Authors: Henrique Carneiro*
For decades, arbitral tribunals and legal commentators have debated about the proper way to interpret the word “obligation” or “commitment” in international investment treaty umbrella clauses. The plain text of such clauses seems, at first glance, to classify as a treaty violation even the most run-of-the-mill contract breaches between sovereigns and foreign investors. Consider the following example: “Either Contracting Party shall constantly guarantee the observance of the commitments it has entered into with respect to the investments of the investors of the other Contracting Party.”
The wording in these clauses differs on the margins—“guarantee” vs. “observe,” “obligation” vs. “commitment,” etc.—but they are all susceptible to a capacious, plain text-based construction that would have far-reaching economic consequences. There seems to be a mismatch between, on the one hand, the apparent simplicity, lack of ambiguity, and non-prominence of these clauses, and on the other hand, the far-reaching effects on state sovereignty that would result from the application of their plain terms.
It is therefore not surprising that many tribunals have decided against such literal interpretations. But perhaps due to the lack of a central, authoritative appellate body, some arbitral tribunals have gone in the opposite direction, unabashedly adopting the literal and capacious interpretations. Professor Julian Arato has summarized the jurisprudential approaches to answering this legal question. As he explains:
[T]here seem to be three main approaches relevant to the question of how far the umbrella clause transforms the municipal state contract into an instrument of international law: one, the outlier, being exceedingly restrictive; another completely embracing the internationalization thesis; and the third adopting a more nuanced, hybrid approach. The three approaches are nicely captured by a trio of cases decided between 2003 and 2012, brought by the same claimant, the Swiss customs inspection company Société Générale de Surveillance (“SGS”), against three different states (Pakistan, the Philippines, and Paraguay).
The purpose of this blog post is to advance a modest but concrete methodological solution to this legal question, which I call the empirics-based hybrid approach. This “hybrid” approach to umbrella clause interpretation would be based primarily on readily discoverable empirical facts about the investment relationship between parties to the applicable international investment agreement. And the following rule of construction should emerge from such an inquiry: when the size of the investment relationship between the two parties is relatively large, it is unlikely that a basic umbrella clause like the one referenced above encompasses run-of-the-mill breaches of contract by the sovereign. In contrast, when the size of the investment relationship between the two parties is relatively small, it is more likely that such an umbrella clause applies to breaches of contract—no matter how run-of-the-mill these may be, and regardless of any conflicting contractual forum selection clauses between foreign investors and host countries. The mechanics of this approach, as well as its possible variations in practice, will be discussed in more depth below.
It is “modest,” because, as I will explain, it should only be deployed in limited settings. This “hybrid” approach should be applied in the specific case where (1) a tribunal is tasked with interpreting an umbrella clause that is similar to those I have cited above, and (2) an inquiry into the drafting context (or legislative history, so to speak) of the clause does not yield clear answers about the drafters’ intent. Only when these two conditions are fulfilled should tribunals explicitly consider the size of the investment relationship between the parties to the investment treaty in attempting to interpret the term “obligation” in an umbrella clause.
When the approach does apply, it can serve as a middle ground between the aforementioned, seemingly incompatible “restrictive” and “internationalizing” schools of thought, because its principled application will sometimes result in a restrictive result, and other times in an internationalizing result. It accommodates seemingly conflicting jurisprudential positions in one single interpretive theory that is grounded in general principles of international law and treaty interpretation, as well as in the general principles underlying ICSID itself. This is why I am referring to it as “hybrid.”
The empirics-based hybrid approach should yield many benefits which will also be explained below in more depth. First, it should increase the likelihood of correct interpretation. Second, it is designed to produce a sovereignty-enhancing effect in the long term and thereby revitalize the umbrella clause as a behavior-incentivizing legal device that can be used by drafters and negotiators. Third, it can help to resolve the tension between finality and consistency that is inherent in international investment arbitration. And fourth, it can help to safeguard the credibility of international arbitration as a practice by signaling to nations and investors that arbitrators will seek to treat each umbrella clause as a completely unique provision, deserving of a full and searching interpretation, regardless of whether other clauses in other treaties contain identical wording. But in order for this proposal to be used by arbitral tribunals in a practical setting, some theoretical issues must be addressed, and I will also do so below.
It should also be said that this position assumes that umbrella clauses should not be categorically dismissed as mere rhetorical appeals, incapable of carrying an independent and legally enforceable substantive content.
The remainder of this blog post will consist of (1) a brief summary of the existing jurisprudential approaches to this topic; (2) an outline and theoretical explanation of the empirics-based hybrid approach I am advancing; (3) a discussion of the benefits of this proposal; and (4) answers to possible challenges that may reasonably be lodged against the proposal.
- Existing Approaches
The restrictive approach is characterized by the proposition that an umbrella clause cannot be interpreted so as to “transubstantiate” a breach of contract between the sovereign and the foreign investor into a violation of the BIT—a violation of international law—without “clear and persuasive evidence that such was in fact the intention of both [parties]” to the BIT. This has been referred to by Professor Arato as a “fringe position,” although he himself acknowledges two other arbitral awards whose tribunals have endorsed this reasoning.
The “nuanced, hybrid approach” is something of an abstention doctrine. Tribunals following this approach suspend judgment as to the content of an “obligation” or a “commitment,” deferring to the applicable municipal law for that determination. According to this view, what the umbrella clause governs is the performance or observance of the obligation or commitment once the content is determined by a local court. For example, in SGS v. Philippines, the Tribunal ruled that the relevant umbrella clause did confer jurisdiction over breach of contract claims, however, it decided that it:
should not exercise its jurisdiction over a contractual claim when the parties have already agreed on how such a claim is to be resolved, and have done so exclusively. SGS should not be able to approbate and reprobate in respect of the same contract: if it claims under the contract, it should comply with the contract in respect of the very matter which is the foundation of its claim. The Philippine courts are available to hear SGS’s contract claim. Until the question of the scope or extent of the Respondent’s obligation to pay is clarified […] a decision by this Tribunal on SGS’s claim to payment would be premature.
Finally, tribunals adopting the “internationalization” approach see “no basis on the face of the clause to believe that it should mean anything other than what it says,” and they view such a “plain text” interpretation as “necessary to give the clause purpose and effect.” Professor Jarrod Wong has also advanced this interpretation both (1) on interpretive grounds by arguing—like many tribunals and litigants have—that the umbrella clause would be meaningless if not interpreted strictly; and (2) on normative grounds, stating that “the host State stands to benefit from the adoption of BIT provisions such as the umbrella clause because they foster a more hospitable, and therefore, more attractive, environment for foreign investment.” In addition to SGS v. Paraguay, the ICSID tribunal in Noble Ventures, Inc. v. Romania provided support for this view. This case concerned a dispute between Noble Ventures, a Maryland-based company engaging in mining-related consulting services, and the Romanian State Ownership Fund arising out of a privatization agreement whereby Noble Ventures was to acquire the shares of Combinatul Siderurgic Resita, a state-owned mining company. There, the tribunal was tasked with interpreting the United States-Romania umbrella clause, which provided that “[e]ach Party shall observe any obligation it may have entered into with regard to investments.”
The tribunal emphasized the principle that the umbrella clause “introduces an exception to the general separation of States obligations under municipal and international law.” And it relied on both text and purpose in holding that:
a breach of the [share purchase agreement] is, as a matter of law, capable of constituting a breach, attributable to the Respondent, of the BIT by reason of the inclusion in the BIT of Article II(2)(c), and, in the judgment of the Tribunal, whatever, if any, limitation is to be placed on the meaning of “obligation” where that expression is used in Article II(2)(c) of the BIT (cf. paragraph 61 above), a breach that is attributable to the Respondent of Article 7.4.2 of the [share purchase agreement] would constitute a breach by the Respondent of Article II(2)(c) of the BIT.
- Introducing the empirics-based hybrid approach
As explained above, the empirics-based hybrid approach to interpreting umbrella clauses should be reserved for situations where the meaning of the clause cannot be conclusively elucidated with reference to drafting documents, context, and other purposive evidence. And it also assumes that the inquiry should not end just because the umbrella clause is phrased in simple and unambiguous terms. Article 31 of the Vienna Convention—regarded by the United States, at least as of 2017, as constituting “customary international law on the law of treaties”—provides that as a matter of law, textualism without context is not proper for treaty interpretation.
Arbitrators should first consider whether an umbrella clause can be categorized as a “simple umbrella clause,” like the one in the Philippines-Switzerland BIT or the one in the Romania-U.S. BIT. For example, the Guatemala-Netherlands BIT umbrella clause provides that “[e]ach Contracting Party shall observe any obligation it may have entered into with regard to investments of investors of the other Contracting Party.” As discussed, the Romania-U.S. BIT has a similar umbrella clause. And Saudi Arabia’s BITs with Germany, Switzerland, Belgium-Luxembourg, Korea, and Switzerland, provide that “[e]ach of the contracting parties shall ensure compliance with any other obligation, which it has in respect of its territory assumed with regard to investors of the other Contracting Party.”
In contrast, the Austria-Saudi Arabia BIT adds that “disputes under the umbrella clause ‘shall be decided, absent other agreement, in accordance with the law of the Contracting Party, party to the dispute, including its rules on the conflict of laws, the law governing the authorization or agreement and such rules of international law as may be applicable.’” The former, but not the latter, would satisfy this first step of the inquiry.
What this introductory point has shown is that in a manner that resembles the Chevron inquiry conducted by federal courts in the United States, arbitral tribunals would need to determine when the text of an umbrella clause is similar enough to these basic, broad provisions I have been citing in order to proceed to “step 2,” as it were, which in this case would be the assessment of empirical, economic considerations about the investment relationship between the BIT parties. Frequently, umbrella clause provisions are worded in such a way that they occupy a middle ground between provisions that are specific and unambiguous, and those that are very broad and capacious. So, a relatively difficult judgment call emerges with regards to these provisions that occupy a middle ground.
After determining that the clause can be categorized as a “simple umbrella clause,” and if the substantive scope of the provision cannot be readily determined through other means, a tribunal should turn to an assessment of the size of the investment relationship between the parties. Many methodological alternatives are possible. For example: tribunals may consider the number of companies from Contracting Party A that have investments in Contracting Party B, and vice versa. Tribunals may choose to consider all such investments, or only count investments that are valued above a minimum threshold. Another alternative would be to calculate the percentage of one BIT party’s share of the other BIT party’s foreign direct investment volume. It is because of this elevated amount of choice, that in the beginning of this essay I phrased the rule of construction in terms of a sliding scale rather than as consisting of discrete, bright-line standards.
Third, and most importantly, it must be determined whether, in order to assess the investment relationship between the parties to the investment treaty, tribunals may only consider the investment relationship at the time of the drafting and signing—an originalist analysis, so to speak. It might appear to follow logically that the empirics-based approach would require this, because the task is to use the size of an investment relationship to estimate the probability that the word “obligation” as it appears in an international investment treaty was intended to cover contracts. And this analysis would only make sense if the point of departure was the investment relationship during the period of drafting, negotiation, and signing. Any other approach would seem to pose insurmountable rule-of-law problems by allowing for the meaning of a clause to evolve over time absent any formal amendments to the treaty—a sort of “living” investment treaty, so to speak.
Although there undoubtedly are formidable rule-of-law problems any time the meaning of a textual provision is allowed to “evolve with the times,” the costs of such a characteristic do not necessarily outweigh its benefits in this specific context. Regardless, I do not here adopt a conclusion as to whether the “originalist” element is a non-negotiable aspect of an empirics-based interpretive approach.
- Benefits of the approach
First of all, because it is based in probability, and because it should only be invoked when meaning and intent cannot be clearly established through more traditional means, the empirics-based hybrid approach increases, even if only marginally, the likelihood of an objectively correct interpretation of the word “obligation” in a basic umbrella clause. If two nations that have a very sparse investment relationship sign a Bilateral Investment Treaty containing an umbrella clause, the clause must encompass breaches of contract if it is to mean anything at all. This is because, the smaller the investment relationship between the two nations, the smaller the total amount of individual “obligations” the Contracting state will owe to foreign investors from that other country. This increases the probability that each individual contract between the host state and the foreign investor was the object of the clause—indeed, it would be absurd to think otherwise, unless there is clear evidence to the contrary—evidence, for example, that the parties did not intend for the umbrella clause to create any independently enforceable rights.
Second of all, this approach has the potential to marginally increase consistency and predictability. Inconsistency and unpredictability has plagued umbrella clause litigation. While parties should not expect to predict results based on a quick glance at the text of an umbrella clause, adoption of this approach would offer at least a marginal increase in consistency by theoretically unifying seemingly inconsistent interpretations of similarly worded, or even identically worded, treaty provisions.
A third benefit of this approach is that it provides a high-resolution, non-strained path to differentiating between different BITs—even those that are very similarly worded. An example of a strained exercise in differentiating BITs can be found in a decision already discussed above—namely, the Noble Ventures decision. In that case, the tribunal rested a portion of its reasoning in support of an internationalizing interpretation on a distinction between (1) the text of the Romania-U.S. clause, which read “[e]ach Party shall observe any obligation it may have entered into with regard to investments,” and (2) the text of the Philippines-Switzerland clause, which read “[e]ach Contracting Party shall observe any obligation it has assumed with regard to specific investments in its territory by investors of the other Contracting Party,” which was the provision at issue in SGS v. Philippines. The tribunal wrote:
In the present case, the formulation adopted at Art. II(2)(c), which is even more general and straightforward than that in the bilateral investment treaty that fell to be considered in SGS v. Philippines, clearly falls into the category of the most general and direct formulations tending to an assimilation of contractual obligations to treaty ones; not only does it use the term “shall observe” but it refers in the most general terms to “any” obligations that either Party may have entered into “with regard to investments”.
To borrow from Professor Martin H. Redish, this argument is an example of “selective literalism.” Simply, it is difficult if not impossible to meaningfully distinguish the Romania-U.S. and the Philippines-Switzerland clauses on the basis of text alone. And one does not seem “more general and straightforward” than the other. An explicit and more searching inquiry into empirical facts about the investment relationship between the two parties to a BIT, however, would give tribunals the opportunity to avoid awkward textual arguments like this one.
Finally, a specifically non-originalist version of this approach would have a sovereignty-enhancing effect by allocating the costs of sovereignty restrictions onto parties that are least likely to actually appraise them as costs. One can imagine that reasonable state signatories might be open to being hailed into an ICSID tribunal to litigate contract breaches at Time 1, when perhaps there are not as many possible claimants and, by extension, less of a deleterious effect on sovereignty, while being against the availability of such claims at Time 2, at a more developed stage. Countries that share small investment relationships, on the other hand, will be comparatively better equipped to litigate run-of-the-mill contract disputes in arbitral tribunals. While they suffer an infringement on their sovereignty by having to litigate a decision to, for example, cancel a concession contract, these claims will also be fewer and further between. And in having to face such claims, they signal to potential investors that arbitration is viable, thereby attracting more investment. If this trend continues over time, the investment volumes will grow to such an extent that tribunals would, by applying the empirics-based hybrid model, no longer interpret the umbrella clause in that same way, thereby limiting the negative effects on sovereignty.
- Addressing Counterarguments and Challenges
The first objection that can be levied against this approach is that it does not remedy but rather contributes to the “inconsistent and problematic” jurisprudence surrounding umbrella clauses. How could it be that the same exact language should be interpreted differently in different cases, just based on the parties to the treaty?
The problem with this objection is that it inappropriately invokes principles underlying rule of law and statutory interpretation that, while very relevant to domestic lawmaking and legal interpretation, do not translate as well into the context of international law. We can take the United States as an example. In 1890, the United States Congress enacted the Sherman Antitrust Act, which contained, for example, a provision stating that “[e]very contract, combination in the form of trust or otherwise, or conspiracy, in restraint of trade or commerce among the several States, or with foreign nations, is declared to be illegal.” This provision is notoriously concise yet difficult to pin down. And there would be an obvious rule-of-law problem if the federal appellate circuits were to hand down diverging interpretations of this same provision. The U.S. Congress passes one law for the entire nation—it does not mean to, nor could it, enact one law for California, another for New York, another for Wyoming, and so on.
The same is not true of treaty language that is shared by more than one treaty. When two nations develop a BIT, they are creating a new and unique legal instrument, no matter how much overlap there may be between the language of this new instrument and that of another BIT that already exists in the world. To place dispositive weight on the particular wording of a BIT, and on the fact that it is identical to another instrument, would represent an undue attenuation of the interpretive process in the name of a fairness that is simply inapplicable in this context. Given that the same umbrella clause language has been interpreted in radically different ways in different treaties, it would not be reasonable at this stage for parties to expect arbitral tribunals to understand that their specific choice of wording means that they intended to mean the same exact thing as another BIT using that wording.
A second objection is that this approach asks arbitrators to engage in a “sea of doubt” by opining about topics that are outside their expertise. But a similar response as the one to the first objection is also available here. The democratic and philosophical principles underlying the various strains of judicial restraint in domestic contexts simply do not apply to ICSID arbitration, which is an opt-in system of dispute resolution based more on finality than on anything else. When nations consent to ICSID arbitration, they are knowingly submitting themselves to a system that has an extremely limited appellate system, and they are also subjecting themselves to the judgment of jurists from all around the world, each with a unique view of the law. These facts should temper concerns about judicial overreach and overconfidence.
Still a third objection might be that, because the aforementioned rule of construction that would emerge from this empirics-based hybrid approach is simultaneously specific and laden with standards, it should only be used if the parties have themselves clearly specified it in the treaty. My response to this objection is that each of the other jurisprudential approaches would also benefit from more unambiguous textual “help” from parties to investment treaties. And this proposed approach finds support in the dictates of the treaty interpretation provisions of the Vienna Convention, which encourage interpretive techniques that go above and beyond strict textualism.
The approach I propose in this essay is not intended to displace the entire field of umbrella clause interpretation. It is designed to be applied only in limited settings, as an epistemic aid to arbitral tribunals. And it is ultimately intended to offer an alternative to highly restrained textualist approaches.
Investment arbitration is an expensive ordeal; cases may go through multiple procedural stages, arbitrators are sometimes chosen from the opposite end of the globe and are hosted in luxurious hotels in very expensive cities, and astronomical amounts of money are usually at stake for the litigants. In light of these facts, it seems appropriate to encourage arbitrators to engage in a searching, explicit analysis of empirical economic data as an interpretive heuristic. The approach I advocate for does this, while building in methodological checks and balances to minimize the risk of judicial overreach. The alternative—allowing complicated cases to be disposed of at the jurisdictional stage on more superficial, literalist grounds—is unacceptable.
*Henrique Carneiro is a 2022 graduate of Northwestern Pritzker School of Law.
 Pakistan-Switzerland BIT, art. 11; see also, e.g., Guatemala-Netherlands BIT, art. 4(4) (“Each Contracting Party shall observe any obligation it may have entered into with regard to investments of investors of the other Contracting Party.”); Armenia-Austria BIT, art. 9(1) (using similar language).
 Julian Arato, Corporations as Lawmakers, 56 Harv. Int’l L.J. 229, 252 (2015) [hereinafter Arato, Corporations as Lawmakers] (citing Société Générale de Surveillance S.A. v. Pakistan, ICSID Case No. ARB/01/13 [hereinafter SGS v. Pakistan]; Société Générale de Surveillance S.A. v. Philippines, ICSID Case No. ARB/02/06, Decision on Jurisdiction (Jan. 29, 2004) [hereinafter SGS v. Philippines]; Société Générale de Surveillance S.A. v. Republic of Paraguay, ICSID Case No. ARB/07/29, Award (Feb. 10, 2012) [hereinafter SGS v. Paraguay]. In addition, Professor James Crawford has indicated that the level of dissensus between arbitrators on the proper scope of the standard umbrella clause is analogous to a “carpet [that] looks very much as if different people have started from different ends without many common threads—a crazy quilt rather than a Persian rug.” James Crawford, Treaty and Contract in Investment Arbitration, 24 Arb. Int’l 351, 353 (2008).
 See supra note 1.
 It is possible that this is the purpose of one or another particular umbrella clause. As Professor Crawford writes, “[t]here is no such thing as the umbrella clause; rather, there are umbrella clauses.” Crawford, supra note 2, at 355. It is not uncommon for statutes, constitutions, and other legal documents to contain legally unenforceable preambles, statements of policy, and legislative fact-finding. But the default rule is that “treaty language is presumed to have its natural and ordinary meaning in its context.” Id. See also Vienna Convention on the Law of Treaties, art. 31(1), opened for signature May 23, 1969, 1155 U.N.T.S. 331.
 SGS v. Pakistan, supra note 2, at ¶ 172 (“The modes by which a Contracting Party may ‘constantly guarantee the observance of’ its contractual or statutory or administrative municipal law commitments with respect to investments are not necessarily exhausted by the instant transubstantiation of contract claims into BIT claims posited by the Claimant”).
 Id. at ¶ 173.
 Arato, Corporations as Lawmakers, supra note 2, at 254.
 Id. at n.89 (citing El Paso Energy v. Argentina, ICSID Case No. ARB/03/15, Decision on Jurisdiction (Apr. 27, 2006); Pan American Energy LLC v. Argentine Republic, ICSID Case No. ARB/04/8, Decision on Preliminary Objections (July 27, 2006)). In addition, Professor Jarrod Wong discusses Joy Mining v. Egypt, a case in which the tribunal, in dicta, wrote that “it could not be held that an umbrella clause inserted in the Treaty, and not very prominently, could have the effect of transforming all contract disputes into investment disputes under the Treaty, unless of course there would be a clear violation of the Treaty rights and obligations or a violation of contract rights of such a magnitude as to trigger the Treaty protection[.]” See Jarrod Wong, Umbrella Clauses in Bilateral Investment Treaties: Of Breaches of Contract, Treaty Violations, and the Divide between Developing and Developed Countries in Foreign Investment Disputes, 14 Geo. Mason. L. Rev. 137, 161 (2006) (citing Joy Mining Mach. Ltd. v. Egypt, Award on Jurisdiction, ICSID (W. Bank) Case No. ARB/03/11 at ¶ 81 (2004)).
 Arato, Corporations as Lawmakers, supra note 2, at 252.
 See, e.g., SGS v. Philippines, supra note 2, at ¶ 155; see also Bureau Veritas, Inspection, Valuation, Assessment, and Control, BIVAC B.V. v. The Republic of Paraguay, Decision on Jurisdiction, ICSID Case No. ARB/07/9 (May 29, 2009), at ¶ 132-61 (finding jurisdiction but abstaining in favor of domestic courts).
 SGS v. Philippines, supra note 2, at ¶ 155.
 SGS v. Paraguay, Decision on Jurisdiction, supra note 2, at ¶ 168.
 Id. at ¶ 170.
 Wong, supra note 8, at 141.
 Noble Ventures, Inc. v. Romania, ICSID Case No. ARB/01/11, Award (Oct. 12, 2005).
 Id. at ¶ 1–31.
 Romania-United States BIT, art. II(2)(c).
 Noble Ventures v. Romania, supra note 15, at ¶ 55.
 Id. at ¶ 136.
 U.S. Department of State, Vienna Convention on the Law of Treaties, https://2009-2017.state.gov/s/l/treaty/faqs/70139.htm (last visited Nov. 24, 2021).
 Guatemala-Netherlands BIT, supra note 1.
 Overview, Investment Treaty Arbitration: Saudi Arabia, Global Arb. Rev., https://globalarbitrationreview.com/insight/know-how/investment-treaty-arbitration/report/saudi-arabia. (last visited Nov. 24, 2021) (citing BITs between Saudi Arabia and Germany, Switzerland, Belgium-Luxembourg Economic Union, Austria, Korea, and Switzerland).
 Chevron U.S.A., Inc. v. Natural Resources Defense Council, Inc., 467 U.S. 837 (1984).
 See, e.g., Susan D. Franck, The Legitimacy Crisis in Investment Treaty Arbitration: Privatizing Public International Law Through Inconsistent Decisions, 73 Fordham L. Rev. 1521, 1568–74 (2005) (characterizing the SGS decisions as “inconsistent”). At the time Professor Franck’s article was published, the SGS v. Paraguay decision had yet to be published, and, as described previously, that decision only added to the sense of inconsistency.
 I will explain why tribunals should not necessarily treat like provisions equally in the next section.
 Philippines-Switzerland BIT, art. X(2).
 Noble Ventures v. Romania, supra note 15, at ¶ 59–60.
 Martin H. Redish, Text, Structure, and Common Sense in the Interpretation of Article III, 138 U. Pa. L. Rev. 1633, 1635–36 (1990) (criticizing a different scholar’s argument for its undue reliance on several words in a constitutional provision).
 Jonathan B. Potts, Stabilizing the Role of Umbrella Clauses in Bilateral Investment Treaties: Intent, Reliance, and Internationalization, 51 Va. J. Int’l L. 1005, 1028 (2011).
 15 U.S.C. § 1.
 In one of the earliest decisions interpreting the Sherman Act, then-Judge William Howard Taft declined to adopt an interpretation of the Sherman Act that would require an assessment of the reasonableness of a particular price that was fixed by conspirators. In doing so, he wrote: “It is true that there are some cases in which the courts, mistaking, as we conceive, the proper limits of the relaxation of the rules for determining the unreasonableness of restraints of trade, have set sail on a sea of doubt, and have assumed the power to say, in respect to contracts which have no other purpose and no other consideration on either side than the mutual restraint of the parties, how much restraint of competition is in the public interest, and how much is not. The manifest danger in the administration of justice according to so shifting, vague, and indeterminate a standard would seem to be a strong reason against adopting it.” United States v. Addyston Pipe & Steel Co., 85 F. 271, 283-284 (6th Cir. 1898).
 When the size of the investment relationship between the two parties is relatively large, it is unlikely that a basic umbrella clause like the one referenced above encompasses run-of-the-mill breaches of contract by the sovereign. In contrast, when the size of the investment relationship between the two parties is relatively small, it is more likely that such an umbrella clause applies to breaches of contract.