Emory Law Journal

Rebalancing Legitimacy and Sovereignty in International Investment Agreements
Julia Hueckel Executive Notes and Comments Editor, Emory Law Journal; J.D. Candidate, Emory University School of Law (2012); A.B., Duke University (2005). I am grateful to Dean Robert Ahdieh for his patience and help in developing the foundation of this Comment. I am also indebted to Alan White and Robert Carlton, whose thoughtful suggestions helped hone my argument and whose encouragement pushed me to constantly improve it. Finally, I would like to thank the editorial staff of the Emory Law Journal, particularly Danny Reach, Andrew McKinley, James Spung, and Amy Dehnel for their precise and thorough editing.

Abstract

In investor–state arbitration, foreign investors can bring states before an arbitral tribunal and claim damages for losses incurred due to prohibited state action. This mechanism for dispute resolution is present in most international investment agreements and has experienced a surge of popularity in the last twenty years. However, increased use of investor–state arbitration has generated increased criticism. The sharpest criticism has come from the state parties themselves. Some question the system’s legitimacy, while others worry that it does too little to protect state sovereignty.

This Comment applies principal–agent theory to respond to these divergent concerns and argues that states must fundamentally change how power is delegated and controlled in international investment agreements. States should draft international investment agreements with clearer, more rule-like provisions where possible and should include interpretative guidelines where rules are not feasible. More specific drafting ex ante will in turn support more effective controls ex post, in the form of a “Review Body.” Together, rule-like provisions and the Review Body will provide clearer and more robust investment law jurisprudence in a way that improves the legitimacy of the system and protects state sovereignty.

Introduction

In investment arbitration, a foreign investor can sue a sovereign state for damages if the state violates an investment agreement with the investor’s home country. While an arbitration provision in an investment agreement represents a state’s consent to arbitration, a number of states have voiced doubt over whether their choice to consent was a wise one. This is particularly so because of the growing tensions between the free-trade system that investment agreements are intended to protect and the rise, and arguable success, of state capitalism.

Indeed, critics doubt the legitimacy of the present system and believe it insufficiently protects their sovereignty. Pessimists need only look to Argentina’s experience for confirmation. In the wake of a major financial crisis in the early 2000s, Argentina implemented several reforms to restore long-term stability. 1William W. Burke-White & Andreas von Staden, Investment Protection in Extraordinary Times: The Interpretation and Application of Non-Precluded Measures Provisions in Bilateral Investment Treaties, 48 Va. J. Int’l L. 307, 309–10 (2008). These reforms adversely affected the investments of many foreign nationals who turned to the international investment agreements (IIAs) signed between Argentina and their home countries for protection. 2 Id. at 310. Using investor–state arbitration provisions in these IIAs, investors sought damages for the losses they had sustained. 3 Id. at 311. By one estimate, these claims could cost Argentina $8 billion. 4 Id. By another, Argentina could be exposed to $80 billion of liability. 5 Id. (citing Wailin Wong, Argentina Treasury Attorney: World Bank Claims Could Reach $80 Billion, Dow Jones Int’l News, Jan. 21, 2005).

Less than ten years later, the U.S. response to the financial crisis of 2008 raised concerns that the United States could be exposed to liability under IIAs, much like Argentina. 6 See Anne van Aaken & Jürgen Kurtz, The Global Financial Crisis: Will State Emergency Measures Trigger International Investment Disputes?, Vale Columbia Center on Sustainable Int’l Investment (Mar. 23, 2009), http://www.vcc.columbia.edu/content/global-financial-crisis-will-state-emergency-measures-trigger-international-investment-dispu (discussing the emergency measures taken by states in the wake of the 2008 financial crisis and their potential implications for investment arbitration). While these worries have proven baseless, some continue to fear that the potential for liability under IIAs could chill other types of regulation. 7Paul Michael Blyschak, State Consent, Investor Interests and the Future of Investment Arbitration: Reanalyzing the Jurisdiction of Investor–State Tribunals in Hard Cases, 9 Asper Rev. Int’l Bus. & Trade L. 99, 117 (2009). Indeed, proceedings have already been brought against the United States for an allegedly biased judicial decision 8 See Loewen Grp., Inc. v. United States, ICSID Case No. ARB(AF)/98/3, Final Award (June 26, 2003), 7 ICSID Rep. 442 (2005) (dismissing a $725 million claim against the United States brought by a foreign investor alleging it had been discriminated against by a Louisiana court when the court admitted evidence of the investor’s foreign status). and against a state regulatory action taken to protect human health and the environment. 9 See Methanex Corp. v. United States, 44 I.L.M. 1345 (NAFTA Ch. 11 Arb. Trib. 2005) (dismissing all charges after six years of proceedings in which Methanex, a Canadian producer of methyl tertiary butyl ether (MTBE), sought $970 million in damages from the United States because California had banned MTBE, asserting that it had contaminated drinking water and endangered human health and the environment). Because an average claimant seeks $300 million or more in damages, claims brought against the United States have caused much concern. 10Susan D. Franck, Development and Outcomes of Investment Treaty Arbitration, 50 Harv. Int’l L.J. 435, 435 (2009).

Yet, despite the potential liability, the United States has long supported IIAs because of the belief that IIAs encourage free trade and offer protections for U.S. investors. Proponents of the model believe that, in time, foreign direct investment (FDI) and restraints placed on states by IIAs will lead to an open investment environment and democratic, transparent governance. 11 See id. For example, the Office of the United States Trade Representative states that IIAs “protect private investment, . . . develop market-oriented policies in partner countries, and . . . promote U.S. exports.” 12 Bilateral Investment Treaties, Off. U.S. Trade Representative, http://www.ustr.gov/trade-agreements/bilateral-investment-treaties (last visited Mar. 27, 2012).

IIAs accomplish these goals through provisions that protect investors from certain types of action by the state. Although they are negotiated between and bind only their signatories, IIAs across the globe tend to use similar, sometimes nearly identical provisions. 13 See Jeswald W. Salacuse, The Emerging Global Regime for Investment, 51 Harv. Int’l L.J. 427, 431–32 (2010). Most mandate an open judiciary and fair treatment of investors and set conditions for money transfers into, out of, and within the country. 14 Id. IIAs prohibit expropriation or regulatory actions that significantly devalue an investment. 15 Id. at 435. The majority of IIAs use investor–state arbitration as an enforcement mechanism and allow an investor to seek damages from a host state that has violated an IIA. 16 Id.

Regardless of form, 17One of the primary differences among IIAs is the number of signatories: bilateral investment treaties (BITs) bind two countries, while free-trade agreements (FTAs) are often, but not always, signed between multiple countries. Jason Webb Yackee, Toward a Minimalist System of International Investment Law?, 32 Suffolk Transnat’l L. Rev. 303, 306 (2009). IIAs can be discussed in the same general terms because they impose similar restrictions and obligations on states. 18 See Jason Webb Yackee, Conceptual Difficulties in the Empirical Study of Bilateral Investment Treaties, 33 Brook. J. Int’l L. 405, 415–17 (2008) (discussing the broad similarities and subtle distinctions between BITs); see also Gus Van Harten, Investment Treaty Arbitration and Public Law 27–28 (2007) (discussing certain distinctions between IIAs but concluding that investment treaties are sufficiently similar to be discussed in the same general terms and as part of a global network). For this reason, at least one commentator has referred to the 3000 IIAs as a “global regime.” 19 See Salacuse, supra note 13, at 431. The United States participates in this global system and, as of March 2012, was a signatory to forty-one bilateral investment treaties 20 Bilateral Investment Treaties, Trade Compliance Center, http://tcc.export.gov/Trade_Agreements/Bilateral_Investment_Treaties/index.asp (last visited Mar. 27, 2012). (BITs) and fourteen free trade agreements. 21 Free Trade Agreements, Trade Compliance Center, http://tcc.export.gov/Trade_Agreements/Free_Trade_Agreements/index.asp (last visited Mar. 27, 2012).

Despite the widespread use of IIAs, the system has been criticized both in the United States and abroad for two primary, but not entirely distinct, reasons. First, some believe investment arbitration is a threat to state sovereignty. 22Salacuse, supra note 13, at 434. Second, there has been a growing concern with the legitimacy of the system. 23 See Susan D. Franck, The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions, 73 Fordham L. Rev. 1521, 1584 (2005). These problems can both be tied to the broad, open-ended standards that are commonplace in IIAs.

When a state obligates itself through broad standards, the application of these standards in particular situations is left to the arbitral tribunals—a result that ultimately gives each tribunal wide-ranging interpretative discretion. 24 See Anthea Roberts, Power and Persuasion in Investment Treaty Interpretation: The Dual Role of States, 104 Am. J. Int’l L. 179, 179 (2010) (“As investment treaties create broad standards rather than specific rules, they must be interpreted before they can be applied. Investor–state tribunals have accordingly played a critical role in interpreting, hence developing, investment treaty law.”). As a consequence, a particular state action that was not contemplated when an IIA was drafted could be found to violate the IIA. 25 Van Harten, supra note 18, at 123. Thus, the state may find its sovereignty constrained in a way that it does not believe it consented to when entering into the IIA. 26 Id. This perceived threat to sovereignty alone is enough for some states to question whether the system is beneficial.

Yet the interpretative discretion afforded tribunals has also led to accusations that the system is not legitimate. In the juridical sense, legitimacy is associated with (1) impartiality, (2) certainty, and (3) accountability. Investment arbitration suffers in each of these areas. Certain critics believe that arbitrators favor the interests of investors from developed countries over developing states’ sovereign prerogatives. 27Franck, supra note 10, at 436. Only investors can initiate arbitration proceedings, and some have argued that this makes the tribunals favor investor interests because the tribunals depend on investors for work. 28 Van Harten, supra note 18, at 172.

Moreover, unbound by precedent, arbitrators have interpreted the same standard inconsistently in different awards, 29Charles N. Brower & Stephan W. Schill, Is Arbitration a Threat or a Boon to the Legitimacy of International Investment Law?, 9 Chi. J. Int’l L. 471, 473 (2009). sometimes exposing states to extreme liability. 30 See Franck, supra note 23, at 1558–82 (describing in detail inconsistent awards under various BITs). Finally, weak accountability mechanisms for arbitrators have done little to rectify the situation. Increasingly, parties have questioned the system, some to the extent of exiting altogether. 31Suzanne A. Spears, The Quest for Policy Space in a New Generation of International Investment Agreements, 13 J. Int’l Econ. L. 1037, 1040, 1043 (2010).

The factors laid out above also highlight an inherent tension in investment arbitration. For the system to be legitimate, it must respect the sovereignty of state participants. Yet, to be legitimate, the arbitral tribunals must remain independent of the states as they resolve individual disputes. Thus states cannot protect their sovereignty without simultaneously considering the effects of their actions on the legitimacy of investment arbitration.

This Comment argues that these sensitive problems can be remedied by addressing the existing structural weaknesses in the way states delegate and control the power they transfer to tribunals. Specifically, this should be done through a concerted move away from open-ended standards. Instead, negotiators should look to more specifically drafted rules and interpretative guidelines to control arbitration outcomes ex ante, while relying on a “Review Body” to monitor awards ex post.

By drafting rules and interpretative guidelines ex ante, states can (1) exercise greater control over the arbitral tribunals, (2) limit unpredictable outcomes for both investors and states, (3) establish a necessary foundation for establishing effective controls ex post, and (4) improve perceptions of legitimacy by providing a transparent system where players can be held accountable.

In addition, the proposed Review Body will monitor the tribunals to ensure that they consistently and correctly apply the rules and interpretative guidelines provided ex ante. Using both of these tools, states can restore legitimacy to the investor–state arbitration model and simultaneously protect their sovereign interests.

Accepting this argument requires rejecting the commonly held belief that standards in arbitration agreements cannot be narrowed. Accordingly, this Comment analyzes the way lawmaking costs are distributed under the present system and argues that changes in the global economy and experience with investment arbitration have made standards the more cost-effective option.

To begin the discussion, Part I sets up the struggle between legitimacy and sovereignty in investor–state arbitration. It begins by discussing the structural weaknesses of the system. It shows how broad standards have led to inconsistent arbitral awards that undermine both the legitimacy of the system and the sovereignty of participating states. Moreover, it explains how the lack of precedent and appellate mechanisms has exacerbated these issues.

Part II explains the political aspects of the problem. It recounts the history of IIAs and examines the common political-economic assumptions on which IIAs were built. It then shows how changes in the global economy have shaken some of these assumptions and put pressure on the current system.

Part III suggests a framework for resolving the competing problems of legitimacy and sovereignty. It first uses principal–agent theory to lay out options for control of arbitral tribunals. It then explains how lawmakers typically analyze the cost of these various options. Because investor–state arbitration contains variables that necessarily change the typical analysis, Part III concludes by modifying the traditional cost analysis and retroactively applying it to the present system of investor–state arbitration.

Finally, Part IV applies this framework and suggests a new way for states to delegate and control the interpretative power given to arbitral tribunals. It first argues that changes in the global economy have shifted the way costs are felt in investment arbitration such that rules are now the more cost-effective way to delegate power. It then shows how a combination of ex ante control in the form of rules and interpretative guidelines, and ex post control in the form of the Review Body will help restore legitimacy and sovereignty to investor–state arbitration.

I. The Struggle Between Legitimacy and Sovereignty

Investor–state arbitration provisions are the most controversial elements of IIAs. If a state violates a provision of an IIA, a private investor, if harmed, can bring a claim against that state. Thus, arbitration provisions mean that IIAs are not only instruments of public international law, but that they also function to enforce private rights. 32 See Roberts, supra note 24, at 183 (“Viewing investment treaty arbitration solely through a public international law, state-to-state prism is unsatisfactory because investment treaties create reciprocal rights and duties for the treaty parties and rights for nonstate actors (investors).”).

This duality of purposes implicates potential conflicts of interest. States must draft provisions that simultaneously protect their sovereign regulatory powers and the investments of their citizens abroad. 33 Id. at 182. Therefore, the key question for states becomes how to delegate power to a tribunal to protect both of these competing interests.

Another way of looking at this question is by asking how states can create a legitimate system that binds them to their IIAs without overly impinging on their sovereignty. Legitimacy and sovereignty are especially complicated because they seem to pull in opposite directions; actions that tend to improve the legitimacy of the system ultimately impinge on states’ ability to protect their sovereignty. 34Tai-Heng Cheng, Power, Authority and International Investment Law, 20 Am. U. Int’l L. Rev. 465, 500–01 (2005).

As a result, the present system was structured to protect state sovereignty at the expense of legitimacy. This decision manifests itself through the limited powers granted to a tribunal and its decisions. Ironically, these structural choices have not had their intended effect because the present structure of arbitration undermines both the legitimacy and the sovereignty of the system.

Before fleshing out the discussion of legitimacy and sovereignty, section I.A provides a brief overview of the mechanics of investment arbitration. Section I.B delves more deeply into the structure of investment arbitration and illustrates how this structure fuels tensions between legitimacy and sovereignty.

A. The Mechanics of Investor–State Arbitration

The unwillingness to cede too much power to a supranational institution is reflected in many of the structural features of investment arbitration and particularly in the composition of tribunals. First, arbitral tribunals are convened on a case-by-case basis when an investor believes that a state has violated an IIA. Once the arbitration proceeding is concluded, the tribunal is dissolved.

The selection of the arbitrators for a particular dispute is governed by the arbitration rules of the forum. 35Franck, supra note 10, at 443. The forum is either agreed upon in advance in the treaty or chosen by the investor when submitting a claim. 36 Id. Arbitration rules are generally written by one of the numerous arbitration centers in the world. 37 Id. The most frequently used rules are those promulgated by the World Bank’s International Centre for Settlement of Investment Disputes (ICSID), the United Nations Commission on International Trade Law (UNCITRAL), and the Stockholm Chamber of Commerce. 38 Id. Under ICSID rules, each party to the dispute chooses an arbitrator. 39 Id. The third arbitrator is chosen by ICSID in the event that parties cannot reach an agreement. 40 Id. Under UNCITRAL rules, the third arbitrator is chosen by the party-appointed arbitrators. 41 Id. In any event, the arbitrators are chosen for that dispute alone.

Another important control on tribunals is their limited mandate. 42 See Andrea K. Bjorklund, Mandatory Rules of Law and Investment Arbitration, 18 Am. Rev. Int’l Arb. 175, 191 (2007). Tribunals are only empowered to decide the matter before them. 43Benedict Kingsbury & Stephan Schill, Investor–State Arbitration as Governance: Fair and Equitable Treatment, Proportionality and the Emerging Global Administrative Law 4 (N.Y. Univ. Inst. for Int’l Law & Justice, Working Paper No. 2009/6, 2009), available at http://iilj.org/publications/documents/2009-6.KingsburySchill.pdf. In addition, tribunal awards are not binding on later tribunals. 44Salacuse, supra note 13, at 467. However, some commentators have suggested that there is an informal system of precedent in investment law because tribunals have been known to cite to other parties’ submissions and tribunal awards as a matter of practice. 45 See id. (“[T]he approximately 300 decisions that have emanated from tribunals are consistently cited by lawyers and other tribunals and have a powerful influence on the making of future regime decisions.”); see also Tai-Heng Cheng, Precedent and Control in Investment Treaty Arbitration, 20 Fordham Int’l L.J. 1014 (2007) (arguing that arbitrators have made a concerted effort to reason and write awards in a way that resembles an informal system of precedent); W. Mark C. Weidemaier, Toward a Theory of Precedent in Arbitration, 51 Wm. & Mary L. Rev. 1895 (2010) (comparing investor–state arbitration, commercial arbitration, and labor arbitration, and noting that precedent operates most strongly in the investor–state model of arbitration). Professor Cheng in particular argues that this development is the arbitrators’ response to criticisms that awards were poorly reasoned or inconsistent. 46 See Cheng, supra note 45, at 1015. However, arbitrators can only do so much to legitimize a system that is intentionally designed to limit their power.

B. Competing Concerns of Legitimacy and Sovereignty

The present structure of investment arbitration has numerous features that contribute to the dueling problems of legitimacy and sovereignty. Legitimacy is discussed in various contexts and comprised of many components. 47For example, legitimacy is related to “determinacy and coherence,” as well as to “fairness[ and] accountability.” Franck, supra note 23, at 1584. It is associated with predictability, accountability, fairness, the willingness of players to use and be bound by the system, 48 Id. and the independence or impartiality of the tribunal. 49 See Laurence R. Helfer & Anne-Marie Slaughter, Toward a Theory of Effective Supranational Adjudication, 107 Yale L.J. 273, 312–13 (1997) (identifying independence as a key criterion of the legitimacy and effectiveness of a supranational adjudicatory body); Ruth Mackenzie & Philippe Sands, International Courts and Tribunals and the Independence of the International Judge, 44 Harv. Int’l L.J. 271, 274 (2003) (“Independence of the judiciary is a critical factor . . . .”). Here, certainty, accountability, and impartiality are used as three key indicators of legitimacy, although other factors weave through the discussion as well.

The “legitimacy crisis” in investor–state arbitration involves all of these factors and has been blamed partially on the vague provisions that make up most IIAs. 50 See Franck, supra note 23, at 1584–85. Typical provisions include promises (1) “not to discriminate on the basis of nationality,” (2) “to provide full protection and security to an investment,” or (3) “to treat investments fairly and equitably.” 51 Id. at 1530–32. IIAs contain broad standards because states could not foresee specific situations that would lead to arbitration. 52 Id. at 1584. Because standards give tribunals broad interpretative discretion, actors are unable to determine the content of law before acting. 53 See Yackee, supra note 18, at 418–19. In close cases, it will be up to the tribunal to decide whether an action that arguably falls within the ambit of a provision is prohibited by that provision.

For example, a state may be unable to discern whether a regulatory taking would be considered unfair or inequitable based on the language of the investment treaty alone. Another question may be whether the promise to provide “full protection and security to investments” covers only physical assets or also includes securities. If it applies to investments that are not physical assets, what does that protection entail?

Previous awards provide little guidance. These awards are notoriously inconsistent for several reasons. 54 See Blyschak, supra note 7. To begin, tribunals are not bound to the decisions of earlier tribunals. 55Tai-Heng Cheng & Robert Trisotto, Reasons and Reasoning in Investment Treaty Arbitration, 32 Suffolk Transnat’l L. Rev. 409, 425 (2009). Even if a tribunal were to look to another award for guidance, awards are not always well reasoned. 56Some IIAs contain a reasoning requirement as a control. Id. at 413. However, because that requirement is not explicitly defined in IIAs, it often fails to achieve its goal. See id. at 414. It may be difficult to discern why a particular tribunal found the way it did and if that logic should apply in later cases. Moreover, later ad hoc tribunals may not necessarily view the same provision in the same way. 57Franck, supra note 23, at 1546. The result is a patchwork of awards that do not form a coherent whole. As such, the system lacks certainty, which in turn undermines its legitimacy.

The unpredictability of the law and inconsistency of outcomes are exacerbated by the weakness of accountability mechanisms. 58 See Thomas M. Franck, The Power of Legitimacy Among Nations 61 (1990). There is no appeals system to correct or clarify mistakes. 59Franck, supra note 23, at 1546. Decisions may be overturned for limited reasons under ICSID rules. 60 See id. at 1547–48. However, this procedure only provides relief to the parties of a particular arbitration and does not clarify the law or hold tribunals accountable. 61 Id.

Moreover, the very fact that IIAs are drafted as standards makes it difficult to hold arbitrators accountable. 62 Franck, supra note 58, at 60–61. Where a provision has multiple plausible interpretations, it is difficult to assess whether a tribunal is doing its job faithfully or acting in a biased manner. 63 See Yackee, supra note 18, at 418–19. Even where the tribunal is acting honestly, it may still be accused of bias if the losing party is unable to discern whether the award was actually decided properly. 64 See Franck, supra note 58, at 64 (noting that institutions may be perceived as illegitimate if they do not “function in accordance with ascertainable principles of right process” (emphasis omitted)). This will often be the case when an award is poorly reasoned or lacks an explanation of reasoning altogether.

Likewise, the ability to select arbitrators gives parties some control over the composition of the tribunal. 65 See Franck, supra note 10, at 443. It allows parties to choose and reward arbitrators who have historically shared their views. 66 See Van Harten, supra note 18, at 167–74. However, it does not necessarily provide a mechanism for ensuring that tribunals stay within their mandate or faithfully interpret IIAs. 67 See supra notes 57–58 and accompanying text.

Some argue that the selection of arbitrators serves to align arbitrator and party interests because arbitrators rely on parties for repeat nominations. 68Roberts, supra note 24, at 197–98. Therefore, the arbitrators are dependent on the parties; however, there is some debate as to the true effects of this control. 69 See Eric A. Posner & John C. Yoo, Judicial Independence in International Tribunals, 93 Calif. L. Rev. 1, 26–27 (2005) (asserting that selection by parties to a dispute makes adjudicators dependent on those parties). Some commentators believe that this gives the arbitrators the incentive to decide issues impartially because state parties will seek neutral tribunals. 70 Id.; see also Roberts, supra note 24, at 197–98 (arguing that the common accusation that arbitrators will seek to create more work for themselves as a group (and thus act in a way that favors investors) must be balanced against the arbitrators’ awareness that reasonable and impartial interpretations are more likely to keep the system alive and expanding). Others believe that arbitrators favor investor interests because investors are the only ones who can initiate arbitration and will only do so if they anticipate a favorable result. 71 E.g., Van Harten, supra note 18, at 172–73.

Discussions of arbitrator incentives are closely linked to criticism of tribunal impartiality. Tribunal bias and lack of independence are often the most powerful critiques of the legitimacy of investor–state arbitration. Many developing countries believe that tribunals favor investors and neoliberal economic policies. 72 See Spears, supra note 31, at 1040–41 (“Most of the IIAs in force today were drafted . . . by . . . capital-exporting states that subscribed to a market fundamentalist or ‘neo-liberal’ version of economic liberalism . . . . [N]eo-liberals do not envision any role for the regulatory state beyond the establishment and protection of property rights—not even for the correction of market failures, the risk of which they discount.”). They also argue that tribunals fail to give sufficient weight to state regulatory needs. 73 See Blyschak, supra note 7, at 116–17.

These arguments overlap with concerns about sovereignty. Because states are “the principal actors in the international legal system and . . . the guardians of the public interest within their own borders,” it is important that they have the ability to protect their citizens. 74Charles H. Brower II, Mitsubishi, Investor–State Arbitration, and the Law of State Immunity, 20 Am. U. Int’l. L. Rev. 907, 919 (2005). However, some critics argue that IIAs impair this core state function. 75Blyschak, supra note 7, at 117. Developing countries believe that investor–state arbitration has become a “sword used by . . . investors to attack legitimate government regulation pursued in the public interest.” 76 Id. (quoting Aaron Cosbey, The Road to Hell? Investor Protections in NAFTA’s Chapter 11, in International Investment for Sustainable Development: Balancing Rights and Rewards 150, 151 (Lyuba Zarsky ed., 2005)) (internal quotation mark omitted).

The incredibly high damages that result from arbitration, together with the uncertainty of outcomes and perceived pro-investor bias of the tribunals, have led some to question whether tribunals have a chilling effect on regulation. 77Id. Some awards have been so large 78 See Catherine M. Amirfar & Elyse M. Dreyer, Thirteen Years of NAFTA’s Chapter 11: The Criticisms, the United States’s Responses, and Lessons Learned, 20 N.Y. Int’l L. Rev. 39, 44 (2007) (discussing the Ethyl Corp. v. Canada, 38 I.L.M. 708 (NAFTA Ch. 11 Arb. Trib. 1999), arbitration where Canada settled the case for $13 million and permitted a chemical that was deemed harmful by the EPA to be sold in Canada); Cheng, supra note 34, at 508 (noting that one tribunal awarded investors $270 million). that “a state’s economy may be debilitated.” 79Cheng, supra note 34, at 508. Damages, in effect, operate as a financial limit on the state’s ability to exercise its regulatory power.

Reforms to the system of investor–state arbitration must, then, address several goals. First, states must feel satisfied that their sovereignty will not be impinged in unpredictable ways. In other words, states must know the type and extent of limitations that an IIA will place on their sovereignty during the negotiation phase. Next, states need to believe in the legitimacy of the system. This can be addressed by improving the logical consistency of awards and by creating structures that foster the independence and impartiality of tribunals. Finally, the system needs to remain true to its original goals of protecting investors and encouraging good governance.

II. The History and Development of International Investment Agreements

To address the concerns raised above, it is important to understand why investment agreements and investment arbitration developed as they did. Indeed, many of the problems plaguing investment arbitration are a function of the common assumptions under which it developed. As mentioned above, IIAs are relatively similar. 80 See supra notes 13–16 and accompanying text. Although each is negotiated by and binds only the parties to the agreement, IIAs share a common history and structure. 81 See Salacuse, supra note 13, at 432–34. In general, IIAs were premised on the economic conditions present during the second half of the twentieth century. 82 See id.

At that time developed countries assumed that (1) free and open trade was the path to economic success and (2) FDI would flow predominantly from developed countries into developing ones. The present model of investment arbitration reflects these views.

However, economic changes in the last ten years have greatly undermined both of these assumptions. Accepting that certain assumptions underlying investment arbitration are no longer valid is critical. This shift both allows and encourages necessary changes that were once regarded as impossible. To better explain this shift, this Part first traces the development of IIAs and investment arbitration, and then examines how IIAs operate in the modern economy.

A. International Investment Agreements: A History

The present system of investment agreements has its roots in the desire to protect investors whose home states were in the process of decolonization. 83 See id. at 437. As colonization receded, leaders in developing (or capital-importing) states required investors to be subject to local laws and remedies. 84 Van Harten, supra note 18, at 16–17. These requirements subjected investors to changing regulations and forced them to seek protections in local courts. 85Salacuse, supra note 13, at 436–39. Fearful of discriminatory measures, expropriation, and regulatory instability, investors sought new assurances that their investments would be safe in the developing world. 86 Id. at 436–37.

Initially, accusations of wrongful treatment of an investor by a state were dealt with through state-to-state arbitration. 87 See Van Harten, supra note 18, at 18. An investor would complain to its home state, which could then choose whether to initiate an arbitration proceeding on the investor’s behalf. 88 Id. While this provided some solace, investors’ interests could be overshadowed by competing political interests in their home states. 89 Id. Furthermore, claims brought by weaker states were sometimes ignored. 90 Id.

Ultimately, this arrangement proved unsatisfactory, and investors pressured capital-exporting states to create a system that would better protect their interests. 91 Id. at 17. Despite the strong impetus, several attempts at multilateral agreements on international investment failed. 92 See id. at 19–21 (discussing the numerous failed attempts to create a multilateral regime for international investment from the 1929 Draft Convention on the Treatment of Foreigners, to the multilateral investment code in the International Trade Organization, to the 1948 Havana Charter, to the Abs–Shawcross Draft Convention on Investments Abroad in 1959, to the 1967 Draft Convention on the Protection of Foreign Property of the Organisation for Economic Co-operation and Development). Capital-importing states felt that the current state of international law reflected Western experiences and priorities. 93Salacuse, supra note 13, at 434. They viewed investment treaties as a modern version of imperialism and banded together to oppose any attempt to create a multilateral agreement based on Western views of investment law. 94 See Van Harten, supra note 18, at 16.

Ironically, this opposition did not prevent the explosion of BITs. 95 See Salacuse, supra note 13, at 440. Professor Salacuse further argues that there may have been valid interests at stake for capital-exporting states. These include “(1) encouraging foreign investment in capital-importing countries; (2) relationship building; (3) economic liberalization; (4) encouraging domestic investment; and (5) improving governance and strengthening rule of law.” Id. at 440–41. Bilateral negotiations succeeded where multilateral negotiations failed for several primary reasons. First, negotiating a bilateral treaty is more straightforward because there are only two sets of competing interests, while the number of states involved in multilateral agreements can require a much more complicated and nuanced juggling of interests. 96Roberts, supra note 24, at 196. Second, the possibility of obtaining a special economic relationship with a capital-exporting state can be a more enticing incentive for negotiation in a bilateral setting. 97Salacuse, supra note 13, at 442. In addition, developing states cannot seek the support of other similarly situated actors when negotiating in bilateral settings. 98 See id. at 440. Once some IIAs were signed, other states may have believed they would be less competitive if they failed to sign a comparable treaty. 99 Van Harten, supra note 18, at 43. Thus, bilateral treaties succeeded where multilateral treaties had failed. 100 See id. at 38–44.

Indeed, during the fifty-year period between 1959 and 2009, over 3000 investment treaties were put into effect. 101Salacuse, supra note 13, at 429–30. As mentioned above, IIAs were initially signed between capital-importing and capital-exporting states. 102 Id. at 433. The primary motivators behind IIAs were investors, more specifically investors from developed states. Thus, the first IIAs were negotiated under the assumption that developed states needed investor protections, while developing states would compromise sovereignty to attract investment and growth. 103 Id. at 435.

B. International Investment Agreements in the Modern Global Economy

The dynamics present during the development of IIAs are slowly changing. At the end of 2005, 644 treaties had been signed between traditionally capital-exporting countries. 104 Id. at 433–34. The North American Free Trade Agreement (NAFTA), signed between the United States, Canada, and Mexico, is frequently used and raises new questions about how IIAs apply to developed economies. 105 See Roberts, supra note 24, at 224.

Answering these questions is pressing because states are increasingly playing both capital-importing and capital-exporting roles. The rapid growth of emerging markets has changed the power dynamics in the global economy. 106 See id. at 196–97. China’s state capitalism, in particular, presents a powerful alternative to the Western free-market philosophy. 107 See Ian Bremmer, The End of the Free Market: Who Wins the War Between States and Corporations? 161–66 (2010). State capitalism is a term used by Ian Bremmer to describe state-led economies that strategically manage players in their economies for political gain. Id. at 23. In addition, the American brand of capitalism that fueled the expansive growth of the last two decades has been blamed for the economic crisis. 108 Id. at 162. As a result, the United States has lost much of its prestige and ability to influence other nations. 109 Id. at 163. Skeptics are less certain than ever that free-market capitalism will bring the success and growth it promises. 110 See id. at 164.

At the same time, a number of developing states, such as the BRIC countries (Brazil, Russia, India, and China), 111The term BRIC was coined in a 2001 report by Goldman Sachs. See Jim O’Neill, Building Better Global Economic BRICs 1 (Goldman Sachs Econ. Research Grp., Global Economics Paper No. 66, 2001), available at http://www2.goldmansachs.com/our-thinking/brics/brics-reports-pdfs/build-better-brics.pdf. developed dramatically during the first decade of the 2000s. 112 See Tetsufumi Yamakawa et al., BRICs Lead the Global Recovery, BRICs Monthly (Goldman Sachs, N.Y.C., N.Y.), May 29, 2009, at 1, available at http://www2.goldmansachs.com/our-thinking/brics/brics-reports-pdfs/brics-lead-recovery.pdf. Notably, these countries employ varying degrees of interventionist economic policies ranging from subsidies to full-scale state control of certain corporations and investment funds. 113 See Bremmer, supra note 107, at 164–65. Ian Bremmer, president of Eurasia Group, 114According to its website, “Eurasia Group is the world’s leading global political risk research and consulting firm.” What We Do, Eurasia Group, http://www.eurasiagroup.net/about-eurasia-group (last visited Mar. 27, 2012). Eurasia Group’s “analysts monitor political, economic, social, and security developments . . . . They also examine cross-border issues such as trade, energy and other commodities, financial regulation, climate change, and global health.” Id. calls this state capitalism “a system in which the state dominates markets, primarily for political gain.” 115 Bremmer, supra note 107, at 43. While most governments intervene in markets in some way, the distinction between state capitalist and free-market capitalist is one of degree and purpose. 116 See id. In a state-capitalist economy, “governments play the role of lead economic actor,” while in free-market economies, there are “established legal limits on the state’s ability to regulate the actions of private companies and investors.” 117 Id. at 46.

Based on this definition alone, it is easy to see how IIAs that embrace limits on state action come directly in conflict with economic systems that rely on state action. Moreover, for those skeptical of IIAs, the success of state-capitalist economies undermines the idea that free trade is necessary for growth.

These changes in the global political economy have strong implications for investment treaties. First, many developing countries are no longer willing to sacrifice their sovereignty in IIAs, because they believe the costs outweigh the purported trade benefits. Already several developing countries have become so distrusting of investor–state arbitration that they have chosen to abandon their investment-treaty regimes. 118Salacuse, supra note 13, at 469–70. Bolivia and Ecuador withdrew from the ICSID in 2007 and 2010, respectively. Id. at 469. Both plan on renegotiating a number of their IIAs, but Ecuador went as far as to leave nine IIAs. Id. at 470. Second, the growth of developing economies means that FDI will increasingly flow in both directions. Traditionally capital-exporting states will find themselves as host countries, meaning that they may also begin to see IIAs as threats to sovereignty.

Thus, to enter into or renegotiate effective investment treaties, proponents of free trade will have to not only convince potential partners that these treaties and their arbitration clauses are legitimate but also address criticisms from their domestic constituencies, as worries about sovereignty continue to grow.

III. Choices for Control

The problems discussed above can be solved by implementing new structural controls in how IIAs delegate power to arbitral tribunals. To illustrate how these controls might operate, section III.A lays out the various choices for control as envisioned by principal–agent theory. Section III.B then examines these choices from a cost-oriented perspective. Finally, section III.C analyzes the present structure of investor–state arbitration through these frameworks.

A. Principal–Agent Theory

A common approach to examining delegations of power is principal–agent theory. Under principal–agent theory, a principal delegates powers to an agent because the agent is able to better resolve a problem than the principal would be. 119Mark Thatcher & Alec Stone Sweet, Theory and Practice of Delegation to Non-Majoritarian Institutions, 25 W. Eur. Pol. 1, 4 (2002). In the case of investor–state arbitration, the agent (the arbitral tribunal) ensures that state parties act in accordance with the commitments enumerated in IIAs. 120 See id. (noting that the agent “resolve[s] commitment problems . . . [by] work[ing] to enhance the credibility of promises made . . . between multiple principals”). Because a commitment in an IIA may run counter to a state’s interests at a given moment in the future, a state cannot be trusted to abide by its commitments without the enforcement of arbitral tribunals.

The principal can only reap the benefits of a principal–agent relationship by granting the agent power. 121 Id. This power can best be understood as the freedom of the agent to choose the best outcome within the range of outcomes permitted by its mandate—in short, discretion. 122 See id. at 4–5. Giving the agent too much discretion risks that the agent will select outcomes that are unsatisfactory to the principal. 123 Id. at 4. For example, if the point of creating investor–state arbitration tribunals is to allow them to enforce investors’ rights against encroachment from states, then granting states too much control over how those rights are enforced will likely defeat the purpose of convening the tribunal. However, if the agent does not have enough discretion, the benefits of the agency relationship may not be worth the costs. 124 See id. at 4–5. Thus, to ensure the proper amount of control and benefit, the principal must decide “what powers to delegate and what institutional control mechanisms to craft.” 125Jonas Tallberg, Delegation to Supranational Institutions: Why, How, and with What Consequences?, 25 W. Eur. Pol. 23, 28 (2002).

For example, a common means to control an adjudicatory body’s discretion is through rules that limit the decision maker’s role to factual questions. As Professor Frederick Schauer acknowledged, “A decision-maker instructed to make decisions according to a set of rules is thereby instructed not to consider certain facts, certain reasons, and certain arguments. We now understand that such instructions to ignore the otherwise relevant are usefully seen as withdrawals of decisional jurisdiction . . . .” 126 Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule-Based Decision-Making in Law and in Life 158 (1991) (footnote omitted). According to Professor Schauer, this withdrawal of “decisional jurisdiction” occurs because parties want to “narrow the range of permissible variance in decision” or because they distrust decision makers. 127 Id.

Another way to look at rules is as legal commands that “give content to the law . . . before . . . individuals act.” 128Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557, 560 (1992) (emphasis omitted). When law is conveyed as a rule, the law promulgator controls the content. 129 See id. at 559–60. The law enforcer is thus limited in her discretion. 130 See id. at 609. For example, a rule that says, “you must be eight years old to get on the Ferris wheel,” does not permit the enforcer to allow someone who is seven on the ride. Even if the reason behind the rule is that children should be a certain height and weight to safely enjoy the ride, the enforcer cannot allow a tall seven-year-old to ride with older friends.

Rules are traditionally contrasted with standards. Standards are broad legal commands whose content is supplied ex post. 131 See id. at 560. Generally, law enforcers provide standards with content when faced with unique factual situations. 132 See id. at 609. Standards thus give law enforcers the power to specify the content of the law based on their own discretion. 133 Id. at 570. Contrast the rule above, supra note 130, with a standard that says, “you must be tall enough to safely ride the Ferris wheel.” Under these circumstances, the ride operator could allow our precocious seven-year-old on the ride. The risk in this situation is that the ride operator is not qualified to make the decision and would mistakenly allow a child who is too small on the ride.

For the purpose of explanation, standards and rules are described as two alternative, all-or-nothing choices. 134 Id. at 561. In reality, there are many choices in between. 135 Id. Laws can contain elements that are both rule-like and standard-like, depending on the lawmaker’s preference. 136 See id. at 568 (stating that, when enacting a law, a government decides whether to promulgate each aspect as a rule or standard). The key is that, where laws are written as standards, the agent has a lot of discretion.

In systems that use precedent, precedent converts standards to rules. 137 Id. at 577. If a particular factual situation is resolved in a certain way under a standard, that resolution then becomes a rule to be applied to all analogous fact patterns. 138 See id. Thus, precedent helps to control inconsistent applications of standards and limits the range of the enforcer’s discretion to only one determination. 139 See id. at 611. However, if the system does not allow for precedent, each new decision calls for a new assessment. 140 See id. at 577.

Once power has been delegated to the agent, the principal can exercise some control over the agent ex post. Principal–agent theory emphasizes two types of ex post control—monitoring and sanctions. 141Tallberg, supra note 125, at 28. “Monitoring affects agent behaviour by making it less likely that shirking will go unnoticed. Sanctions, in turn, encourage the agent to fulfil its functions faithfully, by raising the costs of non-compliance.” 142 Id.

Traditionally, the judiciary operates within a number of structural ex post controls that serve to both monitor the system and sanction participants. For example, domestic courts have a developed appellate system. 143 See Cheng, supra note 45, at 1021. An appellate system allows errors to be corrected and leads to consistent law. 144 Id. Accordingly, lower court judges tend to be constrained to certain outcomes by a fear of being overruled.

Precedent is a strong force in common law systems. 145 See id. at 1017–18 (indicating that judges in common law systems need compelling reasons to deviate from precedent). Jurisprudence that has developed over years also conveys certain norms that shape judges’ thinking. As Professors Posner and Yoo put it, “[C]itizens who become judges share most of the values and expectations of the political community.” 146Posner & Yoo, supra note 69, at 13–14. Those who act radically outside these norms will suffer reputational harm. 147 See id. at 14 (listing potential ramifications for judges who do not share the values of the political community). In this way precedent serves an informal sanctioning function.

Both ex ante and ex post controls are weak in the present system of investment arbitration. Broad standards delegate significant interpretative power to the agent tribunals. In addition, mechanisms for monitoring the tribunals and states’ ability to sanction arbitrators are weak. Without an appellate mechanism, precedent, or strong reputational incentives, the present structures fail to ensure that the agent tribunals are accountable to their principals.

B. The Legal Costs of Control

In considering reforms to the present system, the cost of any change must be factored in. Principals often choose what types of controls to implement depending on cost and the type of principal–agent relationship being crafted. 148 See Thatcher & Sweet, supra note 119, at 4. The following section analyzes the legal costs of ex ante delegation and pragmatic considerations helpful in crafting ex post controls.

Ex ante controls are commonly associated with rules. 149 See Kaplow, supra note 128, at 559–62. Rules are commands whose content is provided by the lawmaker ex ante. 150 Id. Standards, on the other hand, are commands whose content is provided ex post by an interpreting body. 151 Id. at 560–62. According to Professor Louis Kaplow, the choice whether to convey law through standards or rules (or some hybrid) depends largely on cost. 152 See id. at 571–77. In this analysis, “costs include the harm caused by individuals’ acts and legal costs—the costs of promulgating the law, the costs of legal advice sought by individuals, and the costs of enforcement proceedings.” 153 Id. at 572. The analysis highlights three points in time during which costs can be a factor: law promulgation, choices of individuals, and law enforcement. 154 Id. at 568–71.

Professor Kaplow’s cost analysis at each stage can be diluted to three factors: (1) the amount of information available, (2) the cost of obtaining additional information, and (3) the number of times the law will be provided with content. 155 See id. at 568–86. In this shorthand version, information means data or experience with how the law will be applied and how it should guide proper outcomes. 156 See id. at 569.

At the promulgation stage, cost depends on how easily actors can determine the appropriate content for a law. 157 See id. Where the content of the law is not obvious, information gathering and cooperation costs can be high. 158 Id. At this stage, rules tend to be more costly to promulgate than standards because rules require more information and agreement. 159 Id.

The choice-of-individuals stage revolves around whether individuals will choose to learn the content of the law before acting. 160 See id. These costs can vary greatly under both rules and standards. According to Professor Kaplow, “[A]dvice is more costly under a standard.” 161 Id. at 571. Because a standard provides little information as to how the law will be applied, actors will need to seek more information before acting. This may occur frequently or infrequently depending on the situation. However, the choice-of-individuals stage will only contribute to costs if individuals choose to become informed, and this assessment may differ from situation to situation. 162 See id. at 571–77 (explaining different cost scenarios depending on whether individuals choose to become informed under a rule or a standard).

Finally, the enforcement phase centers on how much it costs to determine the sanction for violating the law. 163 Id. at 570. In the case of standards, a violation will be more expensive. 164 Id. The cost of determining the sanction will include the cost of determining the content of the law—which requires an assessment of information costs. 165 See id. at 570 & n.23. Frequency can also drive up costs ex post because the same costly exercise of determining the content of the law may have to be repeated. 166 Id. at 621.

The costs of providing law are generally not the same ex ante and ex post. 167 Id. at 570. Accordingly, the principal will apply this analysis at all three stages and compare costs before deciding whether to constrain the agent’s discretion through rules or standards. 168 See id. at 621. The principal will also weigh ex post controls that may allow the principal to effectively limit the agent’s action. If the principal can create inexpensive standards and subsequently ensure an acceptable set of outcomes through ex post control, she will likely do so.

While Professor Kaplow’s analysis serves as a starting point, the cost analysis is more complicated in the case of investment arbitration. There, two principals must agree on how best to delegate power. The choice will depend largely on power dynamics between the principals and the extent to which their goals, incentives, and perceptions of cost converge. In addition, the dynamics of investor–state arbitration require that a state analyze costs both as a sovereign and as a potential litigant. 169 See Roberts, supra note 24 (describing the dual role of states as both drafters of law and parties to future arbitrations). Moreover, states must weigh both the economic and political costs of a particular choice when determining how to delegate and control power. In many instances, too much ex post control will defeat the purpose of the delegation. 170Tallberg, supra note 125, at 28. This is especially true in the case of adjudicatory bodies, where the traditional presumption is that they “require independence as a prerequisite of their legitimacy and the successful fulfillment of their responsibilities.” 171Jacob Katz Cogan, Competition and Control in International Adjudication, 48 Va. J. Int’l L. 411, 414 (2008).

However, the value of independence in an international adjudicatory body is debated. As noted, independence is considered the hallmark of a strong judiciary in democratic domestic systems, and this notion has been applied to international adjudicatory bodies as well. 172Posner & Yoo, supra note 69, at 13. Some scholars advocate for independence of all international adjudicatory bodies, even if incrementally achieved. 173 See, e.g., Helfer & Slaughter, supra note 49; Mackenzie & Sands, supra note 49. Others have argued that independence is undesirable because “states will be reluctant to use international tribunals unless they have control over the judges.” 174Posner & Yoo, supra note 69, at 7; accord Blyschak, supra note 7, at 99 (advocating a more moderate view than Posner and Yoo, but encouraging a more state-centric approach to arbitration).

These competing viewpoints underscore the tension between independence (legitimacy) and control (sovereignty). 175Cogan, supra note 171, at 414. However, even commentators who advocate independent international adjudicatory bodies recognize that domestic courts enjoy independence due to existing controls that international adjudicatory bodies currently lack. 176 Id. at 414–15.

Developed domestic systems contain institutions and actors whose mandates and voices limit the discretion of the judiciary to an acceptable range of options. For example, in the American system, the notion of separation of powers limits the reach of the judiciary. 177 See Posner & Yoo, supra note 69, at 13. Even in systems that do not espouse such principles, the presence of a centralized lawmaker and law enforcer acts as a check on the judiciary. 178 Id. Because these controls are not built into the investor–state model, states that delegate power to the tribunals must either commit to creating such institutions or find other ways of controlling tribunals.

C. Control Choices in the Current Model of Investor–State Arbitration

By retrospectively applying the modified version of Professor Kaplow’s economic analysis of lawmaking to investor–state arbitration, one can examine the individual factors and perceptions that shaped the conventional structure of IIAs.

The following discussion is somewhat speculative in its nature; however, several factors are more certain. First, states likely had access to limited information when drafting the first IIAs. 179 See Yackee, supra note 17, at 309–10. Next, states could not have predicted the incredible growth of investor–state arbitration. 180 See Van Harten, supra note 18, at 13–44 (describing the development of IIAs and the “explosion” of claims in the mid-1990s). Finally, the power dynamics of the last half of the twentieth century tended to favor the interests of capital-exporting states. 181 See Spears, supra note 31, at 1042.

This power structure was conducive to negotiating investor protections. 182 See Van Harten, supra note 18, at 37–41. Developing countries’ access to Western funds was dependent on accepting a free-market-capitalist model. 183 Id. This was true of funding from institutions such as the International Monetary Fund and the World Bank—and, to some degree, of foreign investment. 184 Id. at 41. The United States and other developed countries were eager to press the free-market model on these capital-starved states to increase opportunities for Western investors. 185 See Salacuse, supra note 13, at 433–34.

Most IIAs were drafted under the assumptions discussed in Part II: investors would originate primarily from developed nations, and developing nations would benefit from FDI in exchange for accepting certain constraints on their behavior. 186 See Andrea K. Bjorklund, Reconciling State Sovereignty and Investor Protection in Denial of Justice Claims, 45 Va. J. Int’l L. 809, 832 (2005). The assumption that investments would flow predominantly in one direction meant that developed states viewed their interests as largely aligned with those of investors, while developing states were more concerned with restrictions on sovereignty. 187Salacuse, supra note 13, at 435. Thus, the global power structure favored investor protections.

In addition, legal costs favored the provision of law ex post. As mentioned, there was a lack of information on how investor–state arbitration would operate in practice. 188 See Yackee, supra note 17, at 309–10. In addition, creating rules without knowledge may not have been favorable because rules tend to be both over- and under-inclusive. 189 See Kaplow, supra note 128, at 586–96.

Moreover, developed states believed that they would rarely be subject to liability under IIAs. 190Spears, supra note 31, at 1042. Accordingly, it is unlikely that they were concerned whether a particular standard helped a state make an informed decision regarding potential liability under investment law before acting. The more powerful bargaining position of these states likely overshadowed the developing states’ concerns about liability. These power dynamics likely discounted the cost attributed at the individual-choice stage.

Finally, determining the content of a standard ex post offered a distinct advantage over establishing a rule ex ante—there would be more information available ex post because the decision of how to apply the standard would be made with reference to a specific factual context. 191 See Kaplow, supra note 128, at 583. Moreover, the cost of providing law ex post was likely underestimated because drafters of IIAs did not predict that investor–state arbitration would occur with such frequency. 192 See Spears, supra note 31, at 1042. As a result, the promise of lower costs ex post compared favorably to the high costs associated with providing content for the law ex ante. On balance, this analysis supports the use of broad standards like those that are present in IIAs today. 193 See id. at 1040; see also Brower, supra note 74, at 917 (discussing the use of broad provisions in investment treaties).

However, the choice to supply the content of the law ex post carries significant consequences. The most glaring of these is that content supplied ex post is not supplied by the states, but rather by the tribunals. 194Kaplow, supra note 128, at 560–65. As discussed in Part II, the choice to transfer interpretative power to the tribunals in this way has resulted in widespread criticism of the investor–state model. Part IV addresses alternative choices for control that resolve the resulting tension between legitimacy and sovereignty.

IV. Control in Investor–State Arbitration: A New Approach

Because criticism of investor–state arbitration is not new, there has already been some effort on the part of states and arbitrators to address problems within the existing framework. These efforts have appeared in several forms, including interpretative guidelines attached to IIAs, 195 See Spears, supra note 31, at 1048 (“[I]nterpretative language [included in newer IIAs] seeks to refine . . . standards and encourage arbitrators to conduct a balancing exercise to determine whether government action taken in response to legitimate regulatory concerns has violated investors’ rights.”). interpretative statements issued by bodies such as the Free Trade Commission (FTC), 196 See Amirfar & Dreyer, supra note 78, at 50. and informal precedent. 197 See supra note 45 and accompanying text.

However, more fundamental changes to the structure of investor–state arbitration are necessary. In particular, states should make a concerted effort to control tribunals ex ante and monitor this narrower delegation through additional controls established ex post. Through the use of strong ex ante and ex post controls, states can cooperate to guide tribunal jurisprudence.

Ex ante controls should take two forms. First, where parties can agree, they should draft more rule-like provisions. Where it is more difficult for parties to agree, they should include interpretative guidelines for tribunals to apply in their decision making. Ex post controls should take the form of a Review Body. This Body would combine some features of an appellate body and some features of an interpretative body, such as the FTC.

For this proposed system to function, arbitral awards would have to be publically available. In the original model of investor–state arbitration, awards could remain private if parties so desired. 198 See Vaughan Lowe, Private Disputes and the Public Interest in International Law, in International Law and Dispute Settlement: New Problems and Techniques 3, 10 (Duncan French et al. eds., 2010). However, the United States has already adopted a position that advocates for transparency in investor–state arbitration and public awards. 199 Id. Therefore, this Comment assumes that awards would be publically available.

The following sections lay out this prescription in greater detail. Because using rules has often been dismissed as infeasible in IIAs, section IV.A applies the modified Kaplow analysis to show that changes in the global economy have shifted the costs and incentives associated with negotiations. Section IV.B then argues that ex ante rules would improve the legitimacy of the system, allow states to better protect their sovereign interests, and allow for more effective ex post controls. Section IV.C discusses the role of the Review Body and how it would further protect the competing interests at stake in investor–state arbitration. Section IV.D offers a comparison between this proposal and other suggested models.

A. Legal Costs Favor Rule-Like Provisions

The use of rule-like provisions in IIAs has generally been rejected because it would require the “mammoth” task of renegotiating treaties. 200Franck, supra note 23, at 1588–89; accord Yackee, supra note 17, at 315 (arguing that correcting tribunal interpretations would be difficult because it would require renegotiating existing treaties). This section argues that a cost analysis supports renegotiation. As discussed above, the initial structure of IIAs was premised on several assumptions that have changed over time. 201Bjorklund, supra note 186, at 832–33. There is now a significant body of information in the form of awards and analyses that can guide states in drafting provisions. 202 See Kingsbury & Schill, supra note 43, at 2. Imbalances in power and wealth, although present, no longer favor capital-exporting states as extremely as they did during the last half of the twentieth century. 203 See Roberts, supra note 24, at 196. Finally, the belief that open markets and FDI are the exclusive paths to development does not carry the global support that it once did. 204 See Salacuse, supra note 13, at 470 (arguing that “[m]any parts of the world have lost faith” in the economic model that drove the IIA boom of the 1990s). In fact, some states are so displeased with the present system that they have chosen to abandon or renegotiate their IIAs. 205 Id. at 469–70. These changes require a revision of the system because they significantly alter how costs are felt and perceived at each stage of the economic analysis.

At the ex ante stage, the collective experience of over three hundred arbitrations 206Kingsbury & Schill, supra note 43, at 2. provides states with substantial information about what kinds of actions may trigger arbitration and with what effects. Despite the inconsistencies of individual arbitrations, the experience of past awards will help states draft more rule-like provisions. Furthermore, mutual concerns for preserving certain state regulatory actions incentivize detailed negotiations up front. 207 See Roberts, supra note 24, at 196 (“These developments increase the prospect that at least some states will appear on both sides of the investment treaty equation and that a convergence of states’ views will thus be facilitated.”).

At the individual-choice stage, the reality that both developing and developed states could potentially face arbitration for regulatory actions 208 Id. has made obtaining information before acting a wise choice. A system that is ruled by broad standards and does not honor precedent makes it very expensive—and even impossible—to obtain information before acting. 209 See Yackee, supra note 18, at 418–19. A lack of information may lead an actor to unknowingly breach a provision. Because the average claimant demands $300 million, the cost of breach may be high. 210Franck, supra note 10, at 435. The fear of incurring high costs may, in turn, chill regulation and carry additional social and environmental costs. 211 See Blyschak, supra note 7, at 117. In short, uncertainty at the individual-choice stage is expensive.

The ex post stage also underscores the need for more rule-like provisions in modern IIAs. As arbitrations have multiplied, so too has the cost of supplying content ex post. 212 See Kaplow, supra note 128, at 621 (“If conduct will be frequent, the additional costs of designing rules—which are borne once—are likely to be exceeded by the savings realized each time the rule is applied.”). Although states are free to look to other IIAs for information about how certain provisions may later be interpreted, tribunals must still analyze each provision with reference to its particular IIA and the drafting states’ intentions. 213 See Blyschak, supra note 7, at 113 (arguing that IIAs “are [a] lex specialis and as such only have effect between the parties”). A tribunal may refer favorably to awards issued under other IIAs, but those awards are persuasive at best. Even if the persuasive value of these awards lessens the cost of obtaining information to a certain degree, 214 See Kaplow, supra note 128, at 614 (“Prior cases—both their outcomes and any written opinions—may reduce the costs and increase the accuracy of legal advice and adjudication even if no ‘binding precedent’ is created.”). This may be less true in a system that produces inconsistent decisions. tribunals must still supply the content of the law and must do so on an increasingly frequent basis. 215 See Franck, supra note 10, at 435 (noting “a four-fold increase” in investment-treaty arbitration within the last decade). As a result, costs at the ex post stage remain high. On balance, this analysis shows that drafting more specific rules ex ante is less costly than continuing to rely on standards that lead to high costs at each stage.

While rule-like provisions are preferable under this general analysis, it is likely that cost will bear out differently for each provision. In other words, for some provisions, the availability of information ex ante and converging incentives would make specific negotiation possible. For other provisions that deal with more contentious, complicated, or previously unknown questions, or where increased flexibility may be necessary, a clear rule may not make sense. 216 See Charles H. Brower, II, Structure, Legitimacy, and NAFTA’s Investment Chapter, 36 Vand. J. Transnat’l L. 37, 88 (2003) (“[T]he use of simple rules to deal with complex matters often produces absurd results at the margins, creating . . . a paradoxical reduction of legitimacy.”).

For example, the question of whether a security was considered an investment for the purposes of “full protection and security” could be answered by a definition. But the question of what types of actions constitute a breach of that provision may be better addressed by a more flexible approach. In the latter instance, states should supply tribunals with interpretative guidelines. 217A similar approach has already been taken by the United States in its 2004 Model BIT, but the expansion of this technique has been controversial. See Franck, supra note 23, at 1588–89; Roberts, supra note 24, at 180. Interpretative guidelines would narrow the tribunal’s discretion to a certain degree by instructing arbitrators how to balance interests. 218 See Spears, supra note 31, at 1048. Yet, interpretative guidelines would also provide tribunals with some flexibility to deal with more complex situations. 219 See id. Moreover, as shown in section C below, interpretative guidelines, like rules, would provide a baseline of review for the Review Body. Thus, where negotiating rule-like provisions is not feasible, states should still use interpretative guidelines issued ex ante to narrow a tribunal’s discretion.

B. The Benefits of Rule-Like Provisions

As shown above, an analysis of cost alone strongly suggests that parties should draft IIAs with more rule-like provisions. Rule-like provisions would make significant strides toward restoring legitimacy to investor–state arbitration, while simultaneously preserving sovereignty. Moreover, rules can more clearly commit states to a framework that promotes FDI. 220 See Franck, supra note 58, at 54 (“[T]he more determinate a standard, the more difficult it is to justify non-compliance.”).

First, rules lead to more predictable law. This is true for both the individual-choice stage and the law-determination stage. 221 See Kaplow, supra note 128, at 585. For actors at the individual-choice stage, rules convey the content of the law. 222 See id. at 577. They let actors know the parameters of acceptable and protected action—such clarity allows actors to work within the law. 223 See id. This is important for states contemplating regulation, as well as for investors planning a business venture.

At the law-determination stage, rules constrain tribunals to the set of outcomes contemplated by states ex ante. 224 See Schauer, supra note 126, at 158–59. Accordingly, awards should fall consistently within this range of outcomes. Consistent outcomes are a hallmark of adjudicatory legitimacy because they show that tribunals apply the law the same way to similar situations. 225 See Franck, supra note 23, at 1589 (asserting that rule-like provisions have textual determinacy that in turn leads to predictable and consistent law). Consistent outcomes also allow parties to rely on the system. 226 Id. at 1584–85.

Certainly, rules drafted ex ante will require states to cede some sovereignty. However, by using clearly drafted rules, states will be able to control the range of the tribunals’ decisions and the extent of sovereignty they give up. 227 See Schauer, supra note 126, at 158. Interpretative guidelines can have a similar effect. By instructing tribunals how to interpret provisions and weigh interests, states can control how much weight tribunals afford matters of state sovereignty. 228Spears, supra note 31, at 1048. As a result, arbitral awards should reflect the carefully negotiated policy decisions made by states ex ante.

C. Ex Post Control: The Review Body

As seen above, the use of rules, instead of broad standards, addresses many of the criticisms leveled at investor–state arbitration. Rules also facilitate ex post controls. This section proposes a framework for ex post control in the investor–state model in the form of a Review Body. The first portion discusses the powers of the Review Body. The next portion discusses structural controls on the Review Body. The final portion shows how the Review Body would address the legitimacy and sovereignty problems faced by investor–state arbitration.

Parties to an IIA should elect a standing Review Body to oversee the awards issued by tribunals. The Review Body would be limited to examining questions of law and would serve three main functions. First, it could overturn awards, but only if the tribunal misapplied the law laid out in the IIA. Second, it could, by consensus, determine that certain awards are precedent. Third, it could amend awards where the outcome was correct but the law was improperly applied.

The combination of these competencies would allow the Review Body the flexibility to expand the law by creating precedential awards when prudent. However, when assigning precedential value would have uncertain or controversial effects, the Review Body would be empowered to merely overturn or correct the awards. By attributing precedential status to certain awards, the Review Body would, in effect, create two tiers of awards: precedential and persuasive.

While the awards that were not designated as precedent would not be binding on the parties, they would still indicate the Review Body’s position on a particular issue. Corrected awards would suggest that the Review Body felt that a particular approach was correct in a certain case, while awards that were left as is would bear the implication of being correctly decided, though nonbinding.

The Review Body would review any award brought before it by one of the disputing parties. If the error were immaterial, the Review Body would issue an amended award that would include a correction. However, if a material error were found, the Review Body would overturn the award and remand it to the tribunal with instructions. Both actions would require a supermajority.

In addition, the Review Body would review all awards for precedential value. Here, the entire Body could agree to assign precedential value to all or part of an award. To determine whether an award should be treated as precedent, the Body would examine the relevant interpretative guidelines and other awards dealing with the same issue. Once an award was deemed precedent, this decision would be binding on all pending and future arbitrations. When amending awards, overturning awards, or declaring precedent, the Review Body would be required to issue opinions that clarify its reasoning.

While the Review Body’s mandate would be limited, it would still have substantial power and would need to be subject to some controls. These controls would take several forms: (1) the composition of the Review Body, (2) the tenure of its members, (3) the provisions of IIAs and precedent, and (4) the timing of review decisions.

First, the composition of the reviewing Body must reflect a balance of state interests and institutional independence. Professors Posner and Yoo view the independence of an adjudicatory body as a sliding scale dependent on several factors. 229Posner & Yoo, supra note 69, at 26–27 & tbl.1. These include (1) whether the body is standing with membership for a fixed term or convened ad hoc for a particular dispute, (2) whether its members are appointed by the states or by other parties, and (3) whether state parties can express their dissatisfaction with a member’s performance through refusing to renominate. 230 See id.

To reflect the values of both independence and sovereignty, the Body should be composed of two classes of members—Members and Independent Members. These classes would be structured to contain features of both dependence and independence described above. Each state party would nominate two Members. Then the state parties would choose by consensus as many Independent Members as there are states in the treaty (two for BITs, for example). The Independent Members could not be citizens of any of the treaty’s party states, while there would be no restriction on the citizenship of Members. 231The citizenship of the Members would make them more likely to share the values of their home countries, while Independent Members may be less likely to adopt the same views as treaty parties. See id. at 12.

Members would serve shorter terms (for example, five years), while Independent Members would serve for longer terms (for example, ten years). The Independent Members’ longer terms would promote stability and allow jurisprudence to develop, 232 See Franck, supra note 23, at 1624. while the Members’ shorter terms increase accountability to the parties. The Members could serve multiple consecutive terms, making them dependent on parties’ approval for renomination. 233 See Posner & Yoo, supra note 69, at 21 (discussing differing incentives for adjudicators who are nominated by state parties). Independent Members could not be renominated, because the potential for renomination could shape Independent Members’ incentives in a way that would undermine their independence. However, if this system were adopted by multiple IIAs, then Independent Members could seek nomination for other reviewing bodies.

Although Members would be incentivized to represent state interests through both their nomination and potential for renomination, they would not be agents of the treaty parties. To increase the legitimacy of the Body, Members could not make ex parte contacts regarding arbitration or take direct orders from treaty parties. 234 See Franck, supra note 23, at 1624 (“[Standards of judicial conduct] promote[] confidence in the neutrality, independence, and the integrity of the judges.”). Given the nature of investment treaties and the complex interests they represent, Members would likely be engaged in balancing the interests not only of their home government but also of investors. 235Roberts, supra note 24, at 197. Because any decision would require at least a majority of all members, Members from a given state would need the votes of at least one Independent Member or one Member from another state before undertaking any action.

Independent Members would serve to balance the opinions of Members. The incentive structure encourages Independent Members to serve the system, rather than the individual parties. 236 See id. at 197–98; see also Posner & Yoo, supra note 69, at 21 (discussing how perceived bias undermines the legitimacy of adjudicatory systems). Independent Members who performed their duties successfully could be nominated for positions on other reviewing bodies. To compete for such a nomination, Independent Members would need to show that they are able to balance interests well and do their jobs without bias.

In addition to structural and reputational incentives, rule-like provisions, interpretative guidelines, and precedent would also constrain the Review Body. In determining whether an award was correctly decided, the Review Body would be limited to the tools provided in the IIAs and subsequent awards. The Review Body would also have to explain its reasoning when writing opinions. 237 Cheng, supra note 45, at 1021. Accordingly, the Review Body would be more likely to treat similar awards similarly and robustly justify any discrepancies, leading to consistent opinions. 238 See id.

Consistency is crucial for the legitimacy of the Review Body. 239 See Franck, supra note 23, at 1585. If observers see that the Review Body is acting within the mandate created by a given IIA, they will be reassured that the system is functioning properly. 240 See id. at 1616–17 (“[S]ubjecting awards to public scrutiny is another check on the discretion of a private body which provides an incentive for thorough, considered, and well-reasoned awards.”). Because rule-like provisions and interpretative guidelines help communicate the content of the law, they will make it easier for observers to be informed about the state of investment law and hold the Review Body accountable. 241 See Kaplow, supra note 128, at 562–63.

Finally, the timing of decisions by the Review Body would limit its ability to manipulate outcomes. 242 See Roberts, supra note 24, at 212 (“When states adopt an interpretation early on, there is a good chance that they will be perceived as acting legitimately . . . . When treaty parties enter into an interpretative agreement after an alleged breach or filing of a claim, there is a fair risk that the interpretation will be viewed skeptically, as an attempt to influence an ongoing case in which one is a respondent.”). If an investor seeks review of an award, the investor would be required to submit a request to the Review Body within thirty days of the final award. The Review Body must then act within sixty days. After that portion of review is complete, the Review Body would have an additional twenty days to determine whether to assign precedential value to the award. If the Review Body did not receive a review request from a litigant, it would have sixty days to determine whether to assign precedential value to the award.

Similarly short time frames are used by the World Trade Organization’s (WTO’s) appellate mechanism. 243Surya P. Subedi, The WTO Dispute Settlement Mechanism as a New Technique for Settling Disputes in International Law, in International Law and Dispute Settlement: New Problems and Techniques, supra note 198, at 173, 180. In that model, panels decide factual issues, and the appellate body can only review questions of law. 244 Isabelle Van Damme, Treaty Interpretation by the WTO Appellate Body 12 (2009). As a result, any appeal can be conducted rather quickly—within a maximum of ninety days. 245Subedi, supra note 243, at 180. Because the division of roles is similar in this proposed model for arbitration, it is feasible for review to be expedient.

The timing is critical for several reasons. First, finality is a key concept in arbitration. 246Brower, supra note 216, at 92. An extra layer of significant delay would be unwelcome. 247 Id. Yet where there are important interests at stake, a ninety-day delay strikes the appropriate balance between finality and correctness of law. Because arbitration is a very long process, parties would likely not mind waiting an additional ninety days, especially if it provides desired relief. 248 See Franck, supra note 23, at 1622 (arguing that a small sacrifice in finality would be acceptable to participants due to the vast benefits of an appeals process). Finally, if the Review Body had to act quickly, it may not have time to look at the specifics of other ongoing arbitrations. Thus, it would be less likely to assign precedential value to a particular award to influence the outcome of an ongoing arbitration. Even if the Review Body wanted to impact other arbitrations, the 110-day maximum window would limit the number of ongoing arbitrations the Body could influence.

As shown above, the mandate and structure of the Review Body would allow it to remain somewhat independent while still answering indirectly to the treaty parties. Within this structure, the Review Body fulfills the monitoring function envisioned by principal–agent theory 249 See supra notes 141–201 and accompanying text. and, in doing so, furthers important goals. First, the Review Body would lead to an increase in well-reasoned, consistent awards and the overall development of jurisprudence in a way that is less threatening to state sovereignty. The Review Body would also facilitate ongoing clarification of provisions written with interpretative guidelines. 250 See Franck, supra note 58, at 64 (noting that rules with lower levels of textual clarity can become legitimized through judicial processes). Significantly, the Review Body could help accomplish these goals within a framework that would improve the overall legitimacy of the investor–state system.

By sanctioning certain awards as precedent, the temporary Review Body would create a strong set of incentives for arbitrators to draft well-reasoned awards. 251 See Cheng, supra note 45, at 1024–25. In addition, the ability to overturn awards would also encourage arbitrators to stay within the mandate of the IIA. 252 See id. On the one hand, arbitrators would strive to author precedential awards. Authoring an award sanctioned as precedent would signal the quality of the arbitrator and would serve as an important credential for arbitrators who must compete for future selection. 253Roberts, supra note 24, at 197–98. Because well-reasoned decisions would be considered more valuable as precedent, arbitrators would have the incentive to write well-reasoned awards.

On the other hand, when arbitrators venture into areas of interpretative uncertainty, where questions are governed by standards and interpretative guidelines, they would have incentives to justify their reasoning thoroughly. Arbitrators would strive to convince the Review Body of the soundness of their reasoning in order to avoid being overturned and to secure future appointments. 254 See id. If arbitrators produced awards that did not comport with the tribunal’s mandate, the Review Body would overturn the award. This action and the subsequent negative effects on an arbitrator’s career would act as a sanction and would encourage compliance with the regime.

Moreover, the proposed system would allow the Review Body to communicate with tribunals about what types of reasoning and interpretations it finds persuasive. 255 See Kaplow, supra note 128, at 614 (noting that even opinions that are not precedent help to clarify and communicate the content of the law). The Review Body could also choose to approve an award without amending it or giving it precedential value. As well-reasoned awards present the issues to the Review Body, that Body would be able to read various awards and evaluate the logic used in each. This option gives the Review Body the ability to examine different outcomes over time without having to commit to precedent. An accrual of consistent persuasive awards may operate as unofficial precedent that would help actors learn the content of the law. 256 See id. However, this system allows both tribunals and the Review Body to be flexible when unanticipated factual scenarios arise. This process will clarify the law without prematurely committing states to undesirable interpretations.

D. Other Suggestions Only Resolve Problems of Legitimacy or Sovereignty

The solution described above offers to restore legitimacy to investor–state arbitration while making important structural choices that protect states’ sovereignty. Because the solution presents a compromise between divergent goals, it is unlikely to please critics who believe that adjudicatory bodies should be fully independent or those who believe that states should retain complete control over their sovereign prerogatives. This section compares the model proposed here with two other models. The first, based largely on the FTC, emphasizes state sovereignty. The second, based on extant international courts, focuses more on legitimacy.

Some IIAs, most notably NAFTA, have implemented a type of reviewing committee that oversees tribunals. 257 See Amirfar & Dreyer, supra note 78, at 50. The FTC is made up of state representatives and issues binding interpretations of NAFTA that help narrow tribunals’ mandates. 258 Id. While this can guide the tribunals’ jurisprudence, critics have called the FTC’s use of Interpretive Notes “self-serving” and have accused the FTC of stepping beyond the scope of its power. 259Franck, supra note 23, at 1604. Specifically, critics have accused NAFTA parties of using “the Commission as a tool to gratify narrow self-interests.” 260Brower, supra note 216, at 82. FTC interpretations have “appeared at a critical juncture in several arbitrations” and have been treated as binding on pending disputes. 261 Id. at 81. As Professor Charles H. Brower II put it, “[T]he Free Trade Commission’s work appears less of a principled exercise in international governance than a clever fait accompli designed to avoid liability.” 262 Id. The experience of the FTC highlights the types of delegitimizing actions that states may take when faced with a threat to their sovereignty.

The system proposed by this Comment avoids the main pitfalls of the FTC that have led observers to criticize its legitimacy. 263 See id. (criticizing the FTC as a nonlegitimate exercise of state power). First, in the proposed model, states can only change the treaty through official drafting or amendment. Direct control over the tribunals occurs ex ante, when it is deemed the most legitimate. 264 See Roberts, supra note 24, at 212–13. Likewise, the Review Body acts at a particular time in the process and cannot merely issue interpretations when it is convenient to do so. The timing of reviews also limits the extent to which the Review Body can manipulate the outcomes of ongoing arbitrations. 265 See id. Second, the Review Body is comprised of both Independent Members and party-aligned Members. This structure insulates it from direct control by state parties. Accordingly, the Review Body is less subject to political manipulation by states seeking to avoid damages.

On the other hand, the Review Body is not as responsive to states’ needs as the FTC. The Review Body cannot be used to amend the IIA ex post. At least one observer has noted that the ability to amend NAFTA through subsequent interpretation “may, in fact, make the Chapter 11 regime inherently more legitimate” insofar as it allows states to shape the regime to reflect the needs of their citizens. 266Yackee, supra note 17, at 334.

Generally, neither the FTC nor the Review Body emphasizes legitimacy as much as the proposals for a cross-cutting appellate body contemplated by other commentators. 267 See, e.g., Franck, supra note 23, at 1617–25 (proposing a permanent Investment Arbitration Appellate Court, comprised of independent members appointed with public input and implemented through an international convention or through the Free Trade Agreement of the Americas). These proposals include several factors of independent, and thus legitimate, tribunals. 268 See id. For example, such proposals tend to include a multinational treaty or institution, as well as a body composed solely of disinterested parties. 269 Id.

However, proposals for an appellate body face several challenges that are avoided by the structural changes presented here. First, establishing a multilateral appellate body would likely be more difficult than renegotiating or amending individual treaties. As the past has shown, states that oppose free-market ideology are more amenable to negotiation in bilateral settings than in multilateral ones. 270 See Van Harten, supra note 18, at 17–43 (offering an overview of why negotiations for a multilateral investment code failed while smaller scale investment treaties succeeded). Moreover, by negotiating more rule-like provisions, developed states can show their developing partners that they are willing to make compromises on areas of mutual interest.

Compromise may be indispensible because the evolving global economy provides decreasing incentives for developing nations to participate in IIAs. 271 See supra notes 104–18 and accompanying text. As a result of such compromise, however, it is certainly possible that IIAs will start to resemble individualized contracts that represent the intentions of the particular parties. For commentators who believe that IIAs should advance the goal of creating a unified body of investment law, this possibility may make an overarching appellate body a more attractive solution. 272Franck, supra note 23, at 1617–18.

However, an appellate body would do little to calm states’ sovereignty concerns. As Professor Brower points out, commentators who suggest a standing appellate body without more “have overlooked the problem of ‘control’” in their proposals. 273Brower, supra note 216, at 90. As evidenced by Professors Posner and Yoo’s study, states are less likely to use or comply with the rulings of a body that does not protect their interests. 274Posner & Yoo, supra note 69, at 13, 48 (noting that, in cases where the WTO ruled against either the European Union or the United States, compliance occurred only thirty-three percent of the time). Accordingly, the only way to ensure participation in an international adjudicatory system is to afford state parties some certainty that outcomes will be acceptable. 275 Id.

This Comment proposes a system that does precisely that, but that also improves the legitimacy of the existing model. While the resulting solution is a compromise that may not please critics at either extreme, it is workable and addresses the primary shortcomings of the present model of investor–state arbitration.

Conclusion

The solution proposed by this Comment is premised on two competing notions: (1) sovereign states will not submit to a system that they do not trust to represent their interests, and (2) a system that does not carry a perception of legitimacy will not last. Today’s model of investor–state arbitration is suffering from a legitimacy crisis that has been exacerbated by changes in the global economy. As arbitral awards continue to unpredictably penalize states for regulation, states will exit the system. For investor–state arbitration to survive, a compromise between legitimacy and sovereignty must be reached. By controlling tribunals through rule-like provisions drafted ex ante and a Review Board constituted ex post, states can protect both their sovereign interests and restore legitimacy to the investor–state arbitration model.

Footnotes

1William W. Burke-White & Andreas von Staden, Investment Protection in Extraordinary Times: The Interpretation and Application of Non-Precluded Measures Provisions in Bilateral Investment Treaties, 48 Va. J. Int’l L. 307, 309–10 (2008).

2 Id. at 310.

3 Id. at 311.

4 Id.

5 Id. (citing Wailin Wong, Argentina Treasury Attorney: World Bank Claims Could Reach $80 Billion, Dow Jones Int’l News, Jan. 21, 2005).

6 See Anne van Aaken & Jürgen Kurtz, The Global Financial Crisis: Will State Emergency Measures Trigger International Investment Disputes?, Vale Columbia Center on Sustainable Int’l Investment (Mar. 23, 2009), http://www.vcc.columbia.edu/content/global-financial-crisis-will-state-emergency-measures-trigger-international-investment-dispu (discussing the emergency measures taken by states in the wake of the 2008 financial crisis and their potential implications for investment arbitration).

7Paul Michael Blyschak, State Consent, Investor Interests and the Future of Investment Arbitration: Reanalyzing the Jurisdiction of Investor–State Tribunals in Hard Cases, 9 Asper Rev. Int’l Bus. & Trade L. 99, 117 (2009).

8 See Loewen Grp., Inc. v. United States, ICSID Case No. ARB(AF)/98/3, Final Award (June 26, 2003), 7 ICSID Rep. 442 (2005) (dismissing a $725 million claim against the United States brought by a foreign investor alleging it had been discriminated against by a Louisiana court when the court admitted evidence of the investor’s foreign status).

9 See Methanex Corp. v. United States, 44 I.L.M. 1345 (NAFTA Ch. 11 Arb. Trib. 2005) (dismissing all charges after six years of proceedings in which Methanex, a Canadian producer of methyl tertiary butyl ether (MTBE), sought $970 million in damages from the United States because California had banned MTBE, asserting that it had contaminated drinking water and endangered human health and the environment).

10Susan D. Franck, Development and Outcomes of Investment Treaty Arbitration, 50 Harv. Int’l L.J. 435, 435 (2009).

11 See id.

12 Bilateral Investment Treaties, Off. U.S. Trade Representative, http://www.ustr.gov/trade-agreements/bilateral-investment-treaties (last visited Mar. 27, 2012).

13 See Jeswald W. Salacuse, The Emerging Global Regime for Investment, 51 Harv. Int’l L.J. 427, 431–32 (2010).

14 Id.

15 Id. at 435.

16 Id.

17One of the primary differences among IIAs is the number of signatories: bilateral investment treaties (BITs) bind two countries, while free-trade agreements (FTAs) are often, but not always, signed between multiple countries. Jason Webb Yackee, Toward a Minimalist System of International Investment Law?, 32 Suffolk Transnat’l L. Rev. 303, 306 (2009).

18 See Jason Webb Yackee, Conceptual Difficulties in the Empirical Study of Bilateral Investment Treaties, 33 Brook. J. Int’l L. 405, 415–17 (2008) (discussing the broad similarities and subtle distinctions between BITs); see also Gus Van Harten, Investment Treaty Arbitration and Public Law 27–28 (2007) (discussing certain distinctions between IIAs but concluding that investment treaties are sufficiently similar to be discussed in the same general terms and as part of a global network).

19 See Salacuse, supra note 13, at 431.

20 Bilateral Investment Treaties, Trade Compliance Center, http://tcc.export.gov/Trade_Agreements/Bilateral_Investment_Treaties/index.asp (last visited Mar. 27, 2012).

21 Free Trade Agreements, Trade Compliance Center, http://tcc.export.gov/Trade_Agreements/Free_Trade_Agreements/index.asp (last visited Mar. 27, 2012).

22Salacuse, supra note 13, at 434.

23 See Susan D. Franck, The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions, 73 Fordham L. Rev. 1521, 1584 (2005).

24 See Anthea Roberts, Power and Persuasion in Investment Treaty Interpretation: The Dual Role of States, 104 Am. J. Int’l L. 179, 179 (2010) (“As investment treaties create broad standards rather than specific rules, they must be interpreted before they can be applied. Investor–state tribunals have accordingly played a critical role in interpreting, hence developing, investment treaty law.”).

25 Van Harten, supra note 18, at 123.

26 Id.

27Franck, supra note 10, at 436.

28 Van Harten, supra note 18, at 172.

29Charles N. Brower & Stephan W. Schill, Is Arbitration a Threat or a Boon to the Legitimacy of International Investment Law?, 9 Chi. J. Int’l L. 471, 473 (2009).

30 See Franck, supra note 23, at 1558–82 (describing in detail inconsistent awards under various BITs).

31Suzanne A. Spears, The Quest for Policy Space in a New Generation of International Investment Agreements, 13 J. Int’l Econ. L. 1037, 1040, 1043 (2010).

32 See Roberts, supra note 24, at 183 (“Viewing investment treaty arbitration solely through a public international law, state-to-state prism is unsatisfactory because investment treaties create reciprocal rights and duties for the treaty parties and rights for nonstate actors (investors).”).

33 Id. at 182.

34Tai-Heng Cheng, Power, Authority and International Investment Law, 20 Am. U. Int’l L. Rev. 465, 500–01 (2005).

35Franck, supra note 10, at 443.

36 Id.

37 Id.

38 Id.

39 Id.

40 Id.

41 Id.

42 See Andrea K. Bjorklund, Mandatory Rules of Law and Investment Arbitration, 18 Am. Rev. Int’l Arb. 175, 191 (2007).

43Benedict Kingsbury & Stephan Schill, Investor–State Arbitration as Governance: Fair and Equitable Treatment, Proportionality and the Emerging Global Administrative Law 4 (N.Y. Univ. Inst. for Int’l Law & Justice, Working Paper No. 2009/6, 2009), available at http://iilj.org/publications/documents/2009-6.KingsburySchill.pdf.

44Salacuse, supra note 13, at 467.

45 See id. (“[T]he approximately 300 decisions that have emanated from tribunals are consistently cited by lawyers and other tribunals and have a powerful influence on the making of future regime decisions.”); see also Tai-Heng Cheng, Precedent and Control in Investment Treaty Arbitration, 20 Fordham Int’l L.J. 1014 (2007) (arguing that arbitrators have made a concerted effort to reason and write awards in a way that resembles an informal system of precedent); W. Mark C. Weidemaier, Toward a Theory of Precedent in Arbitration, 51 Wm. & Mary L. Rev. 1895 (2010) (comparing investor–state arbitration, commercial arbitration, and labor arbitration, and noting that precedent operates most strongly in the investor–state model of arbitration).

46 See Cheng, supra note 45, at 1015.

47For example, legitimacy is related to “determinacy and coherence,” as well as to “fairness[ and] accountability.” Franck, supra note 23, at 1584.

48 Id.

49 See Laurence R. Helfer & Anne-Marie Slaughter, Toward a Theory of Effective Supranational Adjudication, 107 Yale L.J. 273, 312–13 (1997) (identifying independence as a key criterion of the legitimacy and effectiveness of a supranational adjudicatory body); Ruth Mackenzie & Philippe Sands, International Courts and Tribunals and the Independence of the International Judge, 44 Harv. Int’l L.J. 271, 274 (2003) (“Independence of the judiciary is a critical factor . . . .”).

50 See Franck, supra note 23, at 1584–85.

51 Id. at 1530–32.

52 Id. at 1584.

53 See Yackee, supra note 18, at 418–19.

54 See Blyschak, supra note 7.

55Tai-Heng Cheng & Robert Trisotto, Reasons and Reasoning in Investment Treaty Arbitration, 32 Suffolk Transnat’l L. Rev. 409, 425 (2009).

56Some IIAs contain a reasoning requirement as a control. Id. at 413. However, because that requirement is not explicitly defined in IIAs, it often fails to achieve its goal. See id. at 414.

57Franck, supra note 23, at 1546.

58 See Thomas M. Franck, The Power of Legitimacy Among Nations 61 (1990).

59Franck, supra note 23, at 1546.

60 See id. at 1547–48.

61 Id.

62 Franck, supra note 58, at 60–61.

63 See Yackee, supra note 18, at 418–19.

64 See Franck, supra note 58, at 64 (noting that institutions may be perceived as illegitimate if they do not “function in accordance with ascertainable principles of right process” (emphasis omitted)).

65 See Franck, supra note 10, at 443.

66 See Van Harten, supra note 18, at 167–74.

67 See supra notes 57–58 and accompanying text.

68Roberts, supra note 24, at 197–98.

69 See Eric A. Posner & John C. Yoo, Judicial Independence in International Tribunals, 93 Calif. L. Rev. 1, 26–27 (2005) (asserting that selection by parties to a dispute makes adjudicators dependent on those parties).

70 Id.; see also Roberts, supra note 24, at 197–98 (arguing that the common accusation that arbitrators will seek to create more work for themselves as a group (and thus act in a way that favors investors) must be balanced against the arbitrators’ awareness that reasonable and impartial interpretations are more likely to keep the system alive and expanding).

71 E.g., Van Harten, supra note 18, at 172–73.

72 See Spears, supra note 31, at 1040–41 (“Most of the IIAs in force today were drafted . . . by . . . capital-exporting states that subscribed to a market fundamentalist or ‘neo-liberal’ version of economic liberalism . . . . [N]eo-liberals do not envision any role for the regulatory state beyond the establishment and protection of property rights—not even for the correction of market failures, the risk of which they discount.”).

73 See Blyschak, supra note 7, at 116–17.

74Charles H. Brower II, Mitsubishi, Investor–State Arbitration, and the Law of State Immunity, 20 Am. U. Int’l. L. Rev. 907, 919 (2005).

75Blyschak, supra note 7, at 117.

76 Id. (quoting Aaron Cosbey, The Road to Hell? Investor Protections in NAFTA’s Chapter 11, in International Investment for Sustainable Development: Balancing Rights and Rewards 150, 151 (Lyuba Zarsky ed., 2005)) (internal quotation mark omitted).

77Id.

78 See Catherine M. Amirfar & Elyse M. Dreyer, Thirteen Years of NAFTA’s Chapter 11: The Criticisms, the United States’s Responses, and Lessons Learned, 20 N.Y. Int’l L. Rev. 39, 44 (2007) (discussing the Ethyl Corp. v. Canada, 38 I.L.M. 708 (NAFTA Ch. 11 Arb. Trib. 1999), arbitration where Canada settled the case for $13 million and permitted a chemical that was deemed harmful by the EPA to be sold in Canada); Cheng, supra note 34, at 508 (noting that one tribunal awarded investors $270 million).

79Cheng, supra note 34, at 508.

80 See supra notes 13–16 and accompanying text.

81 See Salacuse, supra note 13, at 432–34.

82 See id.

83 See id. at 437.

84 Van Harten, supra note 18, at 16–17.

85Salacuse, supra note 13, at 436–39.

86 Id. at 436–37.

87 See Van Harten, supra note 18, at 18.

88 Id.

89 Id.

90 Id.

91 Id. at 17.

92 See id. at 19–21 (discussing the numerous failed attempts to create a multilateral regime for international investment from the 1929 Draft Convention on the Treatment of Foreigners, to the multilateral investment code in the International Trade Organization, to the 1948 Havana Charter, to the Abs–Shawcross Draft Convention on Investments Abroad in 1959, to the 1967 Draft Convention on the Protection of Foreign Property of the Organisation for Economic Co-operation and Development).

93Salacuse, supra note 13, at 434.

94 See Van Harten, supra note 18, at 16.

95 See Salacuse, supra note 13, at 440. Professor Salacuse further argues that there may have been valid interests at stake for capital-exporting states. These include “(1) encouraging foreign investment in capital-importing countries; (2) relationship building; (3) economic liberalization; (4) encouraging domestic investment; and (5) improving governance and strengthening rule of law.” Id. at 440–41.

96Roberts, supra note 24, at 196.

97Salacuse, supra note 13, at 442.

98 See id. at 440.

99 Van Harten, supra note 18, at 43.

100 See id. at 38–44.

101Salacuse, supra note 13, at 429–30.

102 Id. at 433.

103 Id. at 435.

104 Id. at 433–34.

105 See Roberts, supra note 24, at 224.

106 See id. at 196–97.

107 See Ian Bremmer, The End of the Free Market: Who Wins the War Between States and Corporations? 161–66 (2010). State capitalism is a term used by Ian Bremmer to describe state-led economies that strategically manage players in their economies for political gain. Id. at 23.

108 Id. at 162.

109 Id. at 163.

110 See id. at 164.

111The term BRIC was coined in a 2001 report by Goldman Sachs. See Jim O’Neill, Building Better Global Economic BRICs 1 (Goldman Sachs Econ. Research Grp., Global Economics Paper No. 66, 2001), available at http://www2.goldmansachs.com/our-thinking/brics/brics-reports-pdfs/build-better-brics.pdf.

112 See Tetsufumi Yamakawa et al., BRICs Lead the Global Recovery, BRICs Monthly (Goldman Sachs, N.Y.C., N.Y.), May 29, 2009, at 1, available at http://www2.goldmansachs.com/our-thinking/brics/brics-reports-pdfs/brics-lead-recovery.pdf.

113 See Bremmer, supra note 107, at 164–65.

114According to its website, “Eurasia Group is the world’s leading global political risk research and consulting firm.” What We Do, Eurasia Group, http://www.eurasiagroup.net/about-eurasia-group (last visited Mar. 27, 2012). Eurasia Group’s “analysts monitor political, economic, social, and security developments . . . . They also examine cross-border issues such as trade, energy and other commodities, financial regulation, climate change, and global health.” Id.

115 Bremmer, supra note 107, at 43.

116 See id.

117 Id. at 46.

118Salacuse, supra note 13, at 469–70. Bolivia and Ecuador withdrew from the ICSID in 2007 and 2010, respectively. Id. at 469. Both plan on renegotiating a number of their IIAs, but Ecuador went as far as to leave nine IIAs. Id. at 470.

119Mark Thatcher & Alec Stone Sweet, Theory and Practice of Delegation to Non-Majoritarian Institutions, 25 W. Eur. Pol. 1, 4 (2002).

120 See id. (noting that the agent “resolve[s] commitment problems . . . [by] work[ing] to enhance the credibility of promises made . . . between multiple principals”).

121 Id.

122 See id. at 4–5.

123 Id. at 4. For example, if the point of creating investor–state arbitration tribunals is to allow them to enforce investors’ rights against encroachment from states, then granting states too much control over how those rights are enforced will likely defeat the purpose of convening the tribunal.

124 See id. at 4–5.

125Jonas Tallberg, Delegation to Supranational Institutions: Why, How, and with What Consequences?, 25 W. Eur. Pol. 23, 28 (2002).

126 Frederick Schauer, Playing by the Rules: A Philosophical Examination of Rule-Based Decision-Making in Law and in Life 158 (1991) (footnote omitted).

127 Id.

128Louis Kaplow, Rules Versus Standards: An Economic Analysis, 42 Duke L.J. 557, 560 (1992) (emphasis omitted).

129 See id. at 559–60.

130 See id. at 609. For example, a rule that says, “you must be eight years old to get on the Ferris wheel,” does not permit the enforcer to allow someone who is seven on the ride. Even if the reason behind the rule is that children should be a certain height and weight to safely enjoy the ride, the enforcer cannot allow a tall seven-year-old to ride with older friends.

131 See id. at 560.

132 See id. at 609.

133 Id. at 570. Contrast the rule above, supra note 130, with a standard that says, “you must be tall enough to safely ride the Ferris wheel.” Under these circumstances, the ride operator could allow our precocious seven-year-old on the ride. The risk in this situation is that the ride operator is not qualified to make the decision and would mistakenly allow a child who is too small on the ride.

134 Id. at 561.

135 Id.

136 See id. at 568 (stating that, when enacting a law, a government decides whether to promulgate each aspect as a rule or standard).

137 Id. at 577.

138 See id.

139 See id. at 611.

140 See id. at 577.

141Tallberg, supra note 125, at 28.

142 Id.

143 See Cheng, supra note 45, at 1021.

144 Id.

145 See id. at 1017–18 (indicating that judges in common law systems need compelling reasons to deviate from precedent).

146Posner & Yoo, supra note 69, at 13–14.

147 See id. at 14 (listing potential ramifications for judges who do not share the values of the political community).

148 See Thatcher & Sweet, supra note 119, at 4.

149 See Kaplow, supra note 128, at 559–62.

150 Id.

151 Id. at 560–62.

152 See id. at 571–77.

153 Id. at 572.

154 Id. at 568–71.

155 See id. at 568–86.

156 See id. at 569.

157 See id.

158 Id.

159 Id.

160 See id.

161 Id. at 571.

162 See id. at 571–77 (explaining different cost scenarios depending on whether individuals choose to become informed under a rule or a standard).

163 Id. at 570.

164 Id.

165 See id. at 570 & n.23.

166 Id. at 621.

167 Id. at 570.

168 See id. at 621.

169 See Roberts, supra note 24 (describing the dual role of states as both drafters of law and parties to future arbitrations).

170Tallberg, supra note 125, at 28.

171Jacob Katz Cogan, Competition and Control in International Adjudication, 48 Va. J. Int’l L. 411, 414 (2008).

172Posner & Yoo, supra note 69, at 13.

173 See, e.g., Helfer & Slaughter, supra note 49; Mackenzie & Sands, supra note 49.

174Posner & Yoo, supra note 69, at 7; accord Blyschak, supra note 7, at 99 (advocating a more moderate view than Posner and Yoo, but encouraging a more state-centric approach to arbitration).

175Cogan, supra note 171, at 414.

176 Id. at 414–15.

177 See Posner & Yoo, supra note 69, at 13.

178 Id.

179 See Yackee, supra note 17, at 309–10.

180 See Van Harten, supra note 18, at 13–44 (describing the development of IIAs and the “explosion” of claims in the mid-1990s).

181 See Spears, supra note 31, at 1042.

182 See Van Harten, supra note 18, at 37–41.

183 Id.

184 Id. at 41.

185 See Salacuse, supra note 13, at 433–34.

186 See Andrea K. Bjorklund, Reconciling State Sovereignty and Investor Protection in Denial of Justice Claims, 45 Va. J. Int’l L. 809, 832 (2005).

187Salacuse, supra note 13, at 435.

188 See Yackee, supra note 17, at 309–10.

189 See Kaplow, supra note 128, at 586–96.

190Spears, supra note 31, at 1042.

191 See Kaplow, supra note 128, at 583.

192 See Spears, supra note 31, at 1042.

193 See id. at 1040; see also Brower, supra note 74, at 917 (discussing the use of broad provisions in investment treaties).

194Kaplow, supra note 128, at 560–65.

195 See Spears, supra note 31, at 1048 (“[I]nterpretative language [included in newer IIAs] seeks to refine . . . standards and encourage arbitrators to conduct a balancing exercise to determine whether government action taken in response to legitimate regulatory concerns has violated investors’ rights.”).

196 See Amirfar & Dreyer, supra note 78, at 50.

197 See supra note 45 and accompanying text.

198 See Vaughan Lowe, Private Disputes and the Public Interest in International Law, in International Law and Dispute Settlement: New Problems and Techniques 3, 10 (Duncan French et al. eds., 2010).

199 Id.

200Franck, supra note 23, at 1588–89; accord Yackee, supra note 17, at 315 (arguing that correcting tribunal interpretations would be difficult because it would require renegotiating existing treaties).

201Bjorklund, supra note 186, at 832–33.

202 See Kingsbury & Schill, supra note 43, at 2.

203 See Roberts, supra note 24, at 196.

204 See Salacuse, supra note 13, at 470 (arguing that “[m]any parts of the world have lost faith” in the economic model that drove the IIA boom of the 1990s).

205 Id. at 469–70.

206Kingsbury & Schill, supra note 43, at 2.

207 See Roberts, supra note 24, at 196 (“These developments increase the prospect that at least some states will appear on both sides of the investment treaty equation and that a convergence of states’ views will thus be facilitated.”).

208 Id.

209 See Yackee, supra note 18, at 418–19.

210Franck, supra note 10, at 435.

211 See Blyschak, supra note 7, at 117.

212 See Kaplow, supra note 128, at 621 (“If conduct will be frequent, the additional costs of designing rules—which are borne once—are likely to be exceeded by the savings realized each time the rule is applied.”).

213 See Blyschak, supra note 7, at 113 (arguing that IIAs “are [a] lex specialis and as such only have effect between the parties”).

214 See Kaplow, supra note 128, at 614 (“Prior cases—both their outcomes and any written opinions—may reduce the costs and increase the accuracy of legal advice and adjudication even if no ‘binding precedent’ is created.”). This may be less true in a system that produces inconsistent decisions.

215 See Franck, supra note 10, at 435 (noting “a four-fold increase” in investment-treaty arbitration within the last decade).

216 See Charles H. Brower, II, Structure, Legitimacy, and NAFTA’s Investment Chapter, 36 Vand. J. Transnat’l L. 37, 88 (2003) (“[T]he use of simple rules to deal with complex matters often produces absurd results at the margins, creating . . . a paradoxical reduction of legitimacy.”).

217A similar approach has already been taken by the United States in its 2004 Model BIT, but the expansion of this technique has been controversial. See Franck, supra note 23, at 1588–89; Roberts, supra note 24, at 180.

218 See Spears, supra note 31, at 1048.

219 See id.

220 See Franck, supra note 58, at 54 (“[T]he more determinate a standard, the more difficult it is to justify non-compliance.”).

221 See Kaplow, supra note 128, at 585.

222 See id. at 577.

223 See id.

224 See Schauer, supra note 126, at 158–59.

225 See Franck, supra note 23, at 1589 (asserting that rule-like provisions have textual determinacy that in turn leads to predictable and consistent law).

226 Id. at 1584–85.

227 See Schauer, supra note 126, at 158.

228Spears, supra note 31, at 1048.

229Posner & Yoo, supra note 69, at 26–27 & tbl.1.

230 See id.

231The citizenship of the Members would make them more likely to share the values of their home countries, while Independent Members may be less likely to adopt the same views as treaty parties. See id. at 12.

232 See Franck, supra note 23, at 1624.

233 See Posner & Yoo, supra note 69, at 21 (discussing differing incentives for adjudicators who are nominated by state parties).

234 See Franck, supra note 23, at 1624 (“[Standards of judicial conduct] promote[] confidence in the neutrality, independence, and the integrity of the judges.”).

235Roberts, supra note 24, at 197.

236 See id. at 197–98; see also Posner & Yoo, supra note 69, at 21 (discussing how perceived bias undermines the legitimacy of adjudicatory systems).

237 Cheng, supra note 45, at 1021.

238 See id.

239 See Franck, supra note 23, at 1585.

240 See id. at 1616–17 (“[S]ubjecting awards to public scrutiny is another check on the discretion of a private body which provides an incentive for thorough, considered, and well-reasoned awards.”).

241 See Kaplow, supra note 128, at 562–63.

242 See Roberts, supra note 24, at 212 (“When states adopt an interpretation early on, there is a good chance that they will be perceived as acting legitimately . . . . When treaty parties enter into an interpretative agreement after an alleged breach or filing of a claim, there is a fair risk that the interpretation will be viewed skeptically, as an attempt to influence an ongoing case in which one is a respondent.”).

243Surya P. Subedi, The WTO Dispute Settlement Mechanism as a New Technique for Settling Disputes in International Law, in International Law and Dispute Settlement: New Problems and Techniques, supra note 198, at 173, 180.

244 Isabelle Van Damme, Treaty Interpretation by the WTO Appellate Body 12 (2009).

245Subedi, supra note 243, at 180.

246Brower, supra note 216, at 92.

247 Id.

248 See Franck, supra note 23, at 1622 (arguing that a small sacrifice in finality would be acceptable to participants due to the vast benefits of an appeals process).

249 See supra notes 141–201 and accompanying text.

250 See Franck, supra note 58, at 64 (noting that rules with lower levels of textual clarity can become legitimized through judicial processes).

251 See Cheng, supra note 45, at 1024–25.

252 See id.

253Roberts, supra note 24, at 197–98.

254 See id.

255 See Kaplow, supra note 128, at 614 (noting that even opinions that are not precedent help to clarify and communicate the content of the law).

256 See id.

257 See Amirfar & Dreyer, supra note 78, at 50.

258 Id.

259Franck, supra note 23, at 1604.

260Brower, supra note 216, at 82.

261 Id. at 81.

262 Id.

263 See id. (criticizing the FTC as a nonlegitimate exercise of state power).

264 See Roberts, supra note 24, at 212–13.

265 See id.

266Yackee, supra note 17, at 334.

267 See, e.g., Franck, supra note 23, at 1617–25 (proposing a permanent Investment Arbitration Appellate Court, comprised of independent members appointed with public input and implemented through an international convention or through the Free Trade Agreement of the Americas).

268 See id.

269 Id.

270 See Van Harten, supra note 18, at 17–43 (offering an overview of why negotiations for a multilateral investment code failed while smaller scale investment treaties succeeded).

271 See supra notes 104–18 and accompanying text.

272Franck, supra note 23, at 1617–18.

273Brower, supra note 216, at 90.

274Posner & Yoo, supra note 69, at 13, 48 (noting that, in cases where the WTO ruled against either the European Union or the United States, compliance occurred only thirty-three percent of the time).

275 Id.

Executive Notes and Comments Editor, Emory Law Journal; J.D. Candidate, Emory University School of Law (2012); A.B., Duke University (2005). I am grateful to Dean Robert Ahdieh for his patience and help in developing the foundation of this Comment. I am also indebted to Alan White and Robert Carlton, whose thoughtful suggestions helped hone my argument and whose encouragement pushed me to constantly improve it. Finally, I would like to thank the editorial staff of the Emory Law Journal, particularly Danny Reach, Andrew McKinley, James Spung, and Amy Dehnel for their precise and thorough editing.