A. The Pre-IIA Era

B. Enter the IIA Era

One might have thought that tribunals adjudicating IIA claims could shed references to lawful or unlawful expropriations. Those meeting the criteria in the treaty would either not lead to litigation or not lead to any liability; and expropriations in violation of the treaty would be identified as treaty violations first and foremost. The remedy for those violations might be a based on a lex specialis for investment law (even in the treaty itself) or on the fallback position of customary law, but the lodestar of the analysis would be that the treaty had been violated.Footnote ³⁶

Yet counsel to investor-state disputes and many tribunals still deploy the distinction terminologically and substantively. The range of tribunal views, and the diversity of investments valued in current arbitrations, makes each case unique and grouping them somewhat difficult.Footnote ³⁷ Moreover, we can never be certain whether the stated position of the tribunal is actually doing any work in arriving at the final damages number given the mystery involved in those figures. Nonetheless, four basic positions seem to have emerged:Footnote ³⁸

The modern IIA cases thus evidence a significant range of approaches to the use of internal versus external standards regarding compensation for expropriation and the relevance of the lawful/unlawful distinction. Nonetheless, most tribunals distinguish between a lawful and unlawful expropriation, and those tribunals doing so equate “unlawful” with an expropriation in violation of the IIA's criteria. The other uses suggested in the earlier case law (e.g., a violation of custom, or a contract) have faded. Still, most tribunals are awarding the treaty formula of FV_DE for expropriations in violation of the IIA—either without explanation (Group 1); because they believe that that amount is also the amount due for an expropriation in violation of the treaty (Group 2); or because the facts of the case make FV_DE the most sensible award level, in particular because the asset has not increased in value since the taking (Group 3).Footnote ⁶⁸ Only three cases—ADC, Yukos, and Quiborax—have used the distinction to award a higher degree of damages than FV_DE, i.e., FV_DA or something like it.Footnote ⁶⁹

One important wrinkle in these cases is that when tribunals purport to apply FV_DE, they are not consistent in whether they are using only information available at the date of the expropriation or information that has become available since that time, in particular when they are calculating the DCF of the investment.Footnote ⁷⁰ Though DCFs based on both available and ex post information are still discounted back to the date of expropriation in calculating FV_DE, those numbers can differ if market conditions change unpredictably after the expropriation. Full value on the date of expropriation using ex post information has been endorsed as early as Amco Asia v. Indonesia Footnote ⁷¹ (which did not concern an IIA) and as recently at Quiborax.Footnote ⁷² Both cases considered that amount to be the quantum of damages required for unlawful acts as a matter of customary international law. However, sometimes tribunals seem to rely on ex ante information for some inputs and ex post information for other inputs into FV_DE.Footnote ⁷³ Academic and other commentaries differ on whether ex post information can be considered for purposes of determining FV_DE.Footnote ⁷⁴ I will return to this point in Part III.A.

C. European Court of Human Rights

A number of IIA tribunals have supported their distinctions between lawful and unlawful expropriations, with the consequences for damages, by reference to European Court of Human Rights (ECHR) jurisprudence. The precedent sometimes cited is Papamichalopoulos v. Greece (1995), where the Court, having previously found that the Greek military government's taking of the applicant's beachfront property was a violation of Protocol I, labeled the dispossession “unlawful.” It then invoked Chorzów Factory and ordered Greece either to provide restitution or to pay the current value of the land and buildings (i.e., the FV_DA), as well as nonpecuniary damages for mental suffering.Footnote ⁷⁵

However, in Guiso-Gallisay v. Italy (2009), the Court, via the Grand Chamber, changed course significantly, ruling that when the state had definitively taken a claimant's property in the public interest such that she no longer had title to it, it need only pay the value as of the date of the expropriation (plus interest).Footnote ⁷⁶ The ruling suggested that some expropriations that violate Protocol I are worse than others, though it is hard to see how exactly it fits into the lawful/unlawful distinction.Footnote ⁷⁷ Investment tribunals do not, however, seem to have made reference to this case.Footnote ⁷⁸

D. A Brief Word on Academic Commentary

Contemporary academic treatments still make reference to the lawful/unlawful distinction as key to the determination of damages. Thus, the Dolzer/Schreuer volume and de Nanteuil's recent treatise restate the four criteria above as rules of customary international law, qualify any violation of those standards as an illegal expropriation, and call for recourse to the rules of state responsibility for such acts, which may lead to damages different from the value of the asset at the time of the taking.Footnote ⁷⁹ Both the latter and Marboe's impressive book also distinguish between compensation for a lawful expropriation with damages (indemnisation) for an unlawful act.Footnote ⁸⁰ Sornorajah agrees that takings must be for a public purpose and nondiscriminatory to be legal, but he does not seem to consider the mere nonpayment of compensation—even in violation of a treaty—to render the expropriation unlawful; and he calls for payment in those cases to be the amount in the treaty.Footnote ⁸¹

E. Crystallizing the Shortcomings in the Case Law

This overview reveals contrasting understandings as to what makes an expropriation legal or illegal as well as a multiplicity of approaches to the damages that follow. For some tribunals, the distinction is not mentioned. For scholars and tribunals that do note it, an expropriation could be unlawful due to a breach of a contract, local law, customary law, or an IIA (the usage most common among tribunals today). If it is unlawful, the remedy still might be based on the standard in the treaty, i.e., FV_DE, or on customary international law. If it is the customary law standard, it might mean FV_DA, FV_DE, or something else, which may or may not incorporate information not known at the time of the expropriation.

The practice and case law makes clear that every expropriation differs in terms of the method used and the impact on the investor; and every investment differs in terms of the best economic way of valuing it.Footnote ⁸² And not all treaties use identical language to describe the required compensation, although the use of FV_DE is overall quite consistent. The problem for the law is that this diversity of expropriations does not map onto the diversity of legal approaches to remedies. There is no fit between the often-irreconcilable principles in the case law and the actions of the states, including their impact on foreign investors. The combination of inconsistency and conceptual lack of clarity means that the case law does not provide legal guidance to participants in the foreign investment process.

The current approaches to remedies thus have a terminological and a substantive element. As a terminological matter, Chorzów Factory (and, to a lesser extent, the competing opinions in Amoco International Finance) still exerts a pull on some tribunals to characterize expropriations as lawful or unlawful, even if they do not use that distinction to determine damages. Tribunals seem to use the terms out of a sense of inertia, even though the only questions typically before a tribunal are: (a) whether an expropriation has taken place; (b) whether it met the criteria in the IIA; and (c) if not, the amount due for the treaty violation.

The terminological confusion has a fairly simple solution. Tribunals deciding expropriation claims under IIAs should describe expropriations as either treaty-compliant or treaty-violative. Whether the expropriation is unlawful in any other sense is not relevant for its consistency with the expropriation requirements of an IIA (except possibly to the extent that interpretation of some of the process conditions for expropriation may take account of customary international law). This terminology will avoid the incoherence of saying, for instance, that a violation of any of the first three (process-based) criteria of an IIA is an unlawful act while a violation of the fourth (payment) criterion alone is a lawful act.

Many tribunals use the term unlawful in this narrow sense of a treaty violation, so the term could be salvaged. But the alternative will also dispel misgivings by developing states that international law somehow questions their sovereign right to expropriate.Footnote ⁸³ By avoiding the moniker of “unlawful” in favor of treaty-violative, tribunals will be sending a clearer message that the state's right to expropriate is not at issue, but only whether the exercise of that right was conformity with treaties to which it is a party.Footnote ⁸⁴

But a terminological fix will not solve the more fundamental disagreement over the proper remedy for a violation of the criteria for a treaty-compliant expropriation. That dispute raises questions such as whether some violations of IIAs are worse than others; whether IIAs provide a lex specialis for remedies and thus the scope of the default rules of state responsibility; and how we ought best to monetize the damages for an expropriation.

A. Doctrinal Constraints?

International law does not offer one obvious doctrinal solution to the problem presented here. Certainly, as a matter of customary international law, a breach of a legal obligation triggers duties by the violator, which the ASRs have identified as including cessation and nonrepetition, along with full reparation. The ASRs define the latter as restitution if possible (and not an undue burden on the violating state) or, otherwise, compensation “for the damage caused” (meaning “financially assessable damage including loss of profits”).Footnote ⁸⁵

But the ASRs do not provide enough guidance for investor-state tribunals for two reasons. First, ASR Article 33 makes clear that the obligations of the responsible state, including that regarding reparation, apply only to other states, and are “without prejudice to any right” enjoyed by a nonstate actor, such as a foreign investor. The tribunal in Wintershall Aktiengesellaschaft v. Argentina recognized this limitation in interpreting the jurisdictional provisions of the U.S.-Argentina bilateral investment treaty (BIT).Footnote ⁸⁶ Zachary Douglas has interpreted this provision, as well as parts of Chorzów Factory, as suggesting that certain forms of reparation are unique to the interstate context and that ICSID tribunals, which address what he calls “the vindication of private interests,” may not have the authority to award the same damages to an investor as a state would receive.Footnote ⁸⁷ Brigitte Stern makes a related point that reparation can benefit only state interests and “perhaps private individuals,” a position she grounds in human rights law.Footnote ⁸⁸

Wintershall’s, Douglas's, and Stern's views on the inapplicability of Chorzów Factory and the ASRs regarding the consequences of breach—including the duty of full reparation—to the violation of investor rights have other arguments in their favor. For one, full reparation might not adequately reflect the current political economy of investor-state relations, where investors who sue might well be very sophisticated in terms of what they should have known about the possibility of expropriation; and states may strain to provide full compensation in the event of expropriations carried out in the national interest. I will return to this point in Part III below.

I need not decide whether all the ASR rules on consequences of a breach are inapplicable to investor-state arbitration. At a minimum, Article 33 leaves open the possibility that rules requiring full reparation might not apply to investors and affirms that if international investment law provides a lex specialis, then that law would prevail.Footnote ⁸⁹ Article 55 confirms that the ASRs do not apply if a lex specialis governs a state's responsibility. So, it is certainly possible for international investment law to have its own rules on consequences of a treaty breach, including recourse to the treaty standard for the quantum of damages.Footnote ⁹⁰

Second, even if a tribunal decides to use the ASRs to determine damages, the concept of compensation “for the damage caused,” or Chorzów Factory’s “wiping out all the consequences,” is open to significant interpretation and does not point to a single answer regarding the date of valuation or the inclusion of other losses.

The result is three significant gaps in the case law and doctrine. It does not tell us: (a) whether international investment law has a lex specialis regarding damages for expropriations; (b) if so, what that lex specialis might require in terms of the quantum of compensation; and (c) if not, what “full reparation” means in the context of different sorts of expropriations.

B. Toward a Coherent Framework

One response to these gaps might be to leave things as they are on the understanding that investor-state tribunals’ principal duty is to solve a dispute between two parties rather than develop any—or much—coherent case law. The origin of IIA-based arbitration in international commercial arbitration might support such a conclusion. Thomas Schultz has argued against consistency on the grounds that it can freeze bad law in place.Footnote ⁹¹ Critiques of consistent legal reasoning across cases raise many questions beyond the scope of this essay; but for my purposes two brief, modest responses will suffice.

First, the various actors in the foreign investment process will benefit from some predictability from arbitral awards. In considering whether to conclude an IIA, a host state will understand what it is accepting if tribunals treat similar words in similar IIAs similarly. In case of a dispute between an investor and a host state, if the parties see how tribunals have handled cases using similar treaties, they will have a better sense of the strength of their claims and the desirability of settlement. Lawyers in disputes will, or at least should, be able to avoid relitigating settled issues. Other interest groups, whether labor unions or environmental activists, will also be able to make more coherent and persuasive claims. Arbitration itself will have more credibility if the damage award is grounded more firmly in legal principles and not seen as just picking a number between those offered by the parties. This sort of minimalist consistency, as opposed to an arbitral free-for-all, is essential to any rules-based system.Footnote ⁹²

Second, we must acknowledge that international investment law has many attributes of a public law system. Although the public/private divide is both overrated and fuzzy, it remains the case that, given the stakes for host states—the key repeat players—in adjusting national policies to IIA standards and in devoting significant resources to pay damages, the need for some predictability on key interpretive issues is essential.Footnote ⁹³ Even if—indeed, especially if—one favors retention of decentralized, party-controlled arbitration, investment law has become an area where stakeholders deserve some coherence in the law. More consistency will not allay many other concerns of host states, e.g., the scope of IIA clauses on fair and equitable treatment, but it can contribute to a system with more acceptability from the key stakeholders.

If some improvement in the status quo is normatively desirable, one obvious doctrinal recourse would be to deploy principles of treaty interpretation in the hope that each IIA will reveal its own set of remedies for breach; or perhaps a common set of remedies will emerge for similarly worded clauses. The practice of states in response to treaty-violative expropriations when there is no arbitration, or the travaux préparatoires, might reveal a treaty-specific, regime-specific, or even a customary law remedy.

Although such an inquiry might yield some holy grail of remedies, it seems extremely doubtful. As for state practice under a treaty, states are unlikely to admit IIA violations if they do eventually pay compensation to investors, thus making it impossible to know whether the compensation paid “establishes the agreement of the parties”Footnote ⁹⁴ —i.e., of the host state and the investor's state as to the meaning of the treaty. And for the same reason, any compensation paid would be without opinio juris for purposes of customary law. Indeed, these same interpretive disputes arose over the relevance of ad hoc or lump sum settlements to the customary law standard of compensation.Footnote ⁹⁵ We can look to judicial opinions and the views of publicists, but those are very source of the confusion.

With the case law inconsistent, the doctrine incomplete, and traditional methods of interpretation unhelpful, development of a framework for remedies could proceed from one of two bases. One, purely deductive and indeed philosophical, would seek to ground remedies in the theories of property, torts, contracts, remedies, and related matters, including global justice—for indeed all are relevant to expropriation. No doubt the ground is fertile for this sort of research.

The other, and the one I adopt here, is to focus on actual community expectations about remedies.Footnote ⁹⁶ That inquiry does not entail a search for an extant, agreed formula for remedies, because both the case law and state practice demonstrate that we do not yet have one. Instead, my theory looks for community expectations at the next level up, identifying the goals that decision-makers agree should be served by the remedies to investors for treaty-violative expropriations. Only if we understand the policy behind, and the point of, those remedies, rather than merely recite the “wipe out all the consequences” mantra, can we begin to have a sense as to what kinds of takings merit what kinds of payment.Footnote ⁹⁷ The result will be a framework for future arbitral decision making, as well as treaty-drafting. I hesitate to use the word doctrine for it suggests a rigidity and formulaic approach to resolving a highly complex set of decisions by states, investors, and tribunals. Doctrine is what got us into this mess in the first place. So, while I aim for coherence and consistency, the model I propose can guarantee neither.

C. Purposes of Remedies for Treaty-Violative Expropriations

In my view, we can identify five goals of remedies as generally accepted by decision-makers in and observers of the process of legal regulation of foreign investment.Footnote ⁹⁸ Decision-makers might differ on how those goals should be operationalized, but I think the goals themselves are generally shared. Although they are not set out in any one document, case law and other discourse in the international investment community help us identify them. Beyond these five purposes, four goals are more controversial and contested.

D. Other Arguable Goals for Remedies

One might advance other goals for remedies for treaty-violative expropriation. Four are worth brief mention, along with the reasons they are not generally accepted.

A. A Valuation Primer

Before explaining the three options, it is important to clarify several economic concepts, in particular because tribunals are often not explicit about them. For simplicity's sake, we may think of four points in the life of the investment dispute, at least with respect to an investment that is a going concern. D0 is the date the investment starts producing income; DE is the date of the expropriation; DA is the date of the award; and DT is the termination date, when the investment stops producing income, which will we, for the sake of simplicity, assume is after the date of the award.Footnote ¹⁴⁸ Thus:

1. To reiterate a point made earlier, the full value of the investment at the date of the expropriation—FV_DE—could mean one of two things: either (a) the value of the investment at DE using information known at DE only, i.e., based on ex ante information—FV_DE (ex ante); or (b) the value of the investment at DE using information known at DA, i.e., based on ex post information—FV_DE (ex post). In the case of creeping expropriations, determining DE itself may be quite complex or it may be necessary to value the investment at a date other than DE.Footnote ¹⁴⁹

2. If FV_DE means FV_DE (ex ante), then it is calculated as the sum of: (a) the DCF from DE to DT, as expected at DE, discounted back to DE; and (b) interest from DE to DA.Footnote ¹⁵⁰

3. If FV_DE means FV_DE (ex post), then it is the DCF from DE to DT, i.e., the sum of: (a) the DCF from DE to DA, based on information known at DA, brought forward to DA; and (b) the DCF from DA to DT, as expected at DA, discounted back to DA.Footnote ¹⁵¹ Thus, this second version of FV_DE—the ex post version—includes performance from DE to DA based on information known at DA, whereas the first version includes the performance from DE to DA based on information known at DE.

4. These two versions of FV_DE differ only if there are events after the expropriation that were not predicted at DE that affect the DCF. If they had been predicted, they would have been priced into the FV_DE. But they can differ for many reasons, e.g., unexpected changes in commodity prices.Footnote ¹⁵²

5. The full value of the investment at the time of the award—FV_DA—is composed of two components: (a) the expected DCF from DA to DT, discounted back to DA, i.e., the discounted income stream after the award, as expected at DA; and (b) the accumulated income from DE to DA, i.e., the income from the expropriation to the award, brought forward to DA.Footnote ¹⁵³ If only (a) were awarded, the investor would lose the income between DE and DA.Footnote ¹⁵⁴

6. It is now clear that the FV _DE based on ex post information is economically the same as the FV_DA. The two consist of the same two components: the actual performance from DE to DA, brought forward to DA (i.e., both 3(a) and 5(b) above), + the expected performance from DA to DT, discounted back to DA (i.e., both 3(b) and 5(a) above).Footnote ¹⁵⁵ As a result, when tribunals use ex post information to calculate the FV_DE, they are actually calculating—or at partially calculating—the FV_DA. Unfortunately, tribunals are often not clear about this move.

7. The key economic distinction, then, between FV_DE (ex ante) and FV_DA is whether we judge value based on the expectations at time DE only; or whether we judge value based on the information we have at DA (both actual returns from the past and anticipated returns in the future).Footnote ¹⁵⁶ As will be seen below, this economic distinction is critical for evaluating valuation options. Indeed, the nub of the disagreement over how to treat treaty-compliant versus treaty-violative expropriations is over the treatment of information not available at DE. Footnote ¹⁵⁷

8. Finally, the formula in the typical IIA for payment to the investor must mean the FV _DE based on ex ante information alone, i.e., the sum noted in number 2 above. This is the amount the state would have to pay the investor under the treaty at time DE. Obviously, the state could base its full valuation under the treaty only on the information it (or the market) had at the time. A treaty could not require payment of compensation based on actual future returns, information that was not yet available to the state or the investor. So, the treaty formula must be the FV_DE (ex ante).Footnote ¹⁵⁸ However, tribunals that say they are applying the treaty formula and not some other value are often not clear about whether they are in fact using only ex ante information. When I use the term FV_DE below, I am referring to the version based on ex ante information only.

B. Option 1: Lex Specialis Within the Treaty—The IIA Formula for All Treaty-Violative Expropriations

An award of compensation based on the IIA standard—FV_DE (i.e., FV_DE (ex ante))—represents a sort of built-in damages formula, under which the treaty itself specifies the consequences for its own violation. Awarding this amount in all situations advances some but not all of the five goals of an expropriation remedy.

1. With respect to repairing the damage to the investor, in one sense this formula fairly considers the harm insofar as it could be said that the investor will receive the value of what was taken, plus interest. Neither the IIA violation itself, nor the character of the violation—i.e., whether the process conditions were violated in addition to the nonpayment of compensation—has an economic effect on the value of the asset.Footnote ¹⁵⁹ Economically, under the Capital Asset Pricing Model for valuing assets, the market price of the asset at a particular time requires the use of ex ante information alone because the market price of an asset reflects (predicted) risk and returns, not actual returns.Footnote ¹⁶⁰ We might thus say that the investor's harm was confined to the value as known at that time, because the taking also removed the risk of loss from him (not merely the risk of gain).Footnote ¹⁶¹

Yet we could identify the harm differently—that what was taken was the ability to hold and dispose of an asset as the investor chose.Footnote ¹⁶² Under that view, the investor lost the possibility that the asset might increase in value in a way that could not be foreseen at the time it was taken, and so the treaty formula is inadequate to repair the damage. Although the investor would lose that possibility even if the state had complied with the treaty—for all the investor can do in that scenario is reinvest the proceeds of the FV_DE paid by the state—it remains the case that the state breaching the treaty has deprived the investor of that choice. In terms of risk-shifting, FV_DE shifts the risk of future, unexpected loss in value to the state; but if the investment increases in value, the FV_DE ends up depriving the investor of those gains.

A famous U.S. domestic analogy concerns the litigation over the abuse of trust by the executors of Marko Rothko's estate, who sold many of the artist's works shortly after his death in a self-interested deal. When Rothko's heirs sued for their recovery after the paintings had skyrocketed in value, the New York State courts awarded them the value of the paintings at the time of the trial and not the much lower value at the time they were sold.Footnote ¹⁶³ The appeals court distinguished the normal rule of trusts, under which a trustee who breaches the trust by selling property at an inadequate price is not liable for any appreciation damages, by noting that here the trustee had a duty to retain the property, so in that sense he “violated an integral condition of the trust … [making the sales] inherently wrongful transfers which should allow the owner to be made whole.”Footnote ¹⁶⁴ The similarity to Chorzów Factory is remarkable in the emphasis in both cases on the violation of an independent duty. Yet we can also ask whether the equation of treaty commitment with a trust, the violation of which allows for such appreciation-based damages, is justified. The heir betrayed by someone with a duty to act on the former's behalf seems harmed in a worse way than the foreign investor protected by a treaty commitment to take property only under certain conditions.

FV_DE also seems inadequate if we view the protections of the IIA in terms of the jurisprudential concepts discussed earlier. In the terms put forth by Jules Coleman and Jody Kraus, we would say FV_DE is the amount to legitimate a taking done in accordance with the three process-criteria (analogous in some ways to the amount due the victim under a liability rule).Footnote ¹⁶⁵ But the amount to legitimate a taking may well be inadequate—or, at least, is not necessarily adequate—as a measure of damages when the state does not act accordance with some or all of the four criteria. We may not know the correct measure of damages, but we can at least say that it should not be the FV_DE.

Moreover, the IIA formula falls short in rectifying some harms to the investor associated with the noncompliance with the treaty. The harms from lack of due process as well as unexpected departure costs and other consequential damages are not captured in it.Footnote ¹⁶⁶

2. With respect to correct incentives regarding compliance, this formula fares rather poorly. If the state pays the FV_DE whether or not it complies with the treaty, with only interest to distinguish prompt payment from tribunal-ordered late payment, then, all other things being equal, the state has less incentive to comply with the IIA when it expropriates.Footnote ¹⁶⁷ States may have other incentives to comply with the treaty, such as the reputational harm from being a scofflaw or the need to attract foreign investment, but, ceteris paribus, a remedy that takes no account of the law violation enhances the likelihood of its repetition.

At the same time, payment of FV_DE, plus interest, should not generally incentivize a state to engage in so-called “opportunistic expropriation,”Footnote ¹⁶⁸ takings where the state believes it will profit more with the investment under its control than it would with the investment under the investor's control. The reason is that if the state and the investor both know that the investment is likely to generate more profits than originally expected, that expectation will be built into the FV_DE even if it is paid later with interest.Footnote ¹⁶⁹ It is, however, possible that the state will know more than the investor, e.g., if it plans some regulatory changes after the takeover that will make the investment more profitable.

3. Regarding the expressive purpose, a remedy for treaty-violative expropriations based on the treaty formula signals that violations will not go unaddressed; for the state will still have to pay for its violation. However, FV_DE does not send a special signal in terms of a sanction for the treaty violation, as opposed to the lack of payment. In that sense, it does not provide the same sort of control intention as would a rule based on a damages formula external to the treaty itself. Put another way, there is a certain sense of unclean hands when the state that goes outside the treaty by breaching it is required to pay only the amount specified in the treaty.

4. The internal treaty standard seems to advance the goal of a stable framework of international investment law accepted by a broad range of parties. When states agree to a treaty that specifies a formula for compensation for expropriation, it would seem to preserve confidence in the system if tribunals awarded compensation based on that formula.Footnote ¹⁷⁰ On the other hand, sophisticated host states ought to know that when they become parties to an IIA, they are agreeing to the possibility that a tribunal could award expropriation damages above FV_DE for treaty-violative expropriations, and that if they wish to preserve FV_DE as the sole formula for compensation, they ought to so specify in the treaty. And perhaps future IIAs will so specify. But for now, when host states are increasingly suspicious of investor-state arbitration, a strong case can be made that the treaty formula strikes a balance that will advance the durability of the system. This idea is expressed in arbitrator Stern's dissent in Quiborax, where she states that relying on FV_DE (which she emphasizes should be determined only using ex ante information) represents a “fair interpretation of international investment law.”Footnote ¹⁷¹

5. Lastly, as for administrability, the treaty standard requires valuation of an asset at the time it was taken. Financial and valuation experts for the parties are working with, in essence, time-stamped (at DE) data on the market value of the asset, which may include a projection of revenue in the case of a going concern.Footnote ¹⁷² As a general matter, this method, based on determining the expectations at DE, seems more difficult than one where the financial experts can consider economic data after DE, which may well account for some tribunals’ willingness to use post-DE data.Footnote ¹⁷³

In sum, the internal lex specialis approach, under which treaty-violative expropriations leads to an award of damages based on the treaty formula of FV_DE (ex ante), plus interest to DA, advances one, or perhaps two, of the goals of a remedy—repairing the damage to the investor, if we to see that damages as limited to the value of the asset taken; and advancing the stability of the investment regime. But it does less well in terms of repairing the full harm to the investor, putting the right incentives on states to comply with treaties, and sending an expressive signal regarding the consequences of law violation, and it may be administratively difficult.

C. Option 2: State Responsibility's Default Rules—Full Reparation for all Treaty-Violative Expropriations

Reliance on the ASRs’ concept of full reparation for damages—”wiping out all the consequences” in Chorzów Factory’s famous phrasing—in the case of any treaty-violative takings meets some of the above-stated goals, but, as with the prior solution, not all. The difference between this option and Option 1 is the possibility for a tribunal to value the investment based on FV_DA as well as to include other damages not reflected in the value of the investment.

1. Full reparation repairs the harm to the target if we see that harm as not merely the loss of an asset in the past. For when the state violates the treaty, it harms the investor by depriving him or her of the choice that the treaty provides to: (a) hold the investment and perhaps sell it later (i.e., the investor's choice if the state does not expropriate); or (b) invest the FV_DE proceeds (i.e., the investor's choice if the state does expropriate). So, if the investor had held the investment and it appreciated, the state's treaty violation has deprived the investor of the opportunity to enjoy that appreciation—although the investor should not enjoy any appreciation due to the state's takeover of the investment (perhaps due to lower labor costs or new regulations). Damages based on FV_DA allow for the possibility of this appreciation. While the investor's foresight in keeping or selling an investment, or investing the proceeds, should not be assumed to be perfect, this remedy reflects that loss.

Valuation on the date of the award, using information known since the expropriation, thus reflects the reality that the loss is ongoing until the date of the award. And if damages are supposed to be a substitute for restitution, then the award date makes the most sense because that is when the property would be returned to the investor, as several tribunals have recognized.Footnote ¹⁷⁴ Though some may object that damages would then “vary with the date of the award,”Footnote ¹⁷⁵ that possibility is not normatively objectionable but inherent in the concept of fully repairing the wrong. (Indeed, total damages will change with the date of the award even when they are based on FV_DE with interest, because of the accumulation of interest).

Yet what if the asset depreciates after the expropriation, due to factors not associated with the expropriation itself? Does full reparation then require an award of lower damages than the treaty standard of FV_DE? It could be argued that the investor should receive less than FV_DE, insofar as when the investment depreciates post-taking (due to forces beyond either the control of the state or the investor), the state's violation of the treaty deprived the investor of only the depreciated value.Footnote ¹⁷⁶ But this argument does not fairly describe the harm to the investor. Rather, the investor had a right, under the terms of the treaty, to the FV_DE. The investor's rights under the treaty make any subsequent depreciation not really “countable” for purposes of the harm, and thus the customary law rule of full reparation would seem to require setting the treaty standard of FV_DE as a floor. Footnote ¹⁷⁷ This choice also means that the risk of loss is borne by the state if it violates the treaty. Another way to state this point is that the investor on the receiving end of a treaty-violative expropriation should not receive less than an investor who has experienced a treaty-compliant expropriation.Footnote ¹⁷⁸

The result is the formula for full reparation introduced by Judge Brower in Amoco International Finance, which sets the FV_DE (again, based on the ex ante information) as a floor on the remedy.Footnote ¹⁷⁹ By giving the investor the choice of FV_DE and FV_DA, the Brower formula shifts the risk of depreciation of the asset to the state and the benefits of appreciation to the investor. This risk-shifting better addresses the harm to the investor from the state's deprivation of the choice described above and thus provides full reparation.Footnote ¹⁸⁰

Full reparation also has the possibility of addressing the damage to the investor from the state's violation of the BIT's process conditions. That is, even if the asset has not appreciated at all and FV_DA is nothing more than FV_DE plus interest, a full reparation model can conceivably address the state's failure to respect these criteria. To offer a colloquial example, if a customer takes an item from a store and simply drops the cash equivalent of the price on the floor, we might say that the shopkeeper has been harmed for a failure to follow the procedures for buying merchandise.Footnote ¹⁸¹

In addition, the full reparation model allows for damages to include not merely the FV_DA, but consequential damages, i.e., incidental losses to the investor from the taking. The treaty formula does not do so.

Yet full reparation has drawbacks in terms of its calibration to the harm to the investor. First, consider the implications of this approach for violations of the fourth, payment condition typical of IIAs. The host state would have to pay extra damages above the FV_DE even if it has paid or offered a payment to the investor, as long as that amount is less than the FV_DE as determined by the tribunal. One could argue that a state should not be able to evade damages by offering an insufficient amount to the investor. But insufficient payment seems materially different from a failure by the state to pay anything at all, and the full reparation model does not capture that possibility. In Coleman and Kraus's terms, if payment according to the three process conditions legitimates the taking, then insufficient payment is at least closer to legitimating the taking than a deliberate refusal to pay, and the two scenarios should not be treated the same in terms of damages. The same could be said if the state has a bona fide dispute with the investor over who is the proper owner of the investment. Some tribunals and scholars have accepted this distinction by asserting that nonpayment of compensation is not always a violation of the IIA in the first place.Footnote ¹⁸² Second, a state can violate the other three criteria in different ways, e.g., with insufficient process as compared to no process at all. Treating all violations the same in terms of the remedy ignores these nuances.

2. The full reparation model also seems better than the treaty formula at putting the right incentives on states to comply with the treaty. If we care not merely that states pay for assets that they expropriate, but keep the promises they have made—including the process conditions—then a full reparation formula offers a better incentive to comply. It requires states to internalize all the costs associated with the violation, including appreciation damages and consequential damages. States cannot merely pay late with interest.

3. Full reparation generally fills the gap in the expressive value left by reliance on the treaty formula. It sends a stronger message of the law's relevance by treating violations as not mere lapses to be corrected by late payment but as wrongs by the state to be assessed by reference to external standards of damages. But it treats violations in a binary way, ignoring possible bona fides by the state. It thus risks crossing the line from acceptable expression to impermissible punishment.

4. The full reparation formula, however, raises significant concerns regarding the goal of furthering the stability of the international investment legal regime. As shown from the examples above, a state can violate the treaty in numerous ways, with different effects on the investor. A reparation formula that does not capture these nuances among treaty-violative expropriations risks upsetting the current expectations among many states regarding their right to expropriate. If tribunals routinely award FV_DA (or FV_DE based on ex post considerations) in a way that increases already large awards against developing states, it could exacerbate the backlash against IIAs.

Certainly, states are bound by agreements they have signed, including those that entrust the determination of violations and remedies to arbitrators. And host state claims that restrictions on their ability to expropriate are a violation of their sovereignty are clearly wrong as a matter of law (at least since Wimbledon). But expropriation remains a power of all states, and tribunals should be careful about offering remedies clearly associated with other unlawful acts when the state may have acted in good faith, albeit inconsistent with the IIA. As noted earlier, there is a good case that tribunals should not apply the Chorzów Factory/ILC reparation rules to investor-state disputes, and the ASRs themselves allow for special regimes of responsibility. The preservation of the system of IIAs through the confidence of all stakeholders is not advanced by requiring the state to pay to the foreign investor (especially one insured), in all circumstances, what it would have to pay another state. Thus, the risk to broad-based acceptability of the system is significant.

5. A full reparation regime also has drawbacks from the perspective of administrability. It requires a tribunal to determine a number of values that it need not under the treaty formula. First, in determining the FV_DA, the tribunal must not use the actual performance of the investment since the expropriation, but rather the performance assuming the investor still owned the investment, thus excluding any increased value due to the government's takeover.Footnote ¹⁸³ For the state should not have to compensate the investor for profits that the investor would not have made, just as the investor should not be prejudiced by a decline in value due to the expropriation. At least one tribunal, in Rumeli Telekom v. Kazakhstan, refused to award the investor for increased value caused by the state, but the practice does not seem pervasive.Footnote ¹⁸⁴ But it may prove difficult for a tribunal to determine whether the government did a better or worse job than the investor would have done in managing the investment.Footnote ¹⁸⁵

Second, if full reparation means the greater of FV_DE and FV_DA, then the tribunal must determine both the FV_DE and the FV_DA, which requires more accounting data (and arguments between the parties) than does the IIA formula. The Yukos tribunal managed—in eighty pages of reasoning—to perform this feat when valuation turned mostly on share prices, but others using DCF or other valuation methods might find this quite difficult.Footnote ¹⁸⁶ Third, the tribunal must determine consequential damages, i.e., costs to the investor from having to abandon its investment.

Thus, full reparation does reasonably well at advancing the first two goals of a remedy—reparation of harm (though it might overcompensate the investor if we see a state's underpayment based on a bona fide disagreement with the investor as a lesser harm) and incentives on the host state; but its binary approach to violations risks turning expression into punishment, it is deficient in terms of respecting the political economy of the international investment law regime, and it can be difficult to administer.

These concerns about Option 2 have significant implications for the continued relevance to international investment law of Chorzów Factory, which remains routinely cited by tribunals as the font of wisdom on remedies. Beyond the concerns noted earlier about the transposability of customary international law on remedies developed in the interstate context to investor-state disputes, and the economic anachronism of Chorzów Factory’s valuation techniques, the above analysis underscores two key shortcomings of applying Chorzów Factory to IIA violations. First, its distinctions between lawful and unlawful expropriations, and its sweeping rule for remedies for the latter, are oblivious to the “infinite varieties” of the way a state may take property. And second, its demand to “wipe out all the consequences” of a treaty violation detracts from several critical purposes of a remedy.

Thus, whatever the merits of the full reparation approach to other regimes of international law, or as a default rule of custom, tribunals should stop assuming its relevance as a starting point for remedies in investment law. There is a certain irony here, for Chorzów Factory originated in an investment dispute; though it has come to stand for a much broader proposition in customary international law, it has also become quite outdated when it comes to the remedies needed in international investment law.

D. Option 3: The Differentiated Approach: Tailoring Remedies to the Contours of the IIA Violations

Under the third approach, the nature of the treaty violation determines the remedy. Rather than defaulting to either the treaty standard or the full reparation standard, the tribunal makes choices about remedies based on what the state has actually done in the course of its breach of the IIA. At this point I will assess this approach in terms of the five goals of remedies. I will not, for now, appraise any single formula that tailors remedies to treaty violations.

1. A differentiated approach repairs the harm to the foreign investor insofar as it links the conduct of the state in concreto, rather than the mere existence of a treaty violation, to the remedy. Investors can be harmed in different ways by treaty-violative expropriations (or even treaty-compliant ones). They may have faced an unfair process that did not allow them to contest the action; been the target of discrimination on racial, religious, or other arbitrary grounds; received partial payment, a promise of payment, or no payment. In addition, they may have incurred large expenses in contesting the action or terminating the investment (the latter of which they might incur in a treaty-compliant expropriation). Each of these violations has a distinct impact on the investor.

2. This approach can also create the incentives on states to comply with their treaty obligations. As noted above, Option 2 does this better than Option 1. At the same time, why should we assume that full reparation (whether FV_DA, or the choice between FV_DA and FV_DE) provides the optimal set of incentives to induce compliance with the treaty? If a state refuses to pay FV_DE but respects the process conditions, then we may be able to create an incentive for them to pay FV_DE (i.e., comply with the treaty) by requiring them to pay the difference with interest, rather than insisting on damages equal to the FV_DA. If a state violates some but not all of the process conditions, a remedy tailored to that violation, as opposed to a default rule of FV_DA, might incentivize the state to follow the conditions they violate. A flexible approach may do better at providing the “sweet spot” in which we both fairly compensate the investor and provide the necessary incentives on states.

3. A differentiated approach can advance the expressive goal of a remedy. The importance of a treaty commitment can be reinforced without insisting that all violations are of the same gravity, i.e., the signal sent by uniform recourse to either the treaty formula or a full reparation rule. For example, a violation of the nondiscrimination criterion might well merit a particularly significant remedy; failure to pay FV_DE due to unsuccessful negotiations between the state and investor would merit a lesser remedy.Footnote ¹⁸⁷

4. At a time when large awards against host states face resistance, any formula for damages beyond the compensation specified in the IIA raises significant concerns. The differentiated approach, insofar as it leaves open the possibility for a tribunal to set damages based on standards external to the treaty, also faces this shortcoming. Yet the possibility that tribunals might use the treaty formula for some violations, while reserving other, external formulas for other violations, mitigates these concerns somewhat. A differentiated approach might also afford more discretion to arbitrators than the earlier two options, which might produce opposition from host states and others concerned about the impartiality of arbitrators.Footnote ¹⁸⁸ On the other hand, arbitrators already have significant discretion in damage determinations. Thus, a differentiated approach contributes to the viability of the regime, but only if arbitrators are able to justify and tailor remedies so as to maintain the confidence of states.

5. Compared to the treaty formula, a differentiated remedy is more complex for arbitrators to administer in terms of the calculation of damages because it requires looking at more than just the ex ante determined FV_DE. Compared to full reparation, that standard could also be more complicated, insofar as a tribunal, instead of calculating the FV_DA (or the FV_DA and the FV_DE, under the Brower formula) and consequential damages, may have to monetize other impacts of the expropriation on the investor. How, for instance, does one put a monetary value on the failure of the state to follow due process?

In sum, the differentiated approach offers significant promise over the other two alternatives. It offers the possibility of tailoring the remedy for treaty-violative expropriations to the harm done to the investor while maintaining the right incentives on states to comply with the treaty. It fulfills the expressive function better than the ASR remedy. It offers flexibility for maintaining the confidence of various participants in the international investment law process. Yet it may prove difficult to administer insofar as some of the violations may be hard to quantify.

A. Rejecting Economically Flawed Formulas for Remedies

While the point may seem obvious, tribunals need to move beyond economically antiquated notions of damages for expropriations. In the era of modern finance, terms like damnum emergens and lucrum cessans have no place in international arbitration, even if civil law systems continue to use them.Footnote ¹⁹⁶ The mere usage of those terms in frequently cited cases such as Chorzów Factory or Amoco International Finance is not a reason for their retention when they lack an economic basis. They also risk the possibility of double counting of damages.Footnote ¹⁹⁷ It is unfortunate that they appear in some serious scholarship on expropriation.Footnote ¹⁹⁸ The World Bank Guidelines offer a useful, economically rational set of valuation methods for different sorts of expropriated investments.Footnote ¹⁹⁹

B. FV_DE as the Remedy for Bona Fide Disagreements over Compensation

When a state violates an IIA by not providing the compensation required by the treaty, i.e., FV_DE, but the violation was due to a bona fide disagreement between the host state and the investor over the FV_DE, and the state respected the process conditions for an expropriation, then the remedy for such a violation should be only the FV_DE (i.e., based on ex ante information), plus interest to the date of the award. By bona fide disagreement, I mean situations where the state is committed to paying the treaty amount but has legitimate disagreements with the investor that have prevented payment. The most obvious cases would be where the host state makes a good faith calculation of FV_DE that the investor rejects, or a there is bona fide disagreement about the identity of the legal owner of the investment, but the principle would apply to similar cases where the state still respects the process conditions. Thus, the host state's domestic law could require the government (or the investor) to follow certain administrative, judicial, or legislative procedures before payment is approved—which can serve as an important protection for the state's treasury. If those procedures have not concluded and the investor seeks recourse to investor-state arbitration, we can view the disagreement as bona fide assuming the state is still committed to paying and the delay is justifiable under the circumstances.Footnote ²⁰⁰

With respect to the five goals I have identified:

1. This solution repairs the most significant damage to the investor, insofar as the only harm was not receiving the FV_DE when it was due. The state did not harm the investor by ignoring the process conditions or rejecting the possibility of a remedy; it did not seek to deprive the investor of the ability to dispose of the asset or the proceeds as the latter chose in the future. The harm to the investor from mere nonreceipt of the amount due under the treaty, absent more aggravated circumstances, is repaired by the payment of FV_DE plus interest. Payment of FV_DE completes the treaty-based process of expropriation and legitimates it.

As a conceptual matter, the idea of paying only the treaty amount for good faith failures to pay resembles the remedy that courts in the United States award when insurance companies fail to pay on their policies. If the failure is based on an incorrect but good faith interpretation of the contract, only the amount in the contract is awarded. If the company fails to pay in bad faith, then a greater amount is awarded (even as courts and commentators disagree on the sort of conduct that would justify additional damages and how much they should be).Footnote ²⁰¹

2. It puts the right incentive on states. States are incentivized to work with the investor in good faith in order to avoid higher damages. These negotiations, if successful, will prevent violations and ensure compliance, as well as prevent litigation. While it could incentivize states to offer insufficient amounts and then negotiate (for the most the tribunal would later order is FV_DE), limiting the FV_DE option to good faith disagreements should constrain manipulation of the process.

3. It sends a signal that the law still matters—losing states still pay damages. At the same time, by treating a bona fide disagreement over payment and a state's clear flouting of the terms of the treaty differently, it expresses the idea that the law treats the former less severely than the latter and avoids a punitive outcome. The remedy reinforces the law's relevance even as it expresses a tolerance for legitimate disagreements between states and investors on the FV_DE.

4. It reflects the political economy of the foreign investment process and thereby promotes states’ willingness to participate in the legal regulation of that process. It recognizes the failure to pay FV_DE as a treaty violation, which respects the legitimate claims of home states and their investors; yet states’ failure to achieve a meeting of the minds with the investor on the amount of the FV_DE or other issues will not be condemned through a harsh remedy.

5. Concerning administrability, on the one hand, the proposed rule requires the tribunal to determine only the FV_DE. On the other hand, that process requires use of dated information, which might be difficult. More important, this proposal requires the tribunal to determine whether the state's failure to pay FV_DE was due to a bona fide disagreement about the price, title-holder, or other factors. This task requires detailed analysis of the evidence regarding the parties’ negotiations and ultimately a verdict on blame, if any, for the failure of payment. This process could extend the litigation. Yet Venezuela Holdings, Tidewater, and ConocoPhillips show that tribunals are capable of making such an inquiry. Those tribunals looked closely at the parties’ interactions to conclude that in the first two, the state had acted in good faith, while in the last one, it had not.Footnote ²⁰²

By offering a different remedy for one form of nonpayment than for others, even though both are treaty violations, this proposal is in tension with the ASRs, which do not link state responsibility and the duty of reparation to intent on the part of the state—leaving such questions for the primary rules—and provide only a limited set of circumstances precluding wrongfulness.Footnote ²⁰³ Yet the ASRs recognize that reparation is, after all, for a harm and even (albeit incompletely) that a remedy should be proportional to the harm.Footnote ²⁰⁴ So clearly less harm means less compensation. The idea of gradations of damages based on the severity of the violations has been accepted in other contexts. Notably, the Eritrea-Ethiopia Claims Commission argued that “the law of State responsibility must maintain a measure of proportion between the character of a delict and the compensation due.”Footnote ²⁰⁵

Finally, it is worth considering whether a state's failure to compensate the investor because the former does not view its actions as an expropriation at all should be treated as a bona fide disagreement. In situations where the state denies that it is expropriating, it will probably not provide due process or compensation. This case seems somewhat closer than the other bona fide disagreements noted above.

As for the harm to the investor, we could limit compensation to FV_DE and address any additional harm from the violation of due process through proposal D below. As for incentives and the expressive value of a remedy, the possibility of paying only FV_DE might incentivize states to use the indirect rather than the direct route to expropriations. But some governments reasonably believe that their regulations later determined by a tribunal to be expropriatory are not so, and FV_DE thereby avoids a punitive element. From the perspective of the viability of the investment regime, limiting compensation to FV_DE will reassure host states that their regulatory actions, even if ultimately found to be expropriatory, are not treated as outright refusals to pay.Footnote ²⁰⁶

The countervailing concern here is administrability in terms of the tribunal's ability to determine the existence of a bona fide disagreement over the expropriatory nature of the government's acts. Tribunals would have to discern the seriousness of the host state's arguments that it was not expropriating indirectly, even though they have rejected those arguments on the merits. The tribunal would have to distinguish between obvious indirect expropriations and less obvious ones. Yet it is certainly possible for tribunals to reach such a conclusion, suggesting that overall, limiting payment to FV_DE for such disagreements is justifiable.

Treating this situation as a bona fide disagreement may, in the end, prove unnecessary to address host state fears that tribunals will issue large awards for indirect expropriations. For tribunals have been increasingly wary of claims of such expropriations, adopting a test for the existence of an expropriation that centers on the loss of control and not merely a diminution of revenue.Footnote ²⁰⁷ And more recent treaties contain a significantly narrower definition of expropriation.Footnote ²⁰⁸ Thus, the primary rules seem to be the best site for addressing host state concerns about too many findings of indirect expropriations.Footnote ²⁰⁹

C. For Outright Host State Refusal to Pay FV_DE, A Choice for the Investor Between FV_DE and FV_DA

If a state fails to pay the investor due to a refusal to apply the treaty standard of compensation, then tribunals should give the investor damages equal to the FV_DA, which must exclude any appreciation that can be traced to the state's ownership of the investment, with a floor of FV_DE (plus interest).Footnote ²¹⁰ Recall, however, that, in general, FV_DA will only differ from FV_DE if the investment performed in a way that could not have been predicted at DE. Although rarely used in practice, nor conditioned on excluding the state's contribution to the increased value of the investment, the choice between FV_DE and FV_DA is superior to the others in terms of the five goals of a remedy.

1. It provides the reparation that best matches the harm to the investor. The state has deprived the investor of the choice to hold the investment, sell it, or invest the proceeds, and the remedy needs to correct this harm. To do so, it needs to consider the value as of the date of the award based on information at that date. FV_DE is justifiable as the floor because the investor was guaranteed that amount under the treaty, placing the risk of post-expropriation loss on the state breaching the treaty; but if the value has increased due to no action by the state, it must pay the (higher) FV_DA.

2. This proposal also places the right incentives on the host state. By shifting the risk of depreciation of the investment from the investor to the state, and the risk of appreciation from the state to the investor, it puts a strong incentive on states to carry out their treaty commitments. At the same time, the state does not pay the investor for the state's own contribution to the increased value, mitigating concerns that full value is punitive.

3. The expressive value of the remedy is also enhanced. By requiring the state to pay the investor potentially more than the FV_DE, it imposes a special consequence for ignoring the treaty, as opposed to simply misinterpreting the compensation requirement. It thereby reasserts the importance of the duty of compensation. Nonetheless, some host states may still view the difference between FV_DE and FV_DA as a form of punishment, especially if the additional amount is significant (though, as noted, the two numbers will often be the same). It is hard to combat such perceptions, but it is also important to note that damages above FV_DE need not be viewed as punitive.Footnote ²¹¹ In this case, the state has refused to pay and violated a core obligation of the IIA, so the better view seems to be that any difference is aimed to address the full harm to the investor rather than punish the state.

4. With respect to the viability of the legal regime on foreign investment, host state governments that believe investors win too often, or receive too much, will resist any remedies that go beyond FV_DE (and even FV_DE itself). In that sense, it is inferior to applying FV_DE in all situations.Footnote ²¹² But the limitation of this remedy to outright refusals to pay is superior to the alternative of awarding FV_DA for any treaty-violative expropriation. Moreover, if tribunals justify their finding that a nonpayment is due to a refusal by the state and not a bona fide disagreement covered by the prior proposal, then the imposition of the risk-shifting remedy could sit better with host states.

5. As for administrability, this remedy is more complex than the other two insofar as it may require a calculation of FV_DA as well as FV_DE. But tribunals will only have to engage in that calculation in a more limited set of cases than under an approach that awards the greater amount in the event of all treaty-violative expropriations.

D. A Case-by-Case Approach to Violations of a Treaty's Process Conditions

A violation of an IIA's process conditions accompanied by payment of FV_DE is unlikely to lead to litigation, so the likely scenario before a tribunal involves the state's breach of those criteria—typically the public purpose and due process requirements (especially the latter)—in addition to nonpayment. In principle, such a violation seems worse than the refusal to pay in isolation. Yet if the refusal to pay justifies a remedy of FV_DA (with FV_DE as a floor) for the reasons stated in the previous section, what additional damages should be awarded for failure to meet the process conditions? The five goals suggest that tribunals need flexibility on this issue.

1. Any additional harm to the investor from a violation of the process conditions would likely involve some sort of process-related mistreatment.Footnote ²¹³ Yet the sort of harm to the investor can greatly vary, and so the remedy should not replicate the approach for an outright failure to pay. Thus, the investor will need to demonstrate the specific harms from these violations to receive additional damages. One solution might be to award some expenses incurred in opposing the abuse of process. A more unconventional idea (that would require amending the IIAs) would be to award additional damages to the investor's state of nationality. If, for instance, the investor faced nationality-based discrimination, such discrimination harmed the investor's state as well.

2. With respect to incentives, an additional amount of damages gives the state an incentive to follow not merely the compensation requirement but the process conditions. States have put these criteria into IIAs, and they should not be treated as surplusage. Indeed, they are listed first in the typical treaty and offer an important measure of protection to the investor and, indirectly, to the investor's home state. Moreover, compliance with them can avoid tensions between the host and home state after an expropriation. Thus, the law should incentivize compliance with them.

At the same time, if, for some reason, the state pays FV_DE but violates the process conditions, and the investor still decides to sue, these incentives argue against treating a violation of the process conditions as the same as an outright refusal to pay. Two reasons argue for a more flexible approach. First, although the host state must meet the process conditions to comply with the treaty, in reality the payment condition is the most important from the perspective of both the investor and its home state. Second, given the first goal of tailoring a remedy to the harm to the investor, and the possibility for a range of harms as noted earlier, incentivizing compliance by requiring the same remedy as an outright refusal to pay would seem punitive.

3. With regard to the expressive purpose of a remedy, the possibility of additional damages sends the signal that IIA commitments are not merely about paying investors if they are expropriated, but about following certain standards in doing so. If the state violates those standards in addition to refusing the pay, the law is reinforced by an additional remedy.

4. Such a solution, like the prior one, raises concerns in terms of the reaction by losing parties. To enhance the proposal's acceptability to host states, tribunals must pay close attention to the legitimate concerns of host states in expropriating property. Thus, for example, the public purpose criterion should be broadly interpreted; discrimination should be limited to obviously invidious grounds; and the due process requirement should not involve importation of the procedural protections we might expect in a human rights regime. If arbitrators reserve the finding of process violations to unambiguous situations, then the possibility of additional damages may be acceptable to host states.Footnote ²¹⁴ In the end, tribunals may choose to refrain from adding anything to the damages—especially if FV_DA turns out to be much higher than FV_DE.

5. In terms of administrability, this proposal places a great deal of discretion within tribunals as to whether to compensate for these violations, and how much. Determining the separate harm to an investor from the violation of the process conditions may be impossible in many situations. This challenge again underlines the importance of clear proof by the claimant of harm from the violation of those criteria.

E. For Situations Other Than Bona Fide Disagreements over Payment, Payment of Demonstrable Consequential Damages

Lastly, where the state has violated the treaty's process or payment conditions without any bona fide disagreement about compensation, then the state should have to pay consequential damages associated with the violation. These damages, recognized by tribunals and scholars alike, address the costs incurred by the investor as a result of terminating the investment under circumstances not provided under the treaty.Footnote ²¹⁵ They include costs associated with unanticipated or abrupt departure of personnel, unexpected transitions to government control or compliance with new rules, and perhaps even loss of reputation.Footnote ²¹⁶ (They do not generally include the legal costs of the arbitration or domestic remedies.) With respect to the five goals:

1. Consequential damages compensate the investor for unanticipated costs (even if, in practice, an investor can plan for or even insure against a breach of the treaty by the other side). At the same time, the investor should incur all the costs associated with an expropriation that is treaty-compliant, even if it objected to the expropriation. In those situations, the harm to the investor from the expropriation is, according to the treaty, fully covered by FV_DE. As for bona fide disagreements over the compensation, consequential damages should not be awarded because the situation seems close to compliance by the host state. Because the state has followed the process conditions and is committed in principle to paying the FV_DE, the investor is incurring only the expected costs of a treaty-compliant expropriation. If these disagreements include bona fide disputes over the existence of an expropriation, the state will not have followed all the process conditions, but on balance the good faith nature of the state's claim suggests nonpayment of consequential damages as well.

2. Compensation for consequential damages puts the right incentive on states insofar as those damages would not be due if the state had followed the IIA. Rather, the investor would have been expected to pay them. The exception for bona fide disagreements incentivizes good faith negotiations.

3. Consequential damages serve the expressive purpose of reaffirming the importance of expropriating only according to the IIA's procedures. A willingness to pay FV_DE should be treated differently from outright rejection of the treaty's requirements.

4. Consequential damages might be resisted by host states, but the amounts are not likely to be so large that they threaten their commitment to the legal regime on foreign investment. In Siemens, the amount was $9 million compared to $208 million for the FV_DA; in Desert Line, it was one million dollars compared to the roughly $20 million on the main claim.Footnote ²¹⁷

5. Lastly, consequential damages are relatively easy for tribunals to assess in that they typically do not involve complex determinations of market value but rather discrete expenses by the investor in connection with the treaty-violative expropriation. If the expense cannot be linked to the state's taking of the property, then tribunals will not award it.

This five-part proposal thus permits tribunals addressing expropriations to tailor remedies to the nature and severity of any IIA violations, while avoiding the conceptual confusions of the lawful/unlawful distinction and that between damnum emergens and lucrum cessans. It does a better job at advancing the purposes of a remedy than either the treaty standard for all expropriations or Chorzów Factory’s full reparation standard. Although not unambiguously superior on all five goals, the proposal is nevertheless on balance clearly superior. We could even reduce it to three prongs: (1) FV_DE for treaty-violative expropriations characterized by a bona fide disagreement between the parties about the compensation (including over the existence of an expropriation); (2) the higher of FV_DA and FV_DE, along with consequential damages, for refusals by the state to pay compensation; and (3) a case-by-case approach to any damages for violations of an IIA's process conditions.

F. Next Steps for Tribunals and Treaty-Drafters

Tribunals adopting this approach, along with the secretariats that help draft their opinions, will need to make adjustments, both in shedding anachronistic terminology and in determining consequences for the violations of IIAs. Tribunals should be capable of distinguishing refusals to pay from nonpayments where the state and investor could not agree on the FV_DE. For the former situations, some tribunals—Siemens, ADC, Yukos, and Quiborax—have already awarded FV_DA, demonstrating that valuation of an investment as of that date is possible. With regard to violations of the process conditions, tribunals will be in new territory, as they have not focused on isolating damages due to violation of those criteria alone. Consequential damages, on the other hand, are remedies with which tribunals already have familiarity.

Will these new determinations by tribunals undercut the overarching goal of a more predictable and coherent case law? Compared to the current approach, the proposal seems like a significant improvement. We will have a rubric to connect remedies to the terms of the IIA and a more transparent framework that clarifies what losses to the investor should be compensated. Counsel and tribunals can focus on relevant questions about the actions of the host state and their impact on the investor: Did the state have a bona fide dispute with the investor over payment or the existence of an expropriation, or did it simply refuse to pay the treaty amount? Did any violations of the process conditions have an effect on the investor? Did the investor prove any consequential damages? These questions are more pertinent to remedies and more judicially manageable than figuring out whether an action was lawful or unlawful as well as what the PCIJ meant in Chorzów Factory. On the damage calculations, valuation experts will still offer tribunals divergent figures, but claimants and tribunals can at least work from a generally shared appreciation of the economic concepts of FV_DE and FV_DA. That understanding includes the need to exclude post-expropriation information from the former and include it in the latter, the limited circumstances in which the two numbers will differ, and the necessity of excluding from FV_DA any unexpected increase in value due to the government's takeover.

Finally, the approach can be implemented with even greater predictability and authority if states incorporate it in the treaties themselves. Thus, we could imagine a new provision in IIAs, after the standard expropriation article, that tells a tribunal what to award when the state expropriates in a manner inconsistent with each of the four standard requirements. For instance, a new clause might say, “A state's failure to pay the amount specified in the fourth criterion [i.e., FV_DE] due to a bona fide disagreement with the investor [over a defined set of issues] shall not entitle the investor to damages beyond the FV_DE.” Another could note that FV_DE must be calculated, if at all possible, based on information known at DE. And another might say, “A tribunal may award consequential damages to investors if and only if they can demonstrate the additional expenditures due to failure to meet the third [due process] requirement.” Along with reformed case law, any of these clauses would help move toward a true lex specialis for remedies for IIA violations.Footnote ²¹⁸

Yet, as noted earlier, this proposal is not simply transferable to a prescriptive process addressing a regime with different goals. It may well prove a poor fit for a human rights treaty or an investment insurance contract. Indeed, two or more states concluding an IIA might decide among themselves that expropriation should be compensated in yet another way, e.g., FV_DE no matter what the circumstances of the taking.