accorded to nationals of the Contracting Party where the investment has been made.” [fn]See Rivas, at 185, 216-17.[/fn] That is an unusual caveat when juxtaposed against what seems otherwise to be the Model BIT’s general endorsement of the international minimum standard.

Another example, among dozens of subtle but purposeful word choices, is seen in essential security provisions found in certain models such as that used by the United States. Article 18 of the U.S. Model BIT provides in relevant part “[n]othing in this treaty shall be construed to…preclude a Party from applying measures that it considers necessary for the fulfillment of its obligations with respect to the maintenance or restoration of international peace or security, or the protection of its own essential security interests.” [fn]U.S. Model BIT (2012), art. 18.[/fn] According to Lee Caplan and Jeremy Sharpe (authors of the U.S. chapter), the language is intended to make clear that the determination of what is necessary is “within the discretion of” the Party taking the measure.[fn]Lee Caplan and Jeremy Sharpe, United States, reprinted in Commentaries, at 813. The requirement that the provision be applied in good faith, in principle, prevents a State from exploiting opportunistically the self- judging element in the provision.[/fn] This explicitness with respect to the provision’s self-judging character is intended to be an improvement upon the formulation found, for example, in the Argentina-U.S. BIT, article XI of which is not clear on the question who—the respondent State or the arbitral tribunal—decides when a measure was “necessary.” [fn]The Argentina-U.S. BIT is reprinted as an appendix in Lucy Reed, Jan Paulsson, Nigel Blackaby, Guide to ICSID Arbitration 311 (2d ed. 2011).[/fn]

Many of the chapters discuss BIT termination provisions. These treaty provisions have become of interest in recent years with many States reconsidering their investment treaty obligations. By the end of 2013, more than 1300 of existing BITs are by their terms eligible for unilateral termination.[fn]See UNCTAD, International Investment Policymaking in Transition: Challenges and Opportunities of Treaty Renewal, IIA Issue Note No. 4 (June 2013) at 1.[/fn] As one can discover by consulting the Commentaries, however, protection often does not end with the termination becoming effective. For existing investments there is often for both States what might be called a “BIT hangover”—a period of ten or fifteen years during which the BIT’s protections continue to apply.
The market for the Commentaries includes law students, lawmakers, diplomats and arbitration practitioners. Each should find it a fruitful read. The volume provides a wealth of information and thoughtful analysis. Despite having been produced by many different authors from different legal cultures, the chapters are well-integrated and substantive.

Investor-State arbitration has inspired abundant literature, not all of which can be investigated even by the most ardent researcher. Articles, books, and dissertations seem to address every conceivable aspect of the field. With time (and money) ever limited, new books enter the marketplace with a substantial burden. Commentaries discharges its burden easily, and is likely to become an oft-consulted reference by the arbitration community.