Printer Friendly

IS THE WASHINGTON CONSENSUS REALLY DEAD? AN EMPIRICAL ANALYSIS OF FET CLAIMS IN INVESTMENT ARBITRATION.

Abstract

Goodbye Washington Consensus? More than a decade has passed since economists and policy makers have claimed that the attempt to prescribe a set of measures and policies that will enhance economic development, known as the Washington Consensus, was doomed to failure and is no longer relevant. This paper demonstrates that the Washington Consensus is in fact enforced on states through the interpretation and application of international investment agreements (IIAs), and specifically the fair and equitable (FET) standard. Using an innovative empirical analysis of all public investment arbitration awards published until 2015 in which FET claims are discussed (N=120), the results presented here infer that cases involving measures inconsistent with the Washington Consensus have up to a 45 percent higher chance of being found in violation of FET than other cases (p<.01). A more nuanced analysis that examines the role of specific measures in the outcomes of FET claims, found that only cases involving measures that are inconsistent with the Washington Consensus have a statistically significant higher chance of being found in violation of FET compared to other cases. These findings provide a novel empirical basis for legal professionals and practitioners wishing to understand which measures are unlikely to prevail in investment disputes in the current setting. As tribunals still interpret and implement IIAs in line with the Washington Consensus, it seems that unless IIAs are substantially amended--its effects are here to stay.
Table of Contents

I.   Introduction
II.  IIAs, FET and States' Regulatory Space
III. The Rise and Fall of the Washington Consensus
     and its Effects on IIAs and FET
IV.  Empirical Analysis--The Role of the Washington
     Consensus in the Application of the FET Standard
     by Investment Tribunals
     A. Previous Empirical Studies of Investment
        Arbitration
     B. Empirical Strategy
     C. Data
     D. Results
V.   Conclusion


I. Introduction

In a well-known article titled "Goodbye Washington Consensus, Hello Washington Confusion?" published more than a decade ago, Dani Rodrik declared that "it is fair to say that nobody really believes in the Washington Consensus anymore. The question now is not whether the Washington Consensus is dead or alive; it is what will replace it." (1) The Washington Consensus is a term describing the main policies prescribed by international financial organizations such as the World Bank during the 1990s to enhance development. (2) Critics have stressed that the Washington Consensus reflects a neo-liberal economic ideology that does not suit all states and could hinder their development. (3) Throughout the early 2000s the Consensus lost its popularity, with many recognizing its drawbacks and abandoning its guidelines. Consequently, Rodrik and others have claimed the Consensus is dead and no longer plays a role in the policy frameworks of states. (4)

This paper argues that despite Rodrik's declaration, the Washington Consensus is still here. The international investment legal framework it left behind, based on thousands of international investment agreements (IIAs), still calls states to obey this supposedly long-gone Washington Consensus. I explore this possibility here by focusing on one of the most frequently invoked clauses in international investment arbitration disputes: Fair and Equitable Treatment (FET). (5) The article demonstrates that there is a statistically significant greater chance that arbitration tribunals will determine a state violated the FET standard if it adopted measures inconsistent with the Washington Consensus. Accordingly, this paper stresses that if states want to adopt measures that are inconsistent with the Washington Consensus, they should strive actively to allow such measures by adapting their IIAs respectively.

IIAs are agreements between two or more states aiming to afford protection to foreign investors and promote foreign investments. (6) The Washington Consensus's call to abolish barriers impeding the entry of foreign direct investments (FDI) (7) supported the making of IIAs. Consequently, IIAs proliferated during the 1990s as the Washington Consensus reached its peak. Most IIAs include investor-state dispute settlement (ISDS) provisions, which allow investors to submit claims of alleged violations of IIAs to an international investment arbitration tribunal. These tribunals are usually capable of granting investors pecuniary awards, which range from dozens to hundreds of millions of dollars that states are obligated to pay to investors as result of their claim. Tribunals strive to interpret IIAs in line with their objectives and the circumstances upon which they agreed. Therefore, if IIAs were concluded with the intentions of introducing objectives coinciding with the Washington Consensus, then their interpretation and application could align with it as well. Such a conclusion is supported by literature stressing that liberal or neo-liberal ideologies, which align to a certain extent with the Washington Consensus, affect the arbitrators' interpretation of IIAs. (8) I will examine the above argument in this paper by focusing on the application of FET by investment arbitration tribunals.

The FET standard is required by almost all IIAs. While I will discuss further the precise content of the standard below, for now it is sufficient to say that it generally requires states to provide a consistent and transparent legal framework that protects investors' legitimate expectations, among other obligations. (9) Given the indeterminacy of FET, it is applied on a case to case basis at the tribunal's discretion. Since its application is subject to the tribunal's interpretation, (10) it could be subject to the principles set out by the Washington Consensus.

Indeed, the empirical analysis presented in this paper, based on all publicly available decisions on merits up to 2015 (inclusive), (11) infers that tribunals apply FET in correlation with Washington Consensus policies. Particularly, trade regulations, measures that hinder property rights and negative treatment of privatized assets that may imply an effort to withdraw privatization--increase the chances a tribunal will find a violation of FET.

This paper offers three main contributions to existing literature. First, it demonstrates that the FET standard is not as un predictable as generally considered. Though the FET standard can be applied in various circumstances, this paper concludes that there are certain measures that are likely to be found in violation of it--namely, measures that are inconsistent with the Washington Consensus. When considering whether or not to adopt a measure, states often consider its effects on their exposure to investment arbitration claims. Identifying measures that are more likely be in violation with the FET standard could assist states in this process by narrowing the uncertainty that may be involved in arbitration proceedings. Second, it opposes the common perception that the Washington Consensus is no longer relevant by suggesting that the international investment legal framework is still affected by it. This is important, because it implies that if states wish to adopt measures that are inconsistent with the Washington Consensus--for example, withdrawing from privatizations or imposing new trade regulations, such as those recently advocated by the Trump administration--they would be exposed to investment arbitration claims. Mitigating these risks will require amending their IIAs correspondingly, stating their new objectives clearly.

Third and finally, the empirical analysis offers a somewhat new methodological approach compared to existing literature analyzing the outcomes of investment arbitration. Empirical literature examining variables that account for investment arbitration outcomes generally fail to account for arbitration claims' contexts, reflected in the facts of the cases and the cause of action in each case. Rather, outcomes of arbitration cases are mainly predicted by accounting for the identity of the parties to the disputes, the background of the arbitrators, and their legal services. (12) Assuming that the facts of a case and the cause of action do matter, this raises concerns over an omitted variable bias in the results of existing literature. The model adopted in this paper aims to address these concerns by (a) focusing only on one specific cause of action, namely, FET claims; and (b) by using a model that includes variables that represent the measures taken by the respondent states challenged by the claimant, therefore accounting for the factual bases of the claims.

The paper is structured as follows. Part II provides further background on IIAs, focusing on the implications for states' regulatory space resulting from their prescribed ISDS mechanisms and in view of the indeterminacy of provisions such as FET. Part III offers a basic background of the rise and fall of the Washington Consensus and its role in the proliferation of IIAs. Part IV then presents the main argument of this paper: that one can anticipate some correlation between FET and the Washington Consensus. Following this theoretical background, Part V presents empirical support for this claim. Finally, Part VI concludes by discussing the implications of the findings of this paper for states.

II. IIAs, FET and States' Regulatory Space

International investment agreements between two or more states aim to afford protection to foreign investors and promote foreign investments. (13) To achieve thL goal, IIAs include various provisions, requiring, for example, that each state will avoid unlawful expropriation of alien property and afford foreign investors fair and equitable treatment. A unique characteristic of IIAs, when compared to other international law instruments, is that they typically allow investors to submit arbitration claims directly against states that have allegedly breached the provisions set out in the IIA, and to receive compensations for their losses. This ISDS mechanism is quite different from the dispute settlement mechanism provided by trade agreements, which only provide dispute settlement mechanisms between states, and grant preference to amicable agreements between disputing states before allowing states to resort to a binding dispute settlement mechanism. While state-to-state dispute settlement mechanisms are affected by political constraints of states, (14) such restraints are less apparent in ISDS mechanisms. Thus, investors are more likely to turn to a dispute settlement mechanism when states may have violated the provisions of an IIA. Additionally, ISDS may result in more significant remedies, as investment arbitration awards typically amount to millions of dollars. The unique characteristics of investment arbitration tribunals require states who wish to avoid the risk of investment arbitration to be mindful of the measures they choose to adopt or to refrain from. However, the requirement prescribed by certain provisions that appear in most IIAs is not always clear. Provisions that prohibit indirect expropriation could be triggered in various circumstances that are not always easy to anticipate. (15) Moreover, the requirement to afford foreign investors fair and equitable treatment may be interpreted differently by different investment tribunals, creating uncertainties regarding the precise requirements of the FET standard. These uncertainties have elicited concerns that states may avoid adopting regulations over the mere fear of a possible arbitration claim, even when the legal basis of such a claim is shaky.

Many of these concerns were raised as part of discussions regarding the FET standard. Most international investment treaties include a clause requiring the parties to grant FET to foreign investors. (16) The FET clause serves a central role in one of the main objectives of investment treaties: to protect foreign investments. (17) Legal principles such as the FET standard allow flexibility in their application at the price of some uncertainty as to how the principle will be applied in certain situations. (18) The indeterminacy of the FET standard allows tribunals to use it for filling gaps that might be left between specific rules that exist in the treaty in order to ensure a desired level of investor protection. (19) For example, if a tribunal determines that the conditions prescribed by an IIA for an illegal expropriation were not met, it might decree a violation of FET if it finds that the state acted arbitrarily or in a discriminatory manner. (20) Notably, its broad application does create some overlap between FET and other provisions that exist in most IIAs. (21)

The concept of fair and equitable treatment first appeared in international documents in 1948, and has entered since into many IIAs. (22) Granted, the large wave of IIAs during the 1990s, with the influence of the Washington Consensus, also increased the prevalence of the FET standard in international agreements. (23) The first publicly known investment arbitration cases discussing the content of the FET clause occurred in the late 1990s, and the interpretation of the standard by investment arbitration tribunals has evolved ever since. (24)

Though the FET clause exists in almost every IIA, its phrasing may vary, affecting the level of treatment it requires states to afford foreign investors. The main approaches in formulating the FET clause are as follows: (1) requiring fair and equitable treatment to investors with no further specification ("unqualified FET"); (2) requiring FET linked to principles of international law; (3) requiring FET linked to customary international law; and (4) requiring FET linked to the minimum standard of treatment of aliens. (25) The three latter formations are "qualified" FET clauses since they include some additional description of the standard. Some have stressed that the phrasing of the FET clause might affect the threshold for finding a breach of the standard by an arbitration tribunal. (26) Generally, unqualified FET clauses prescribe the lowest threshold to prove a violation of FET. (27) Usually, tribunals examining IIAs that include an unqualified FET clause would determine that it is an autonomous standard which requires states to provide, among other things, a consistent and transparent legal framework that protects investors' legitimate expectations. (28) In contrast, the Free Trade Commission (FTC) established in the context of the North American Free Trade Agreement (NAFTA) has issued a binding interpretation to NAFTA's FET clause that calls for the highest threshold to prove a violation of FET. The FTC determined that the FET does not require more than the "minimum standard of treatment" that prevails in customary international law and that only prohibits shocking, egregious, and outrageous behavior of states. (29)

Some scholars and policy makers have claimed that the low precision of the FET standard creates ambiguities that hinder states' development. Their view is that the FET standard has been implemented inconsistently and at times in a too far-reaching manner. (30) An inconsistent interpretation of the standard may impose obstacles on states' decisions to modify their regulatory structure, even when necessary. (31) According to this line of thinking, a more consistent application of the standard would minimize the negative effect on state development. The United Nations Conference on Trade and Development (UNCTAD) Investment Policy Framework for Sustainable Development has thus offered recommendations that range from specifying the FET clause in investment agreements to simply omitting it. (32) Despite these significant concerns about the effects of the FET clause on state development, there has been no systematic empirical examination of the application of the standard and its alleged inconsistency. The following section posits that the context in which IIAs have proliferated and the possible common ideology of arbitrators may drive tribunals to apply FET standards consistent to the Washington Consensus. According to this argument, the effects of the application of FET standards on state development are somewhat different than suggested above, and largely depend on the desirability of the Washington Consensus. A state might be reluctant to adopt policies that are inconsistent with the Washington Consensus even if they better fit its perceived developmental process, fearing successful FET claims. (33)

III. THE RISE AND FALL OF THE WASHINGTON CONSENSUS AND ITS EFFECTS ON IIAs AND FET

The Washington Consensus is a term coined by John Williamson in 1989 in reference to ten policies prescribed by international financial institutions to developing states, namely, fiscal discipline; re-ordering of public expenditure priorities; tax reforms; liberalization of interest rates; competitive exchange rate; trade liberalization; liberalization of inward foreign direct investment; privatization; deregulation to ease barriers to entry and exit; and ensuring of property rights. (34)

The Washington Consensus attracted criticism, with claims that the policies it prescribed have not, in fact, enhanced development and led to its eventual unravelling. (35) For over a decade now, scholars have claimed that the Washington Consensus no longer exists. (36) However, the role of the Washington Consensus in formulating policies and international legal instruments cannot be ignored--particularly with regard to the large wave of investment treaties of the 1990s, which scholars have attributed to the liberal economic ideologies the Washington Consensus promoted. (37)

As a result, since the early 2000s investors have increasingly used the procedural option granted by these IIAs to turn to an ISDS mechanism that allows foreign investors to submit arbitration claims against governments that allegedly mistreated them. (38) The far-reaching effects of the Washington Consensus are thus still present in the form of IIAs and arbitration claims of investors against states. However, these effects of the Washington Consensus are not limited to the mere existence of IIAs, but to their interpretation as well. A basic principle of treaty interpretation laws is that a treaty must be interpreted in accordance with its context and objectives. (39) Furthermore, a secondary source of interpretation is the circumstances of its conclusion. Because many IIAs were intended to promote policies encouraged by the Washington Consensus, and their preambles often refer to economic ideas the Consensus promoted, such as the importance of market liberalization, (40) one might presume that the interpretation of IIAs would also conform to the Washington Consensus. More specifically, the interpretation and the application tribunals of FET may be influenced by the Washington Consensus.

Aside from the legal expectation of some coherence between the Washington Consensus and FET, characteristics of the international investment community create another factor favoring conformity between the two. Recent literature on the social characteristics of members of international investment arbitrators indicates they are a highly networked community sharing various platforms of communication via conferences, specialized professional and academic publications, and written decisions in arbitrations. (41) A relatively small number of arbitrators accounts for half of the appointments in the most dominant investor-state arbitration institution--the International Centre for Settlement of Investment Disputes (ICSID). Arbitrators of the ICSID social network have sat on arbitral panels with each other, creating a closed community. (42) A tight network of professionals of this kind may foster common norms and ideology, and vice versa--common norms and ideology create conformity among the members of the arbitration community (43) This has led some to doubt the general criticism that international investment arbitration suffers from inconsistencies, and to assert that such claims are largely based on anecdotal examples. (44) The social structure of international arbitration may also lead to doubts about the claims that the application of FET is inconsistent. (45)

At the same time, the indetermination of the FET standard increases its sensitivity to perceptions of arbitrators. (46) If arbitrators share a common view of what is "fair and equitable," this might influence their determination of which policies do not provide investors "fair and equitable treatment."

Though existing scholarship on the underlying norms and ideologies of arbitrators is scarce, studies suggest that arbitrators share a liberal or a neoliberal economic ideology. (47) Recently, Hirsch offered a preliminary analysis of the sociology of international investment law, addressing the common norms and ideologies of the investment arbitration community. He argues that given the social settings in which international investment law emerges and is being interpreted, and given investment lawyers and arbitrators' career paths and the heritage of investment law, "investment lawyers (and arbitrators) are inclined to emphasize the importance of unimpeded flow of capital, legal predictability, and market economy ideology." (48)

Such common economic ideologies among arbitrators may influence their point of departure when examining whether a certain policy or measure implemented by the host state offered the investor "fair and equitable treatment." (49) Since many perceive the Washington Consensus as an expression of neoliberal economic policies, (50) this might imply that interpretation arbitrators would tend to find policies that contradict the Washington Consensus as treatment that is not fair and equitable, thus violating the FET standard. In practice, if the application of FET is affected by the Washington Consensus, this would mean that policies inconsistent with the Washington Consensus would be more likely to be found in violation of FET.

Not all measures can be easily equated to the Washington Consensus policies, since some policies call for broad principles that are dependent on other circumstances, leaving room for discretion. These include achieving a "tax system that would combine a broad tax base with moderate marginal tax rates," fiscal discipline, and competitive exchange rates. (51) However, the Washington Consensus does call for several specific policies, deviation from which would be easily noticed. These include market liberalization, enforcement of strong property rights, liberalization of capital flows to FDI, and privatizations. (52) Therefore, trade regulations like price controls, production and export curtailments, and new import regulations seem to be inconsistent with the Washington Consensus. (53) Similarly, actions that hinder the investor's property rights, such as confiscation or expropriation of property and the imposition of a temporary administrator, are also not consistent with the Washington Consensus. FDI for state development played an important role in the Washington Consensus. While later concern was expressed about the phenomenon of states offering excessive investment incentives, (54) narrowing investment incentives may indicate a restrictive attitude towards FDI, and thus in contradiction with the Washington Consensus. (55) Finally, since the Washington Consensus calls for privatization, measures indicating efforts to withdraw from already established privatizations are also inconsistent with the Consensus.

This leads to the hypothesis examined empirically in the next section: that policies that are inconsistent with the Washington Consensus are more likely to be found in violation of FET than others. As elaborated in the following paragraphs, the analysis infers that such policies have a statistically significant and large chance of being found in violation of FET compared to others.

IV. EMPIRICAL ANALYSIS--THE ROLE OF THE WASHINGTON CONSENSUS IN THE APPLICATION OF THE FET STANDARD BY INVESTMENT TRIBUNALS

As argued above, the application of FET by tribunals could be affected by the Washington Consensus both because of the legal basis of their interpretation and because of possible "disciplinary sensibility" among arbitrators. (56) This section offers an empirical assessment of the hypothesis that policies that are inconsistent with the Washington Consensus are more likely to be found in violation of FET.

A. Previous Empirical Studies of Investment Arbitration

Empirical analyses of investor-state arbitration tribunals are of growing interest. However, just a few distinct subjects have been examined empirically in recent years. Most commonly, empirical research addressed the outcomes of arbitration proceedings and jurisdictional decisions, (57) possible biases of arbitrators, (58) and the role of jurisprudence in investor-state arbitration. (59)

Studies examining factors that might affect the outcomes of investment arbitration are the closest to the topic examined here. These studies usually focus on the final decision of the tribunal (in favor of the claimant or the respondent), and as opposed to the current study, do not focus on the outcomes of specific claims of investors. However, factors that may have affected the final outcome of investment arbitrations are likely to have an effect on the outcomes of specific claims as well. Therefore, it is useful to briefly review existing literature that examined the outcomes of investment arbitration empirically, and compare methodological approaches that were adopted in recent literature on the topic'. The following short summary of relevant studies aims to do so, while highlighting that, generally, existing literature applied limited statistical analyses that did not attempt to control for the unique characteristics that emerge from each case, namely, the factual basis and the legal arguments presented to the tribunal.

Franck examined the role of the development status of the respondent State and arbitrators in several studies, and found no significant relationship between these elements and the outcome of investment arbitration cases. In an early study, she examined the role of development status on investment arbitration outcomes in 52 investment arbitration cases. The simple statistical analysis applied did not attempt to control for additional factors and showed no significant relation between the outcomes of cases and the development status of the parties to the dispute and arbitrators. (60) In a later study, Franck reached similar results in an analysis of over 150 cases, while controlling the democracy level of the countries of the parties to the dispute and arbitrators. (61) Finally, in a more recent study, Franck & Wylie examined several statistical models in search of variables that may predict outcomes of investment arbitration, based on over 150 cases. (62) These models included variables that account for the investor identity (natural person or legal entity), respondent identity (development status and democracy level), experience of parties' legal counsel, arbitrators' gender composition, their composite-development status, whether the chair was a repeat player with multiple appointments, and venue of the arbitration (was the case discussed in ICSID). This study too found no significant relationship between the development status of the parties to the dispute and of arbitrators with the arbitration outcomes. (63) However, the findings did show a significant relationship between the attorneys' experience and the identity of the investor with the outcomes of investment arbitration. (64) Evidently, the growing number of publicly available awards and decisions allowed exploring the role of various factors on the outcomes of investment arbitration cases.

Other studies examined the effects of the institutional structures of investment arbitration on the outcomes of investment arbitration cases. While Franck found no evidence to any relationship between the institutional background of investment arbitration cases, (65) Balan found that NAFTA tribunals tend to dismiss cases more frequently than in other institutional contexts. (66) Both studies, however, did not control for other factors that may have affected these results.

Several studies examined the role of the appointing process of arbitrators, which commonly allows each party to appoint one of the three arbitrators, on the outcomes of investment arbitration cases. Kapeliuk provided descriptive statistics of dozens of investment arbitration cases that may support the possibility that repeat arbitrators are uninfluenced by the need to satisfy parties that appointed them. (67) However, Puig & Strezhnev surveyed over 250 professional arbitrators, and by comparing the responses to hypothetical cases in which part of the participants were told that one of the parties appointed them and others were not, found a significant effect to the common appointing process. (68) This experimental approach allowed the authors to overcome the difficulty in accounting for substantial differences among arbitration cases, among other issues. Nonetheless, one might argue that the behavior of arbitrators could differ in actual cases when the stakes are higher.

Interestingly, most of these studies did not attempt to account for the factual basis and the legal claims raised by the claimant. This could represent a somewhat cynical approach, that the facts of each case do not have a substantial effect on its outcomes. However, if the facts of the case and the legal claims it is based on do contribute to the final decision of the tribunal, the empirical analyses that examined the actual outcomes of investment arbitrations may have suffered from an omitted variable bias. (69)

For example, experienced attorneys may have better abilities to predict outcomes of investment arbitrations than inexperienced attorneys. (70) Therefore, in cases that seem to have a low chance of success, experienced attorneys might recommend their clients not to pursue a costly arbitration, while inexperienced attorneys might fail to recognize that the case entails low chances of success. Consequently, inexperienced attorneys would have a greater chance to lose in arbitration. If an analysis wishes to examine the role of attorneys' experience on the outcomes of arbitration, but fails to account for the factual basis of the claim--it might attribute a significant effect to the outcomes of the procedure to the attorneys' experience, while the factual basis of the case dictated its outcome.

An exception to this is the recent work of Waibel & Wu, (71) who examined the role of non-legal factors on the outcomes of investment arbitration with an extensive model that accounted for the "legal strength" of each case. This was done by asking a panel of experts for their opinion on the legal strength of the cases. Waible & Wu found that policy preferences had a significant effect on liability decisions (but not on jurisdiction decisions). (72) These policy preferences were measured by the development status of the countries arbitrators came from, specialization in public international law, constitutional law or administrative law, and the professional experience of arbitrators the executive branch of government or in corporations. (73) They also found that the development status of the arbitrators' countries had the most prominent effect on the liability decisions made by arbitrators. (74)

An additional approach developed by Van Harten aims to overcome this difficulty by focusing on trends in legal interpretations made by tribunals. (75) Van Harten analyzed 100 cases that resolved issues of jurisdiction and admissibility. (76) The study found statistically significant evidence that tribunals favor the position of claimants over the position of the respondent States, and as well as the position of claimants from major Western capital-exporting states over claimants from other states. (77)

The approach developed in this paper borrows from Van Harten's general approach. The analysis provided here examines the interpretation of FET, by examining its actual application in investment arbitration cases. It also attempts to control differences among cases by accounting for greater variety of circumstances that were included in previous studies that might affect the outcomes of cases. Specifically, the analysis uses a model that includes variables that represent the measures taken by the respondent states challenged by the claimant, therefore accounting for the factual bases of the claims.

B. Empirical Strategy

I examined the hypothesis that policies that are inconsistent with the Washington Consensus are more likely to be found in violation of FET using a number of regression models, analyzing arbitration awards and decisions on liability that included a claim that the respondent state acted in violation of FET.

The null hypothesis is, therefore, that the FET standard has been applied inconsistently by tribunals, therefore finding no certain policy or set of policies that has a statistically significant greater chance of being in violation of FET compared to others.

The examination of this hypothesis is based on an original data set compiled of publicly available FET claims listed in the UNCTAD ISDS database. (78) All publicly available awards and decisions in English were retrieved from UNCTAD's database using references from the database and from Italaw. (79) Each of these awards and decisions was carefully read, with a focus on the factual basis and the FET analysis presented by the tribunal. A list of measures challenged in each case was extracted from the awards. Based on that list, each award received values representing which of the measures in the list were challenged by the investor. Additionally, I examined the type of FET clause the claim was based on, assigning values representing references to international law and minimum standard of treatment, as well as whether it was a NAFTA claim submitted after the FTC adopted its interpretation to FET in NAFTA. (80) Relevant background of the case was also examined based on the details provided in the award; specifically, each award received values representing whether the state claimed the challenged measures served health or environmental concerns, or were the result of a financial crisis or a revolution.

Essentially, all models include independent variables that represent the policies challenged by the claimant that would allow examining the role of the Washington Consensus on the decision of the tribunal of whether the state violated FET. In addition, the models control for several other circumstances that may affect the tribunal's decision whether or not the state indeed acted in violation of FET, such as the type of wording of the FET clause in the IIA that the arbitration dispute is based on.

First, I examined the hypothesis by grouping all cases using a parameter indicating whether the measures challenged in the FET claim were inconsistent with the Washington Consensus. This was done using several possible understandings of which policies should be viewed as inconsistent with the Washington Consensus, which I elaborate in the next paragraphs. If arbitrators tend to find measures/policies that are inconsistent with the Washington Consensus in violation of FET, the group representing such policies should have a significant positive effect on the possibility that the tribunal would find a violation of the FET standard. Notably, this approach is rather sensitive to the definition of the Washington Consensus. Any determination that a certain measure aligns with the consensus may affect the results. Therefore, I have introduced a robustness test examining the effects of a few different definitions of the Washington Consensus, explained below. The null hypothesis--that policies inconsistent with the Washington Consensus do not have a higher chance of being found in violation of FET compared to others--implies that no significant difference should be found between the outcomes of FET claims that challenge different policies. Specifically, consistency with the Washington Consensus should not have any significant effect on the determination that a violation of FET has occurred.

Once establishing that FET claims challenging measures/ policies that are inconsistent with the Washington Consensus have a greater chance of being found in violation of FET than others, a further inquiry was made as to which measures/ policies have the most significant effect on the outcomes of FET claims. This is accomplished by examining whether any of the main policies/presented below that were challenged in FET claims have a greater chance than others of being found in violation of FET. (81)

In all versions of the model, the standard errors are clustered by the respondent state to allow correlations between observations that include the same respondent state resulting from other uncontrolled factors. (82)

The approach this paper offers differs somewhat from most approaches of existing empirical analyses of investment arbitrations. Most of these analyses focus on the outcomes of investor-state disputes and general information on the parties of the dispute (such as the development level of the host state and main properties of the investor) and of the legal officials in the procedures (such as the arbitrators' and counsels' background). Therefore, existing empirical literature on variables accounting for investment arbitration outcomes have generally failed to account for the cause of actions and the factual background of arbitration claims. (83) The model adopted in this paper aims to overcome this concern by (a) focusing only on one specific cause of action, namely, FET claims; and (b) using a model that includes variables representing the measures taken by the respondent states that were challenged by the claimant, therefore accounting for the factual bases of the claims.

1. Dependent Variable

In both models, the dependent variable is a dummy variable indicating whether or not the FET claim was accepted by the tribunal, denoted by . received a value of 1 if the tribunal explicitly stated that the respondent state violated the FET standard, regardless of the amount awarded to the claimant. Cases in which the tribunal did not state whether or not the respondent acted in violation of FET were omitted.

2. Independent Variables

Since there is no consensus on which variables may affect investment arbitration outcomes, a few alternate models are suggested. The basic models offered here include variables that indicate which measures were challenged by the claimant in order to test the hypotheses. Additionally, three variables that were found to reliably predict the outcomes of investment arbitrations by Franck & Wylie are included: the investor's identity (individual or corporation) and counsel expertise of the claimant and of the respondent state.

I tested the hypothesis that tribunals tend to find measures inconsistent with the Washington Consensus in violation of FET by grouping measures that are inconsistent with the Washington Consensus. In accordance with the analysis presented in part III of this paper, WCinconsistent received the value 1 if it included trade regulations, measures that hinder property rights, narrowing investment incentives, or involvement of a privatized asset. WCinconclusive received the value 1 if it included any of the other policies challenged by investors as presented below. Ultimately, some controversy may arise as to which policies should be categorized as inconsistent with the Washington Consensus. As discussed in part III of this paper, trade regulation, hindering property rights, and narrowing investment incentive are considered to unquestionably contradict the Washington Consensus. However, tax increases, monetary policies, and the inclusion of cases whereby the investment involved in the dispute was privatized may be questionable. Therefore, a robustness test is presented examining a few alternatives and their effect on the results.

Additional explanatory variables were added to account for different wordings of the FET clause in the various treaties that were applicable in each case, as described in Part II above. These differences were addressed by adding two dummy variables. FETqualified received the value 1 if the applicable treaty of the dispute included a qualified FET clause. If the applicable treaty was the NAFTA and the award was delivered before the FTC issued its interpretation, FETqualified also received the value 1. If the award was issued after the interpretation was issued, FET qualified received the value 0, and an additional dummy variable denoted by NAFTAFET received the value 1. Seven cases were based on IIAs that had no FET clause, but the tribunal imported an FET clause through the Most Favored Nation (MFN) clause. (84) When the tribunal specified the agreement from which it imported the FET clause, it was coded respectively; however, in two cases the tribunal did not refer to a specific IIA, and these were discarded from this analysis.

Franck & Wylie have found that individuals have a better chance of success in investment arbitrations than corporations do.85 Therefore, the model includes a dummy variable denoted by ClaimantCorporation and scoring 1 if the claimant was a corporation is included in the model. Furthermore, experienced legal teams increased the chances of their client to succeed in the proceeding; two variables indicating the experience of the attorneys of each side to the dispute were therefore produced by adding up the number of cases from the dataset in which each law firm or state agency representing the state was involved (e.g. Canada's and the United States' legal teams), denoted by Claimant Attorney Experience and RespondentAttorney Experience, respectively. This measurement method is similar to that adopted by Franck & Wylie, but instead of ranking the law firms, here the number of cases that involved each firm represented its experience. In most cases, more than one law firm represented each of the disputing parties. In these cases, the variables received the value of the most experienced law firm that represented each party. Since, as demonstrated by Frank & Wylie, these variables have an effect on the final outcome of the procedure, they are accounted for here as well to avoid an omitted variable bias.

This basic model, however, does not account for circumstances that some claim may affect the tribunal's decision when determining whether a violation of FET has occurred. Thus, I offer a more expansive model in addition to the basic model described above. Since some tribunals have noted that violations of other clauses may construct a violation of the FET standard, the model includes an additional dummy variable indicating whether the tribunal found a breach of a different clause, denoted by OtherViolations. Scholars have also stressed that the application of the FET standard on the legitimate expectations of investors may be linked with the development level of the host state. (86) To address this, I included a dummy variable indicating whether the respondent state was a developed state, based on the classification available in UNCTAD's database (denoted by RespondentDeveloped). (87) The notion that the application of the FET standard should balance the investor's legitimate expectations with the states concerns (88) may require variables to reveal the circumstances affecting the respondent state's conduct. Two additional dummy variables were used to control for such circumstances: the first indicating whether the award reveals that the respondent claimed the measures were part of enforcement of environmental or health regulations (denoted by EnviromentalHealth); the second indicating whether the award reveals the measures took place during or after a revolution or economic crisis (denoted by Crisis).

Notably, all circumstances were extracted from the awards, which may lead to concerns about their validity. However, as these factual circumstances were usually undisputed, and therefore seem valid. A possible weakness of these models is that they do not include reference to whether or not state officials made specific promises regarding the investment that were later violated. This aspect may have special importance since many attribute the possibility of legitimate expectations to the existence of such promise. (89) However, given that the existence of such promise is usually disputed between the parties, it is thus difficult to examine objectively. In addition, because there is no relevant representation for determining a violation of the investor's legitimate expectations as long as the exact threshold remains vague, (90) it is difficult to decide what types of representations should be included in the model.

The models examined use both a linear probability model (LPM) and probit regression. Under the LPM assumption, estimated coefficients of each variable measure the change in the probability of a tribunal finding a certain case in violation of the FET standard if that certain variable exists, holding other factors fixed. (91) Although a logit or a probit regression is often better suited for a binary dependent and independent variable, this is not the case when the number of observations is small. (92) Nonetheless, they may serve well for a robustness test of the results obtained from the LPM versions of the models.

Presented below is a further inquiry as to which policies and measures have the most significant effect in the outcomes of FET claims. This is achieved by breaking down the WCinconsistent and WCinconclusive variables into dummy variables that re present the ten main policies challenged by investors, and including all of them in the model. The variable TradeRegulation scored 1 if the investor challenged import and export regulations and restrictions, and policies that involved price regulations such as strict tariff regulation on water and electricity. Property Rights scored 1 if the state took measures to confiscate or expropriate property, and if the state imposed a temporary administrator on certain assets. NarrowingInvestmentIncentives scored 1 if the investor challenged the state's decision to cancel or narrow incentives to promote investments such as tax reliefs. PrivatizedAsset received the value 1 if the award revealed that the disputed investment was privatized by the state. Contractsandconccesions represents whether the respondent state terminated, breached, or annulled a contract with the investor. If the award reveals that the state did not contest this factual basis, Contractsandconccesions scored 1. LicenseandPermits scored 1 if the claimant challenged a license or permits revocation, the denial of permits, enforcement of mandatory standards and bans, or changes in the governmental classification and permitted uses of land (rezoning). Any challenge of a judicial decision was denoted by the variable JudicialDecision. Since tax measure have an ambivalent relation with the Washington Consensus, these were divided in two separate variables. TaxCollection scored 1 if the investor challenged measures that took place as part of the tax collection process. 'Taxlncrease received the value 1 if a certain policy included a tax increase. Tax increases may overlap with other categories: import tax increases may overlap with TradeRegulation, while cancellation of tax reliefs for investors may overlap with NarrowingInvestmentIncentives. Monetary Policy indicates whether the state changed its monetary policy, usually by changing its exchange rate regime fixed to float. Criminalmeasures_fine scored 1 if the award reveals that the state had taken criminal measures, such as a criminal investigation, imprisonment, or imposing a fine against the investor or his investment.

C. Data

Although empirical research of investor-state arbitration proceedings are evolving, it suffers from some constraints, mostly derived from the confidential nature of most arbitration disputes. (93) Thus, the sample of publicly available arbitration decisions is rather small. The recently uploaded UNCTAD database of ISDSs aims to collect information about publicly known disputes from a variety of sources, and allows a more informed analysis, though many of the cases remain classified or unknown. (94) This makes it difficult to carry out complex analyses, and raises concerns of a selection bias when conducting research based on publicly known arbitration decisions. For example, circumstances that affect the parties' decision to publicize the award may also affect the outcome of the dispute, creating biased results that do not represent investment arbitration awards as a whole. Though this concern prevails in the present research, it is partly addressed by the fact that some of the awards that have become public were originally confidential. These awards were made public only sometime after the award was delivered at the initiative of one of the parties to the dispute. Moreover, the present research focuses on the application of the FET standard by arbitrators, who have limited control over whether or not the award will be publicized at later period. Therefore, the results of this research should not be greatly affected by this issue.

The information analyzed below is based on the UNCTAD database of investor-state disputes that recently became available to the public. (95) This database enables easy access to most publicly known arbitration awards, and allows for the extrication of cases in which a breach of the FET clause was raised by the claimant, and in which the tribunal found jurisdiction over the case. These criteria were met in 182 cases. However, not every award on this list was published; twenty-two cases were only granted partial coverage after the parties of the dispute published a short notice. In addition, eleven awards were published in languages other than English, and were not surveyed for this research. In twenty-five of the remaining cases, the tribunal did not discuss the FET allegations, even though they were raised by the claimant. Typically, this was because the tribunal found a breach of a different standard, making FET analysis unnecessary. Finally, in six of the cases the tribunal found that only part of the measures account for a FET breach. Since the unit of observation is each case, including these cases may distort the results, and therefore I have excluded them from this analysis. Thus, only 120 cases are available for examining the general application of the FET standard. Notably, fifteen of the cases involved a common background of the Argentine crisis during the early 2000s, creating a concern of an outlier. The analysis thus examines how the exclusion of the Argentinian cases may affect the results. Because the unit of observation used in the statistical analyses is each case, six cases in which only part of the measures was in violation of FET were discarded from the analyses. (96)

Importantly, the data allows for a control group to examine the effect of the policies that are inconsistent with the Washington Consensus: the sample in the following analysis included both cases in which the challenged policies were aligned with the Washington Consensus (fifty-four cases) and cases where at least part of the policies were inconsistent with the Washington Consensus (sixty-six cases).

Table 1 allows a first glance at measures that were identified in the cases examined. The measures are grouped into categories. (97) It demonstrates that although FET claims have been raised against a large range of measures and policies, they are typically used to challenge just a handful of measures and policies. Often, investors invoke the FET standard when states do not fulfil their obligations under contracts and concessions, or when states have changed, revoked, or denied permits and licenses. This could occur when courts annul a concession or a contract due to corruption or other legal bases, or when a state has failed to complete the terms of the concession or contract. Other judicial decisions have also created FET claims, such as bankruptcy procedures against foreign investors. Trade regulations have frequently been challenged in FET claims, common claims being against policies that imposed trade regulation, such as import and export restrictions and price regulation (mostly on privatized services such as electricity and water). A rarer type comprises claims against the failure of a government to enforce restrictions on the market. Many claims have arisen from actions that hindered the investor's property rights. Most IIAs include a prohibition on expropriation, with certain exclusions. The use of tribunals of the FET clause in these cases can be divided into two types: cases where the FET clause was used when the expropriation claims failed; and cases where the tribunal found a violation of the expropriation clause and has asserted that this is ultimately a breach of the FET clause as well. A few tribunals have however avoided examining the FET standard once finding an illegitimate expropriation. Investors have submitted FET claims when states have narrowed investment incentives, such as tax exemptions; and tax increases and certain measures adopted to collect tax payments have also resulted in FET claims. Several claims were submitted after Argentina adopted certain monetary measures during its financial crisis in the early 2000s. And finally, claims involving privatized assets are common, with a total of forty that involved this issue.

As illustrated in Figure 1, the rate of FET claims accepted by investment tribunals varies between state conducts that were challenged by investors. More specifically, measures inconsistent with the Washington Consensus, as defined in part III of this paper, have a greater success rate than other policies. Nevertheless, this is not sufficient for concluding that tribunals consistently find measures and policies that contradict the Washington Consensus in violation of FET. These variations may result from other factors, such as the experience of the law firm representing the claimant. These factors are accounted for in the statistical analysis presented in the following paragraphs.

D. Results

I will first investigate the effects of policies and measures inconsistent with the Washington Consensus on the outcomes of FET claims. Then, I will examine which measures and policies have the most significant effect on the outcomes of these claims. This latter inquiry allows associating measures and policies with violations of FET. As apparent from the following paragraphs, cases involving policies that are inconsistent with the Washington Consensus have a higher chance to be found in violation of FET.

1. Effects of Policies Inconsistent with the Washington Consensus on the Outcomes of FET Claims

Grouping the policies that are inconsistent with the Washington Consensus allows examination of the possible cultural bias tribunals have towards the Consensus. Table 2 demonstrates, using the assumptions of the basic model, that cases challenging policies that are inconsistent with the Washington Consensus have a large and significantly higher chance of being in violation of FET than other cases. (98) Even after excluding cases that involved Argentina, findings show that cases involving measures and policies inconsistent with the Washington Consensus have a 44% higher chance of being found in violation of FET (pc.001). These results are supported by probit models as well, indicating their robustness.

Table 3 demonstrates that similar results are obtained when applying the expansive model, although the size of the effect of cases involving policies that are inconsistent with the Washington Consensus is smaller.

These results may be sensitive to the exact definition of the Washington Consensus, and to the argument of what exactly the Washington Consensus entails. Therefore, the determination of which measures should be combined under Wcinconsistent is crucial. Therefore, I have proposed and tested a number of different options in Table 4. The table includes six groups of state measures. Three of them--namely, imposing trade regulation, hindering property rights, and narrowing investment incentive--are considered to unquestionably contradict the Washington consensus, and therefore included in every regression. However, as it may be questionable to include tax increases, monetary policies, and cases where the investment involved in the dispute was privatized, the effect of their inclusion in WCinconsistent is tested in Table 4, and option 5 is tested in the results above. Columns 7 to 8 describe the coefficients obtained in the various LPM models offered above. Ultimately, any definition that includes trade regulations, measures that hinder property rights, and the narrowing of investment incentives has a significant effect on the determination of a violation of FET, with small variations of the magnitude of such effect.

The results obtained here of course infer that tribunals tend to find policies that are inconsistent with the Washington Consensus in violation of FET. (99)

2. Associating Measures and Policies with Violations of FET

A more detailed inquiry of the effects of certain measures and policies infers that cases involving measures that hinder property rights have a significant robust effect on the probability that a violation of FET would be determined by the tribunal. Cases involving trade regulations, privatized assets, and tax collection measures may also have a higher chance of being found in violation of FET, though these results are sensitive to various assumptions about which variables should be added into the models and about the p-value adjustments that may be required. These results align with the second proposition, that tribunals tend to find measures that are inconsistent with the Washington Consensus in violation of FET: all three measures that under certain assumptions were found to have a higher chance to be in FET violation by arbitration tribunals are inconsistent with the Washington Consensus. As demonstrated in Table 5, cases involving privatized assets, which may imply an attempt to step back from the privatization, have a 25.5% higher chance of being in violation of FET than other cases. Cases that involved measures that hindered investors property right had an even larger chance, with a 41.1% higher chance of FET violation than other cases. Similar results are found when applying the models with a probit model as well as with LPM, both when Argentina is included in the analysis and when it is excluded from it. Tax collection measures have a negative significant effect on the possibility that a tribunal would find a violation of FET, though this was found only under the LPM assumptions and not when the probit model was used. Importantly, the fact that these models test multiple hypotheses may raise some concern that the null hypotheses would be rejected even if they are true. (100) I addressed this issue by using Bonferroni adjusted p-values, and found that only the negative effect of measures hindering property rights was significant in all four versions of the basic model at a .01 level. The negative effect of cases involving privatized assets was only significant in the probit version of the basic model, which included Argentina at a .05 level, and at a .1 level in the other versions. However, even under the conservative Bonferroni adjusted p-values, the null hypothesis that all policies have no effect on the FET claim is rejected at a .01 significance level: that is, the null hypothesis that tribunals have no tendency to find certain policies in violation of FET is rejected.

Quite similar results are obtained when using the expansive model, as demonstrated in Table 6. Cases involving privatized assets and cases involving measures that hindered the investor's property rights still seem to have a higher chance to be in violation of FET. Under these expansive models, cases involving trade regulations also have a higher chance of being in violation of FET than others. However, under these assumptions, the study found no significant effect for cases that involved tax collection measures. When using Bonferroni adjusted p-values, the negative effect of measures hindering property rights is significant in a .05 level in the LPM version that includes Argentina in the analysis and in both probit versions, though it is just significant in a .1 level under the LPM version that excludes Argentina. Trade regulations, however, have a negative significant effect in the .05 and .1 levels only when Argentina is included in the LPM and probit versions, respectively.

V. CONCLUSION

More than a decade since the Washington Consensus was marked as irrelevant, the international investment legal regime it left behind continues to affect state policy. This paper demonstrates that, along with the existence of thousands of IIAs--which are to some extent the result of the Washington Consensus--the interpretation and application of these IIAs, and specifically the FET standard, de facto enforce the Washington Consensus on states through ISDS mechanisms.

An empirical analysis of all public awards published up to 2015 that appear in UNCTAD's ISDS database (101) reveals that measures and policies that are inconsistent with the Washington Consensus have a greater chance of being found in violation of FET. More specifically, tribunals tend to find a violation of the FET standard when states carry certain measures that undermine concepts of free trade, property rights, and FDIs, and when a privatized asset is involved. As opposed to these, tax collection measures are more likely to be found to be fair and equitable.

This paper suggests that these results could be attributed either to legal interpretation of IIAs that is consistent with their objectives or to the characteristics of the investment arbitration community. Nevertheless, the effects of the Washington Consensus on current awards are clear, which implies that the effects of the Washington Consensus did not diminish even though the Washington Consensus ceased to exist. To that end, the Washington Consensus remains alive simply because the legal instruments it left behind have been evolving slowly. For example, though critics have argued that the Washington Consensus wished to create a one-size suit that did not fit all states, and that states should be granted a greater regulatory space, (102) IIAs have demonstrated little change in the approach of states towards obtaining a greater regulatory space. (103) A change in the perception of policy makers is not enough to diminish the effects of the Washington Consensus on states. If states intend to allow themselves greater regulatory space to adopt measures that are inconsistent with the Washington Consensus, they should amend their IIAs accordingly. The recent adoption of a "tobacco carveout" Trans-Pacific Partnership agreement, which increases regulatory space regarding tobacco products, might indicate first steps in that direction. (104)

As tribunals still interpret and implement IIAs in line with the Washington Consensus, it seems that unless IIAs are substantially amended--its effects are here to stay.

(1.) See Dani Rodrik, Goodbye Washington Consensus, Hello Washington Confusion? A Review of the World Bank's Economic Growth in the 1990s: Learning from a Decade of Reform, 44 J. Econ. Literature 973, 974 (2006). (2.) See John Williamson, What Washington Means by Policy Reform, in Latin American Adjustment: How Much Has Happened? 7 (John Williamson ed., 1990) [hereinafter Williamson, What Washington Means] (presenting Washington Consensus for first time); John Williamson, A Short History of the Washington Consensus, 15 L. & Bus. Rev. Am. 7 (2009) [hereinafter Williamson, Short History] (reviewing the Washington Consensus and examining changes in the approaches of international financial organizations); Joseph E. Stiglitz, Globalization and Its Discontents 16 (2003) (criticizing Washington Consensus); Rodrik, supra note 1 (arguing Washington Consensus no longer exists). (3.) See Stiglitz, supra note 2, at 16. "[T]he economic policies that evolved into the Washington Consensus and were introduced into developing countries were not appropriate for countries in the early stages of development or early stages of transition." Id. See also Charles Gore, The Rise and Fall of the Washington Consensus as a Paradigm for Developing Countries, 28 World Dev. 789 (2000) (highlighting tension between Washington Consensus and concept of Sustainable Development); Joseph E. Stiglitz, Is there a Post-Washington Consensus Consensus?, in The Washington Consensus Reconsidered: Towards a New Global Governance 41 (Narcis Serra & Joseph E. Stiglitz Eds., 2008) [hereinafter Stiglitz, Post-Washington Consensus] "[0]ne-size-fits-all policies are doomed to fail. Policies that work in one country may not work in others. The contrast between the success of the East Asian economies, which did not follow the Washington Consensus, and those that did has become increasingly clear." See Stiglitz, Post-Washington Consensus, supra. See Martin Khor, The World Trading System and Development Concerns, in The Washington Consensus Reconsidered: Towards a New Global Governance 215, 220-21 (Narcis Serra & Joseph E. Stiglitz Eds., 2008) "[A] one-size-fits-all approach will not work and, if enforced, might cause more harm than good." Id. See Rodrik, supra note 1. "Policy reforms of the (Augmented) Washington Consensus type are ineffective because there is nothing that ensures that they are closely targeted on what may be the most important constraints blocking economic growth." Id. See Jose Antonio Ocampo, A Broad View of Macroeconomic Stability, in The Washington Consensus Reconsidered: Towards a New Global Governance 63, 89 (Narcis Serra & Joseph E. Stiglitz Eds., 2008). "[G]iven the multiple objectives and trade-offs faced by macroeconomic authorities, solutions are likely to differ according to the conditions that characterize each country." Id. (4.) See Gary Gereffi, Global Value Chains in a Post-Washington Consensus World, 21 Rev. Int'l Pol. Econ. 9, 14-15 (2014) (explaining how the evolvement of global value chains has changed global governance and ended the era of the Washington Consensus); Rodrik, supra note 1; Stiglitz, Post-Washington Consensus, supra note 3, at 41. "If there is a consensus today about what strategies are most likely to promote the development of the poorest countries in the world, it is this: there is no consensus except that the Washington Consensus did not provide the answer." See Stiglitz, Post-Washington Consensus, supra. See also Eric Sheppard & Helga Leitner, Quo vadis neoliberalism? The Remaking of Global Capitalist Governance after the Washington Consensus, 41 Geoforum 185, 186-87 (2010) (arguing Washington Consensus no longer prevails). (5.) See Rudolf Dolzer, Fair and Equitable Treatment: Today's Contours, 12 St. Clara J. Int'l. L. 7,10 (2013); Nigel Blackaby, Constantine Partasides, Alan Redfern & Martin Hunter, Redfern and Hunter on international arbitration [paragraph] 8.59 (5th ed. 2009) [hereinafter Redfern and Hunter].

(6.) See Andrew Paul Newcombe & Lluis Paradell, Law and Practice of Investment Treaties: Standards of Treatment 1-2 (2009).

(7.) See John Williamson, The Washington Consensus as Policy Prescription for Development, in Development Challenges in the 1990s: Leading Policymakers Speak from Experience 33, 40 (2004) (explaining role of Foreign Direct Investment in Washington Consensus).

(8.) See Amr A. Shalakany, Arbitration and the Third World: A Plea for Reassessing Bias Under the Specter of Neoliberalism, 41 Harv. Int'l. L.J. 419 (2000) (arguing neoliberal economic ideology permeated investment arbitration decisions); Muthucumaraswamy Sornarajah, The Case against a Regime on International Investment Law, in Regionalism in International Investment Law 475, 497-98 (Leon Trakman & Nicola Ranieri ed., 2013) (similar). See also Chester Brown & Kate Miles, Evolution in Investment Treaty Law and Arbitration 633 (2011). "The ideology of neoliberalism dominated the last decade of the twentieth century [...], Many arbitrators sought to give effect to what was the dominant economic thought of the period in their awards." Id.

9. See, e.g., Dolzer, supra note 5, at 16-32 (describing main methods of application of FET in investment arbitration practice); Christoph H. Schreuer, Fair and Equitable Treatment (FET): Interactions with Other Standards, 4 Transnat'l. Disp. Mgmt. 1, 17-18 (2007). "It is widely accepted that the most important function of the FET standard is the protection of the investor's legitimate expectation through the creation of a transparent and stable legal framework." See Schreur, supra. See Redfern and Hunter, supra note 5, [paragraph] 8.61.
   It has been held that failure to ensure due process, consistency,
   and transparency in the functioning of public authorities and the
   lack of a predictable and stable framework for investment contrary
   to legitimate expectations of the investor and commitments made by
   the host State, are breaches of fair and equitable treatment
   standards.


Id. See also Rudolf Dolzer & Christoph Schreuer, Principles of international investment law 133-49 (2008) (describing typical situations may establish violation of FET, while highlighting importance of consistent and transparent legal framework).

(10.) See Ioana Tudor, The Fair and Equitable Treatment Standard in the International Law of Foreign Investment 155 (Oxford Univ. Press 2008). "The role of the judge in the application of FET is crucial, as he is the one identifying the various situations that may be identified by arbitrators." Id.

(11.) See U.N. Conf. on Trade & Dev., Investment Dispute Settlement Navigator, http://investmentpolicyhub.unctad.org/ISDS (last visited Aug. 10, 2017).

(12.) Susan D. Franck, Empirically Evaluating Claims about Investment Treaty Arbitration, 86 N.C. L. Rev. 1 (2007) [hereinafter Franck, Evaluating] (providing descriptive information about dozens of investment arbitration proceedings); Susan D. Franck, Development and Outcomes of Investment Treaty Arbitration, 50 Harv. Int'l. L.J. 435 (2009) [hereinafter Franck, Development] (examining the role of development status in investment arbitration outcomes); Daphna Kapeliuk, Repeat Appointment Factor: Exploring Decision Patterns of Elite Investment Arbitrators, 96 Cornell L. Rev. 47 (2010) [hereinafter Kapeliuk, Patterns] (analyzing judicial behavior of repeatedly appointed arbitrators in investment arbitration); Susan D. Franck, The ICSID Effect? Considering Potential Variations in Arbitration Awards, 51 Va. J. Int'l. L. (2011) [hereinafter Franck, ICSID Effect] (comparing International Centre for Settlement of Investment Disputes (ICSID) awards and investment arbitration awards rendered in other forums); Daphna Kapeliuk, Collegial Games: Analyzing the Effect of Panel Composition on Outcome in Investment Arbitration, 31 Rev. Litig 267 (2012) [hereinafter Kapeliuk, Collegial Games] (exploring the effect of panel composition on the outcome of disputes); Susan D. Franck, Conflating Politics and Development: Examining Investment Treaty Arbitration Outcomes, 55 Va. J. Int'l. L. 13 (2014) [hereinafter Franck, Politics] (analyzing investment arbitration outcomes as a function of raw wins and losses, amounts awarded, and relative investor success); Susan D. Franck & Lindsey E. Wylie, Predicting Outcomes in Investment Treaty Arbitration, 65 Duke L.J. (2015) (examining several models to predict outcomes of investment arbitration that do not include reference to the facts and claims of each case). But see Gus Van Harten, Arbitrator Behaviour in Asymmetrical Adjudication: An Empirical Study of Investment Treaty Arbitration, 50 Osgoode Hall L.J. 211 (2012) (examining trends in legal interpretations of arbitrators while focusing on interpretations granted in specific issues).

(13.) See Newcombe & Paradell, supra note 6, at 1-2.

(14.) The interlinks between the WTO dispute settlement mechanisms and politics are various, allowing countries to find paths that bypass strict legal solutions. See, e.g., Geoffrey & James McCall Smith, The Politics of WTO Dispute Settlement (UCLA International Institute Occasional Paper Series, 2002) (discussing strategic restraint by potential complainants in the cases they choose to file; strategic conciliation by the Appellate Body in decisions involving powerful defendants; and strategic bargaining between losing defendants and winning complainants outside of the DSU regarding the terms and timing of compliance); Sebastiaan Princen, EC Compliance with WTO Law: The Interplay of Law and Politics, 15 Eur. J Int'l L. 555 (2004) (demonstrating the role of internal politics on trade disputes); Lisa L. Martin, International Economic Institutions, in The Oxford Handbook of Political Institutions 654, 664-65 (2008) (summarizing several scholars who emphasized the possibility of countries to evade from trade obligations in light of political pressures). See also Marc L. Busch & Eric Reinhardt, Bargaining in the Shadow of the Law: Early Settlement in GATT/ WTO Disputes, 24 Fordham Int'l L.J. 158 (2000) (discussing variety of incentives for countries to initiate and settle WTO and GATT disputes).

(15.) See, e.g., Eli Lilly and Company v. The Government of Can., (1976) UNCITRAL, ICSID Case No. UNCT/14/2, (referring to claims that legal developments in case law may nullify intellectual property rights and constitutes a prohibited expropriation when applied in practice).

(16.) E.g., Redfern and Hunter, supra note 5,1 8.59; Newcombe & Paradell, supra note 6, at 255.

(17.) See id.

(18.) See, e.g., Pierre Schlag, Rules and Standards, 33 UCLA L. Rev. 379 (1985) (examining the importance and use of both rules and standards in legal systems).

(19.) See Dolzer & Schreuer, supra note 9, at 122; Tljdor, supra note 10, at 155.

(20.) See Schreuer, supra note 9, at 22-23.

(21.) See id.

(22.) See Dolzer & Schreuer, supra note 9, at 120.

(23.) E.g., Newcombe & Paradell, supra note 6, at 78.

(24.) See Dolzer & Schreuer, supra note 9, at 119-21; Tudor, supra note 10, at 15.

(25.) See Tudor, supra note 10, at 15-52 (reviewing the main practices of FET clauses in international investment agreements (IIAs)).

(26.) See United Nations Conference on Trade & Development, Fair and Equitable Treatment: UNCTAD Series on Issues in International Investment Agreements II, 17-29, U.N. Doc. UNCTAD/DIAE/IA/2011/5 (2012), available at http://unctad.org/ en/Docs/unctaddiaeia201 ld5_en.pdf [hereinafter: UNCTAD, Fair and Equitable Treatment]', United Nations Conference on Trade & Development, Investment Policy Framework for Sustainable Development, 83, U.N. Doc. UNCTAD/DIAE/PCB/2015/ 5 (Dec. 23, 2015), available at http://unctad.org/fr/PublicationsLibrary/diaepcb2012 d5_en.pdf (similar) [hereinafter: UNCTAD, Investment Policy]; Tudor, supra note 10, at 15-52 (similar). But see D. Franck, The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law through Inconsistent Decisions, 73 Fordham Law Rev. 1521, 1589 (2005). "Further definition of substantive rights in investment treaties is only a partial solution." Id.

(27.) See UNCTAD, Fair and Equitable Treatment, supra note 26, at 17-29.

(28.) See Dolzer, supra note 5 at 16-32; Schreuer, supra note 9; Redfern and Hunter, supra note 5, 1 8.61; Tudor, supra note 10, at 155; Dolzer & Schreuer, supra note 9.

(29.) See Tudor, supra note 10 at 151; Patrick Dumberry, The Protection of Investors' Legitimate Expectations and the Fair and Equitable Treatment Standard under NAFTA Article 1105, 31 J. Int'l Arb. 47 (2014).
   The position adopted by NAFTA tribunals regarding the
   interpretation of the concept of legitimate expectations clearly
   contrasts with the much more liberal approach that has been taken
   by non-NAFTA tribunals. Thus, only one award supports the view that
   the concept constitutes a stand-alone element of the FET standard
   under Article 1105.


Id.

(30.) See Roland Klager, Revising Treatment Standards--Fair and Equitable Treatment in Light of Sustainable Development, in Shifting Paradigms in International Investment Law: More Balanced, Less Isolated, Increasingly Diversified 65, 67 (Steffen Hindelang & Markus Krajewski eds., 2016). "The dynamic development of international investment has also caused problems and concerns as the case law, especially on fair and equitable treatment [...] is often perceived as being too far-reaching and inconsistent." Id.

(31.) See id. at 67. "States were also held liable for breaching fair and equitable treatment, concerns rose that this provision may constrict a State's sovereignty considerably and that this threatens sustainable development." Id. See also Roland Klager, Fair and Equitable Treatment and Sustainable Development, in Sustainable Development in World Investment Law 241, 241, 250-51 (2011). "The most obvious threat to the promotion of sustainable development posed by fair and equitable treatment is represented by the standard's power to limit the regulatory freedom of host States." Id. See also UNCTAD, Investment Policy, supra note 26, at 83 "The use of FET to protect investors' legitimate expectations can indirectly restrict countries' ability to change investment-related policies or to introduce new policies - including those for the public good - that may have a negative impact on individual foreign investors." Id. See also Rudolf Dolzer, The Impact of International Investment Treaties on Domestic Administrative Law, 37 N.Y.U. J. Int'l L. & Pol. 953, 964 (2004). "Certainly, the principle of fair and equitable treatment may, in practice, have wideranging repercussions for the sovereignty of the host state to determine and apply its administrative law, as it covers all phases of the investments and extends to all areas of domestic law affecting foreign investment." Id.

(32.) See UNCTAD, Investment Policy, supra note 26, at 83.

(33.) See Lorenzo Cotula, Do Investment Treaties Unduly Constrain Regulatory Space?, 9 Questions Int'l L. 19, 21 (2014). "Given the large amounts awarded by arbitral tribunals to investors, the cost of regulation could in principle skew incentives for public action and discourage regulation, and it could do so beyond the state directly involved." Id. See also Stephanie Bijlmaker, Effects of Foreign Direct Investment Arbitration on a State's Regulatory Autonomy Involving the Public Interest, 23 Am. Rev. Int'l Arb. 245 (2012) (discussing possible chilling effect that investment arbitration can cause due to prospects of costly damages and lawsuits); Stephan Schill, Do Investment Treaties Chill Unilateral State Regulation to Mitigate Climate Change?, 24 J. Int'l Arb. 469 (2007) (similar).

(34.) See Williamson, What Washington Means, supra note 2 (presenting Washington Consensus); Williamson, Short History, supra note 2 (reviewing Washington Consensus several years after it was first discussed). See also Stiglitz, supra note 2 at 16 (reviewing Washington Consensus and discussing several criticisms regarding Consensus); Rodrik, supra note 1 (similar).

(35.) See Stiglitz, supra note 2 at 16.
   Capital market liberalization has been pushed despite the fact that
   there is no evidence showing it spurs economic growth. In other
   cases, the economic policies that evolved into the Washington
   Consensus and were introduced into developing countries were not
   appropriate for countries in the early stages of development or
   early stage, of transition.


Id. See also Gore, supra note 2 (discussing processes that led to shift from Washington Consensus to other policies); Rodrik, supra note 1 (discussing drawbacks of Washington Consensus and how process ended Washington Consensus); Ocampo, supra note 3 (similar).

(36.) See Gereffi, supra note 4, at 14-15; Rodrik, supra note 1; Stiglitz, Post-Washington Consensus, supra note 3; Williamson, Short History, supra note 2; Sheppard & Leitner, supra note 4.

(37.) See, e.g., Newcombe & Paradell, supra note 6, at 78; See also Dolzer & Schreuer, supra note 9, at 5. See Lauge N. Skovgaard Poulsen, Bounded Rationality and Economic Diplomacy: The Politics of Investment Treaties in Developing Countries 10, 17 (Ist ed. 2015).

(38.) See Newcombe & Paradell, supra note 6, at 58-59.

(39.) See Vienna Convention on the Law of Treaties art. 31-32, opened for signature May 23, 1969, 1155 U.N.T.S. 331. See, e.g., Anthony Aust, Modern Treaty Law and Practice 207-12 (2d ed. 2013) (setting forth basic treaty interpretation rules under international law).

(40.) See, e.g., Poulsen, supra note 37, at 115-16 (describing use of liberalization requirements in investment chapters in recent preferential trade agreements, as well as in investment agreements of United States, Canada and Japan). See also Kenneth J. Vandevelde, Sustainable Liberalism and the International Investment Regime, 19 Mich. J. Int'l L. 373 (1998) (describing close links between liberal economic ideology and international investment laws).

(41.) See Jason Webb Yackee, Controlling the International Investment Law Agency, 53 Harv. Int'l. L.J. 391, 401-407 (2012) (describing characteristics of investment arbitrators social network). See also Sergio Puig, Social Capital in the Arbitration Market, 25 Eur. J. Int'l L. 387, 403-23 (2014) (analyzing social network of investment arbitrators while describing main characteristics of investment arbitration community).

(42.) See id. at 411-12 (presenting empirical analysis of dozens of investment arbitration cases and finding closed community with several "power-brokers" dominating investment arbitration community). Similar conclusions can be drawn from the data collected for the current research, although a thorough discussion regarding this subject is beyond the scope of this paper. Id.

(43.) See Moshe Hirsch, The Sociology of International Investment Law The Foundations of International Investment Law 142, 146-47, 163-67 (Zachary Douglas, Joost Pauwelyn, & Jorge E. Vinuales eds., 2014) [hereinafter Hirsch, Sociology] (discussing social control mechanisms in investment arbitration community which call for conformism). See also Moshe Hirsch, Invitation to the Sociology of International Law 142-53 (2015) [hereinafter Hirsch, Invitation]; Puig, supra note 41, at 400. "While reputation, persuasion, collegiality, and deference may all play some role in this close-knit community, conformity pressures are also probably common." See Puig, supra. See also Yackee, supra note 41, at 407 ("A small, relatively closed I IL community is more likely relatively ideologically cohesive and better able to coordinate its policymaking efforts."); See also Tom Ginsburg, The Culture of Arbitration, 36 Vand. J. Transnat'l. L. 1335, 1337-38 (2003). "The culture of arbitration typically refers to the gradual convergence in norms, procedures and expectations of participants in the arbitral process." Id. See also Stavros Brekoulakis, Systemic Bias and the Institution of International Arbitration: A New Approach to Arbitral Decision-Making, 4 J. Int'l Disp. Settlement 553 (2013); Daphna Kapeliuk, Social Capital and Arbitral Decision Making, EJIL: Talk! (Sep. 24, 2009), available at http://www.ejiltalk.org/social-capital-and-arbitral-decision-making. "[A]n arbitrator who is part of a closed group of arbitrators, who are repeatedly appointed and appoint each other to sit on panels, may have a strong sense of commitment to peers, which may affect his decision making, and as a consequence the outcome of the dispute." See Kapeliuk, supra.

(44.) See Yackee, supra note 41, at 432-34. "The literature cites only a handful of allegedly inconsistent decisions among the many awards that have been issued to date, and furthermore it is not entirely clear that the cited examples are actually unexplainably inconsistent." Id. See also Jan Paulsson, Avoiding Unintended Consequences, Appeals Mechanism in International Investment Disputes 241, 244-52 (Karl P. Sauvant & Michael Chiswick-Patterson eds., 2008) (arguing only small number of cases resulted in contradicting decisions cannot be explained by variations of relevant factors).

(45.) See Puig, supra note 41, at 403-23. See also Yackee, supra note 41, at 401-407. Some theoretical support to such expectation may be found in research showing that interactions between individuals may affect their perceptions of justice. See also Quinetta M. Roberson & Jason A. Colquitt, Shared and Configurai Justice: A Social Network Model of Justice in Teams, 30 Acad. Manage. Rev. 595-607 (2005).

(46.) See, e.g., Redfern & Hunter, supra note 5, 1 8.59; Dolzer & Schreuer, supra note 9, at 119; Tudor, supra note 10, at 123-132; Dolzer, supra note 5, at 10. See also Kenneth W. Abbott, Robert O. Keohane, Andrew Moravcsik, Anne-Marie Slaughter & Duncan Snidal, The Concept of Legalization, 54 Int'l Org. 401, 412-15 (2000) (discussing use of rules with low precision in legal systems and international law).

(47.) See supra note 8 and accompanying text.

(48.) See Hirsch, Sociology, supra note 43, at 153-54. See also Hirsch, Invitation, supra note 43, at 146.

(49.) See infra note 71 and accompanying text.

(50.) See, e.g., Stiglitz, supra note 1, at 74. "Because in this model there is no need for government-that is free, unfettered, 'liberal' markets work perfectly-the Washington Consensus policies are sometimes referred to as 'neo-liberal,' based on 'market fundamentalism,' a resuscitation of the laissez-faire policies that were popular in some circles in the nineteenth century." Id. Jamie Peck & Adam Tickell, Neoliberalizing Space, 34 Antipode 380,381 (2002) (arguing through Washington Consensus). "Neoliberalism has provided a kind of operating framework or 'ideological software' for competitive globalization, inspiring and imposing far-reaching programs of state restructuring and rescaling across a wide range of national and local contexts." Id. See also John Williamson, Democracy and the "Washington Consensus," 21 World Dev. 1329-36 (1993) (rebutting argument Washington Consensus should be affiliated with neoliberal ideologies).

(51.) See, e.g., Williamson, Short History supra note 2, at 9.

(52.) See id.

(53.) See Stiglitz, Post-Washington Consensus, supra note 3, at 53.

(54.) See Williamson, supra note 7, at 40.

(55.) See Williamson, What Washington Means, supra note 2.

(56.) See supra note 8 and accompanying text.

(57.) See infra notes 59-64, 71 and accompanying text.

(58.) See infra notes 65-68 and accompanying text.

(59.) See Jeffrey P. Commission, Precedent in Investment Treaty Arbitration-A Citation Analysis of a Developing Jurisprudence, 24 J. Int'l Arb. 129 (2007) (examining role of precedent in investment arbitration based on over 200 decisions); Ole Kristian Fauchaid, The Legal Reasoning of ICSI D Tribunals- An Empirical Analysis, 19 Eur. J. Int'l L. 301 (2008) (examining interpretive arguments of ICSID tribunals based on dozens of cases and decisions, and examining their role on development of general international law).

(60.) See Franck, Development, supra note 12.

(61.) See Franck, Politics, supra note 12.

(62.) See Franck & Wylie, supra note 12.

(63.) See id. at 516, n.213.

(64.) See id. at 521-22.

(65.) See Franck, ICS1D Effect, supra note 12, at 398-399.

(66.) See Suha Jubran Bailan, Investment Treaty Arbitration and Institutional Backgrounds: An Empirical Study, 34 Wis. Int'l L.J. 31, 85-87 (2016).

(67.) See Kapeliuk, Patterns, supra note 12, at 70.

(68.) See Sergio Puig & Anton Strezhnev, Affiliation Bias in Arbitration: An Experimental Approach, (Aug. 26, 2016) (unpublished manuscript) (on file https:// ssrn.com/abstract=2830241).

(69.) See Van Harten, supra note 12, at 222-23 "Data on outcomes as a measure of actual behaviour are open to a wide range of possible explanations such as variations in the strength of parties' claims, [and] diversity of fact situations...."

(70.) See Jonas Jacobson, Jasmine Dobbs-Marsh, Varda Liberman & Julia A. Minson, Predicting Civil Jury Verdicts: How Attorneys Use (and Misuse) a Second Opinion, 8 J. Empirical Legal Stud. 99,113-115 (finding attorneys' estimates of verdicts were substantially more accurate than law students' estimates).

(71.) See Michael Waibel & Yanhui Wu, Are Arbitrators Political? Evidence from International Investment Arbitration (Jan. 2017) (unpublished manuscript) (on file at https:// http://www-bcf.usc.edu/~yanhuiwu/arbitrator.pdf).

(72.) See id. at 29-34.

(73.) See id. at 16.

(74.) See id. at 19.

(75.) See Van Harten, supra note 12.

(76.) See id. at 238.

(77.) See id. at 251.

(78.) See Investment Dispute Settlement Navigator, supra note 11.

(79.) See generally, Newly Posted Awards, Decisions & Materials, Italaw, http:// www.italaw.com/ (last visited Aug. 10, 2017).

(80.) See supra notes 25-29 and accompanying text.

(81.) Essentially this tests multiple hypotheses for each measure to influence the outcome of the FET claim, where the null hypotheses are that there is no effect to each of the separate challenged policies on the determination of a violation of FET. Thus, an important caveat is that testing multiple hypotheses increases the probability for rejecting the null hypotheses even if they are true. That is, a statistically significant result might be the mere result of chance. To address this issue, Bonferroni-adjusted p-values levels would be presented alongside the results of this model. The conservative Bonferroni adjustments should guard the analyses from the concern of rejecting the null hypothesis even if it is true, without making further assumptions. However, they might be more conservative than necessary. Therefore, the results presented in this section, inferring that certain measures and policies have a greater chance than others when using Bonferroni adjustments, should be conceived as very reliable. See generally Olive Jean Dunn, Multiple Comparisons Among Means, 56 J. Am. Stat. Ass'n 52 (1961) (describing strategy for addressing issue using Bonferroni adjustment); Neil J. Salkind, 1 Encyclopedia of research design 98-100 (2010) (describing uses of Bonferroni adjustment).

(82.) See Jeffrey M. Wooldridge, Introductory Econometrics: A Modern Approach 483 (5th ed. 2013).

(83.) See supra Part IV.A.

(84.) MFN clauses require a party to an IIA to afford investors of the other party the best treatment it affords investors of other states. Several tribunals interpreted this in a way that allowed them to import FET clauses into IIAs that do not include such clause. See, e.g., Patrick Dumberry, The Importation of the FET Standard through MFN Clauses: An Empirical Study of BITs, 32 ICSID Rev. 116 (2017).

(85.) See Franck (fe Wylie, supra note 12.

(86.) See Nick Gallus, The Influence of lhe Host State's Level of Development on International Investment Treaty Standards of Protection, 6 J. World Inv. Trade 711, 714 (2005); Tudor, supra note 10, at 164-65. Contra Dolzer, supra note 5, at 27 "Respect for a political history of instability or of hostility toward foreign investment, for instance, would not be considered to be conducive to the purposes of an investment treaty...." Id.

(87.) See Investment Dispute Settlement Navigator, supra note 11.

(88.) See, e.g., Tudor, supra note 10, at 128-29.

(89.) See, e.g, PSEG Global Inc. and Konya Ilgin Eletrik Uretim ve Ticaret Limited Sirketi v. Republic of Turkey, ICSID Case No. ARB/02/5, Objections to Jurisdiction, [paragraph] 241 (Jan. 19, 2007), https://www.italaw.com/sites/default/files/case-documents/ ita0695.pdf. "Legitimate expectations by definition require a promise of the administration on which the Claimants rely to assert a right that needs to be observed." See Dolzer, supra note 5, at 26.

(90.) See Michele Potesta, Legitimate Expectations in Investment Treaty Law: Understanding the Roots and the Limits of a Controversial Concept, 28 ICSID Rev. 88, 105-10 (2013).

(91.) See Wooldridge, supra note 82, at 256-57.

(92.) See id. at 596.

(93.) See, e.g., Franck, Evaluating, supra note 12; Franck, Development, supra note 12.

(94.) See Investment Dispute Settlement Navigator, supra note 11.

(95.) See id.

(96.) However, in the following list these conducts were marked in accordance with the tribunal's decision (referring to the outcome of each measure and not by the outcome of the case).

(97.) Alongside the list of measures appears the number of cases in which the conduct was challenged, as well the number of cases in which the conduct was challenged and the tribunal found a breach of the FET standard. For some groups, the number of alleged breaches is smaller than the sum of the conducts. This is because in this table each conduct is counted by the number of cases in which it appears, and in some cases, more than one conduct was alleged to breach the FET clause. It is worthwhile to note that this list is quite similar to one in a 2012 UNCTAD report on Fair and Equitable Treatment, possibly indicating that the current data is reliable. See UNCTAD, supra note 26, at 39-43.

(98.) See Franck & Wylie, supra note 12. When omitting the WCinconsistent and WCinconclusive variables, the Respondent Attorney Experience and Claimant Attorney Experience variables are found to have a significant effect on the outcomes of FET claims. This may imply an omitted variable bias, as discussed in part IV.A.

(99.) The large significant effect of the policies that are inconsistent with the Washington Consensus on the outcome of FET claims requires a further examination into the cases where policies that are inconsistent with the Washington Consensus were challenged but the claim failed. Such inquiry suggests that two main circumstances may explain decisions not to find policies that are inconsistent with the Washington Consensus in violation of FET. The first is whether states promoted health or environmental concerns, and the other is the threshold prescribed by the FET clause or by the developmental stage of the disputing states. Only 6 FET claims that challenged measures that are inconsistent with the Washington Consensus were discussed and dismissed. In these cases the tribunals also dismissed all other claims. Three of these cases involved claims of the respondent state that its actions were driven by health and environmental concerns. The tribunals have accepted these claims and therefore dismissed the FET claims as well as all other claims and found the disputes in favor of the state. During the past decade, international financial organizations have accepted health and environmental considerations as well. Thus, some tendency to accept policies that are inconsistent with the classical Washington Consensus when health and environmental concerns arise may reflect changes in preferences of the international investment regulation community. The remaining three cases include two cases that were dealt under the North American Free Trade Agreement (NAFTA). As claimed above, this may result from the restrictive interpretation NAFTA grants to the FET clause. Respectively, out of 7 cases that involved policies that were inconsistent with the Washington Consensus and were based on NAFTA, only 3 were in violation of FET. Notably, all these cases involved two developed countries (the U.S. and Canada), which may also affect the threshold that should be applied.

(100.) See supra note 81 and accompanying text.

(101.) See Investment Dispute Settlement Navigator, supra note 11.

(102.) See Stiglitz, supra note 3, at 53; Ocampo, supra note 3, at 89.

(103.) See Tomer Broude et al., Legitimation Through Renegotiation: Do States Seek More Regulatory Space in Their BITs? (Jan. 2, 2017), https://ssrn.com/ abstract=2845297.

(104.) See Trans-Pacific Partnership Art. 29.5, (Feb. 4, 2016), United Sates Trade Representative, https://ustr.gov/trade-agreements/free-trade-agreements/ trans-pacific-partnership/tpp-full-text. See also Sergio Puig & Gregory Shaffer, A Breakthrough with the TPP: The Tobacco Carve-out, 16 Yale J. Health Pol'y L. & Ethics 4 (2017).

David Chriki, LL.M, LL.B, The Hebrew University of Jerusalem - Faculty of Law. I am grateful for the valuable guidance and comments I received from Tomer Broude and Yoram Haftel. I also thank Daphna Kapeliuk, Ari Afilalo, Olga Frishman, Ifat Taraboulos, and Edward Worbis for their sincere comments and helpful conversations. Naomi Hausman contributed significant insights that developed the empirical assessment presented in this paper. This research enjoyed the kind support of the Feinberg Fund for International Law at the Law Faculty of the Hebrew University of Jerusalem.

Caption: Figure 1: Acceptance of FET Claims by State Conduct
Table 1: List of Identified State Conducts

Group                      Conduct

Contracts & concessions
                           Termination or breach or annulment of a
                           contract
                           Disqualifying hid in a tender
                           Enforcement of pledge
                           Restructuring treasury bills
Fines, imprisonment &
criminal investigations
Judicial Decision
                           Bankruptcy Measures
                           Annulment of lease/license
                           Constitutional
                           Not enforcing an arbitral award
                           Other
Licenses & Permits
                           License & permits Revocation
                           Denial of permits
                           Rezoning
                           Enforcing mandatory standards & bans
Monetary Policy
Narrowing Investment
Incentives
Hindering Property
Rights
                           Imposition of Temporary Administrator
                           Taking over assets by government (not
                           including taxes)
Tax Measures
                           Tax Increase
                           Tax Collection
Trade- Imposing Trade
Regulation
                           Price Regulation
                           Production & Export Curtailments
                           Import regulations & restriction
Trade--Refraining from
Trade Regulation
                           Failure to ensure exclusivity as
                           promised Failure to regulate prices as
                           promised

Group                      Num. of Cases   FET Breached
                                           (percent of total)

Contracts & concessions    63              38 (60%)
                           57              36 (63%)

                           3               1 (33%)
                           2               0 (0%)
                           1               1 (100%)
Fines, imprisonment &      17              11 (65%)
criminal investigations
Judicial Decision          36              12 (33%)
                           12              3 (25%)
                           8               2 (25%)
                           5               3 (60%)
                           4               1 (25%)
                           7               3 (43%)
Licenses & Permits         42              16(38%)
                           20              8 (40%)
                           17              7 (41 %)
                           11              2(18%)
                           5               0(0%)
Monetary Policy            15              13 (87%)
Narrowing Investment       8               8(100%)
Incentives
Hindering Property         20              16 (80%)
Rights
                           5               5 (100%)
                           17              13 (76%)

Tax Measures               16              10 (62%)
                           11              10(91%)
                           5               0 (0%)
Trade- Imposing Trade      30              24 (80%)
Regulation
                           20              19 (95%)
                           6               5 (83%)
                           6               3 (50%)
Trade--Refraining from     2               0 (0%)
Trade Regulation
                           1               0 (0%)

                           1               0 (0%)

Table 2: Effects of Policies Inconsistent with the
Washington Consensus on the Finding of a
Violation of FET--Basic Model

                                 (1)         (2)
VARIABLES                        LPM         LPM -
                                             Excl. Arg.

WCinconsistent                   0.456 ***   0.440 ***
                                 (0.0807)    (0.0865)
WCinconclusive                   0.140       0.121
                                 (0.107)     (0.109)
Respondent AttorneyExperience    0.00999     0.00391
                                 (0.00653)   (0.00962)
Claimant AttorneyExperience      0.0110      0.0112
                                 (0.00662)   (0.00865)
ClaimantCorporation              0.0975      0.0924
                                 (0.109)     (0.110)
Constant                         -0.0465     -0.00150
                                 (0.164)     (0.175)
Observations                     120         104
R-squared                        0.295       0.218

                                 (3)          (4)
VARIABLES                        Probit       Probit -
                                              Excl. Arg.

WCinconsistent                   1.281 ***    1.194 ***
                                 (0.257)      (0.260)
WCinconclusive                   0.454        0.352
                                 (0.312)      (0.301)
Respondent AttorneyExperience    0.0400       0.0123
                                 (0.0253)     (0.0300)
Claimant AttorneyExperience      0.0438*      0.0350
                                 (0.0239)     (0.0274)
ClaimantCorporation              0.292        0.289
                                 (0.351)      (0.344)
Constant                         -1.707 ***   -1.444 ***
                                 (0.566)      (0.554)
Observations                     120          104
R-squared

Robust standard errors clustered by respondent state in parentheses
* p<.05; ** p<.01; *** p<.001

Table 3: Effects of Policies Inconsistent with the
Washington Consensus on the Finding of a
Violation of FET--Expansive Model

                                 (1)          (2)
VARIABLES                        L.PM         l.PM -
                                              Excl. Arg.

WCinconsistent                   0.359 ***    0.348 ***
                                 (0.0794)     (0.0853)
WCinconclusive                   0.0759       0.0854
                                 (0.110)      (0.113)
Respondent AltorneyExperience    0.00858      0.0156
                                 (0.00776)    (0.00979)
ClaimantAttorneyExperience       0.00860      0.0119 *
                                 (0.00579)    (0.00697)
ClaimanlCorporation              0.127        0.129
                                 (0.104)      (0.107)
OtherViolations                  0.425 ***    0.476 ***
                                 (0.0808)     (0.0822)
Respondent Deve loped            -0.124 *     -0.131 *
                                 (0.0664)     (0.0660)
Health_Enviroment                -0.0401      -0.0365
                                 (0.106)      (0.106)
Crisis                           0.0385       0.113
                                 (0.104)      (0.152)
FET_NAFTA_PostFTC                -0.235 ***   -0.229 ***
                                 (0.0590)     (0.0686)
FET_qualified                    -0.0602      -0.0556
                                 (0.0709)     (0.0916)
Constant                         -0.0166      -0.0911
                                 (0.172)      (0.181)
Observations                     118          102
R-squared                        0.565        0.533

                                 (3)          (4)
VARIABLES                        Probit       Probit -
                                              Excl. Arg.

WCinconsistent                   1.649 ***    1.549 ***
                                 (0.350)      (0.342)
WCinconclusive                   0.598        0.523
                                 (0.545)      (0.531)
Respondent AltorneyExperience    0.0592 *     0.0619 *
                                 (0.0345)     (0.0353)
ClaimantAttorneyExperience       0.0657 **    0.0590 *
                                 (0.0299)     (0.0311)
ClaimanlCorporation              0.450        0.468
                                 (0.517)      (0.514)
OtherViolations                  2.012 ***    1.951 ***
                                 (0.395)      (0.388)
Respondent Deve loped            -0.621 *     -0.613 *
                                 (0.333)      (0.327)
Health_Enviroment                -0.166       -0.165
                                 (0.460)      (0.452)
Crisis                           0.274        0.339
                                 (0.574)      (0.709)
FET_NAFTA_PostFTC                -1.120 ***   -1.128 ***
                                 (0.322)      (0.322)
FET_qualified                    -0.304       -0.328
                                 (0.329)      (0.336)
Constant                         -2.683 ***   -2.536 ***
                                 (0.990)      (0.972)
Observations                     118          102
R-squared

Robust standard errors clustered by respondent state in parentheses
* p<.1; ** p<.05; *** p<.01

Table 4: Testing for effects of Different Washington
Consensus Definitions

           (1)          (2)         (3)          (4)        (5)
           Imposing     Hindering   Narrowing    Tax        Monetary
           trade        property    investment   increase   policy
           regulation   rights      incentives

Option 1   V            V           V            V          V

Option 2   V            V           V            V          V

Option 3   V            V           V            V          X

Option 4   V            V           V            X          V

Option 5   V            V           V            X          X

Option 6   V            V           V            V          X

Option 7   V            V           V            X          V

Option X   V            V           V            X          X

           (6)          (7)           (8)
           Privatized   Basic excl.   Expansive
           asset        Arg.          excl. Arg.

Option 1   V            0.418 ***     0.345 ***
                        (0.0895)      (0.0871)
Option 2   X            0.406 ***     0.328 ***
                        (0.100)       (0.0815)
Option 3   V            0.418 ***     0.345 ***
                        (0.0895)      (0.0871)
Option 4   V            0.440 ***     0.348 ***
                        (0.0865)      (0.0853)
Option 5   V            0.440 ***     0.348 ***
                        (0.0865)      (0.0853)
Option 6   X            0.406 ***     0.328 ***
                        (0.100)       (0.0815)
Option 7   X            0.421 ***     0.322 ***
                        (0.0925)      (0.0815)
Option X   X            0.421 ***     0.322 ***
                        (0.0925)      (0.0815)

Robust standard errors clustered by respondent state in parentheses
* p<.1; ** p<.05; *** p<.01

Table 5: Effects of Challenged Policies on the
Finding of a Violation of FET--Basic Model

                                 (1)          (2)          (3)
VARIABLES                        LPM          LPM -        Probit
                                              excl. Arg.

TradeRegulation                  0.187        0.120        0.842 *
                                 (0.128)      (0.140)      (0.480)
PropertyRights                   0.411 ***    0.416 ***    1,492 ***
                                 (0.104)      (0.108)      (0.424)
NarrowingInvestmentIncentives    0.151        0.205
                                 (0.216)      (0.285)
PrivatizedAsset                  0.255***     0.255 ***    0.898 ***
                                 (0.0896)     (0.0938)     (0.309)
ContractsConccesions             0.139 *      0.127        0.636 **
                                 (0.0743)     (0.0866)     (0.299)
LiccnseandPermits                -0.127       -0.113       -0.243
                                 (0.0853)     (0.0843)     (0.321)
JudicialDecision                 -0.140       -0.129       -0.307
                                 (0.113)      (0.125)      (0.412)
TaxCollection                    -0.284 ***   -0.285 **
                                 (0.101)      (0.106)
Taxlncrease                      0.200        0.272        1.101
                                 (0.207)      (0.257)      (0.778)
MonetaryPolicy                   -0.0881                   0.245
                                 (0.135)                   (0.480)
Criminalmeasures_fine            0.186*       0.196 *      0.776*
                                 (0.104)      (0.103)      (0.406)
Respondent AttorneyExperience    0.0105       0.00871      0.0290
                                 (0.00852)    (0.0106)     (0.0331)
ClaimanlAttorneyExperience       0.00525      0.00633      0.0187
                                 (0.00787)    (0.0102)     (0.0291)
ClaimantCorporation              0.0770       0.0859       0.315
                                 (0.102)      (0.105)      (0.429)
Constant                         0.173        0.158        -1.373 **
                                 (0.131)      (0.149)      (0.571)
Observations                     120          104          109
R-squared                        0.411        0.354

                                 (4)
VARIABLES                        Probit -
                                 excl. Arg.

TradeRegulation                  0.570
                                 (0.539)
PropertyRights                   1.458 ***
                                 (0.424)
NarrowingInvestmentIncentives

PrivatizedAsset                  0.858 **
                                 (0.333)
ContractsConccesions             0.485
                                 (0.309)
LiccnseandPermits                -0.223
                                 (0.340)
JudicialDecision                 -0.324
                                 (0.422)
TaxCollection

Taxlncrease                      1.031
                                 (0.783)
MonetaryPolicy

Criminalmeasures_fine            0.765 *
                                 (0.392)
Respondent AttorneyExperience    0.0184
                                 (0.0378)
ClaimanlAttorneyExperience       0.0162
                                 (0.0313)
ClaimantCorporation              0.366
                                 (0.433)
Constant                         -1.255 **
                                 (0.586)
Observations                     94
R-squared

Robust standard errors clustered by respondent state

* p<.1; ** p<.05; *** p<.01

Table 6: Effects of Challenged Policies on the
Finding of a Violation of FET--Expansive

Model

                                 (1)          (2)
VARIABLES                        LPM          LPM -
                                              Excl. Arg.

TradeRegulation                  0.232 ***    0.216 **
                                 (0.0764)     (0.0886)
PropertyRights                   0.299 ***    0.284 ***
                                 (0.0972)     (0.103)
NarrowingInvestmentIncentives    0.127        0.175
                                 (0.155)      (0.191)
PrivatizedAsset                  0.218 **     0.225 **
                                 (0.0817)     (0.0901)
ContractsConccesions             0.118*       0.108
                                 (0.0605)     (0.0735)
LicenscandPermits                -0.115       -0.126
                                 (0.0703)     (0.0767)
JudicialDecision                 -0.0112      -0.00765
                                 (0.120)      (0.132)
TaxCollection                    -0.138       -0.112
                                 (0.120)      (0.132)
Taxlncrease                      0.205        0.155
                                 (0.144)      (0.173)
MonetaryPolicy                   -0.321 *
                                 (0.167)
Criminalmeasures_fine            0.207**      0.204 *
                                 (0.0972)     (0.102)
Respondent AttorneyExperience    0.0172 **    0.0216 **
                                 (0.00821)    (0.00914)
ClaimantAttorneyExperience       0.00571      0.00791
                                 (0.1X1623)   (0.00782)
ClaimantCorporation              0.140        0.131
                                 (0.113)      (0.114)
OtherViolations                  0427 ***     0.467 ***
                                 (0.0793)     (0.0851)
RespondentDeveloped              -0.0587      -0.0688
                                 (0.0741)     (0.0786)
Health_Enviroment                0.0846       0.0893
                                 (0.107)      (0.108)
Crisis                           0.0878       0.0790
                                 (0.152)      (0.170)
FET_NAFTA_PostFTC                -0.195 **    -0.176 *
                                 (0.0825)     (0.0899)
FET_qualified                    -0.0830      -0.0647
                                 (0.0843)     (0.101)
Constant                         -0.0509      -0.0847
                                 (0.137)      (0.147)
Observations                     118          102
R-squared                        0.619        0.579

                                 (3)          (4)
VARIABLES                        Probit       Probit -
                                              Excl. Arg.

TradeRegulation                  1.680 ***    1.535 **
                                 (0.628)      (0.726)
PropertyRights                   1.363 ***    1.328 ***
                                 (0.452)      (0.453)
NarrowingInvestmentIncentives

PrivatizedAsset                  1.311 ***    1.256 **
                                 (0.488)      (0.515)
ContractsConccesions             0.709        0.647
                                 (0.492)      (0.532)
LicenscandPermits                -0.554       -0.563
                                 (0.437)      (0.448)
JudicialDecision                 0.0408       0.00842
                                 (0.561)      (0.570)
TaxCollection

Taxlncrease                      0.881        0.808
                                 (0.855)      (0.914)
MonetaryPolicy                   -1.636 *
                                 (0.945)
Criminalmeasures_fine            1.232 ***    1.200 ***
                                 (0.425)      (0.414)
Respondent AttorneyExperience    0.104 **     0.102 **
                                 (0.0468)     (0.0458)
ClaimantAttorneyExperience       0.0224       0.0215
                                 (0.0373)     (0.0380)
ClaimantCorporation              0.336        0.349
                                 (0.518)      (0.517)
OtherViolations                  2.177 ***    2.124 ***
                                 (0.357)      (0.364)
RespondentDeveloped              -0.442       -0.434
                                 (0.445)      (0.438)
Health_Enviroment                0.536        0.520
                                 (0.459)      (0.454)
Crisis                           0.558        0.531
                                 (0.794)      (0.805)
FET_NAFTA_PostFTC                -0.742 *     -0.730 *
                                 (0.448)      (0.434)
FET_qualified                    -0.405       -0.384
                                 (0.388)      (0.388)
Constant                         -2.694 ***   -2.595 ***
                                 (0.819)      (0.870)
Observations                     107          92
R-squared

Robust standard errors clustered by respondent state

* p<.1; ** p<.05; *** p<.01
COPYRIGHT 2018 Suffolk University Law School
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2018 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:fair and equitable
Author:Chriki, David
Publication:Suffolk Transnational Law Review
Date:Jun 22, 2018
Words:16339
Previous Article:ADMINISTRATIVE LAW--ELEPHANT TROPHY IMPORTATION BAN CORRECTLY SURVIVES CHALLENGE FROM AN INTERNATIONAL HUNTING CLUB.
Next Article:ENFORCING THE UNENFORCEABLE: HOW TO RELY ON INTERNATIONAL LAW TO CURB CHINA FROM ILLEGAL TERRITORIAL CLAIMS IN THE SOUTH CHINA SEA.
Topics:

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters