The problem of moral hazard and its implications for the protection of ‘legitimate expectations’ under the fair and equitable treatment standard

While the concepts of sovereignty, human rights, the environment and the rule of law are often invoked in public debate about international investment treaties (IITs), there is relatively little discussion of the economic effects of such treaties.[1] One of the most powerful legal protections provided by IITs is the protection of foreign investor’s ‘legitimate expectations’ under fair and equitable treatment (FET) provisions, which are common to most IITs. This article draws on economic theory—specifically, the notion of moral hazard—to elucidate some of the problems with broader interpretations of the doctrine of legitimate expectations.

The concept of efficiency

The concept of efficiency is central to economic analysis of public policy, including the economic analysis of legal rules. Efficiency concerns the maximising of net economic benefits; a policy improves efficiency if its economic benefits exceed its economic costs.[2] Economists have been criticised for focusing exclusively on maximising efficiency, to the neglect of other values.[3] These criticisms are important and well-made, yet they do not undermine the view that efficiency should be one of the criteria by which legal rules are evaluated. A rigorous examination of efficiency is especially important in the context of IIT interpretation, because a common justification for IITs is that they provide economic benefits.

How IITs affect efficiency

The primary means by which IITs affect economic efficiency is by influencing the investment decisions of foreign investors. A more efficient investment decision creates greater net economic benefits, regardless of to whom those benefits accrue. The profitability of an investment project is a first approximation of the efficiency gain of undertaking the project, because profit represents the excess of economic benefits of production over economic costs.[4] This is the foundation of the basic economic argument for IITs—that they give prospective investors confidence that their property and contractual rights will be protected abroad, encouraging them to reallocate their capital from less profitable projects in their home markets to more profitable projects available elsewhere.[5]

But this argument relies on a highly simplified model of economic activity. A more sophisticated model would acknowledge that there are often external costs and benefits of production that do not accrue to the investor. As discussed below, these externalities mean that the (un)profitability of a project does not necessarily imply its (in)efficiency.

The problem of moral hazard and why it reduces efficiency

Moral hazard refers to a situation where economic actors make profit-maximising but inefficient decisions because they are able to avoid costs associated with their conduct. The problem of moral hazard is often associated with insurance—when someone takes out insurance against a given type of harm, they no longer have an incentive to take prudent (efficient) steps to reduce the risk of that harm occurring. In practice, the protections contained in IITs operate as a form of insurance for investors against harm caused by future government conduct. This raises the risk of moral hazard— that investors might undertake projects without adequately assessing the externalities created by their projects, and the associated risk that future governments might redress such externalities when they begin to crystallise.

This problem has been explored in considerable detail in economic literature. The key insight that emerges from this literature is that protecting investors from having to bear the cost of new, efficiency-improving government measures is likely to result in inefficient investment decisions.[6] As such, investors should not be protected from efficient regulatory change, even if it results in the investment becoming unviable. Such legal protections would insure investors against changes in government policy, allowing them to ignore the risk posed to contemplated investments by efficiency-improving policy change.[7]

This scenario is easier to illustrate with an example. It would be inefficient for an investor to sink capital into building a factory which would operate at a profit of one thousand dollars a year by dumping pollutants in a river that cause two thousand dollars a year worth of damage to a downstream oyster industry. The most efficient investment decision would be for the investor not to undertake the investment in the first place and to allocate its capital to some other project. An investor that knew that future governments were free to prohibit dumping without compensating the investor would be less likely to commence such a project. On the other hand, an investor that knew that a future government would be required to pay compensation if it prohibited dumping would be far more likely to undertake the project.

In the example above, a prohibition on dumping pollutants is efficiency-improving because the benefits of the ban exceed the costs. In cases that come before arbitral tribunals there may be considerable evidentiary difficulties in determining whether given government measures are efficiency-improving. I do not suggest that tribunals should attempt to make such determinations on a case-by-case basis. Rather my argument is that the more expansive strands of current ‘legitimate expectations’ jurisprudence are highly likely to result in investors being compensated for losses caused by efficiency-improving government conduct. As such, these broader understandings of the doctrine of legitimate expectations induce moral hazard on the part of investors.

Existing jurisprudence on the protection of ‘legitimate expectations’ under the fair and equitable treatment standard

It is now widely accepted that fair and equitable treatment (FET) provisions, which are found in the vast majority of IITs, protect foreign investors’ legitimate expectations.[8] Despite this apparent consensus, arbitral tribunals have taken markedly different views of the range of expectations that might potentially qualify as ‘legitimate’ expectations. I identify four distinct views about the scope of the doctrine in contemporary arbitral decisions.[9]

The narrowest interpretation seems to require that an expectation be based on specific legal entitlements vested in a foreign investor under the law of the host state in order for such an expectation to be legitimate. For example, in LG&E v Argentina the tribunal held that an expectation would have to ‘exist and be enforceable by law’ to merit protection.[10] According to this view, the doctrine functions as an additional, international layer of protection for existing rights, rather than as a source of new rights.[11] A second view is that a legitimate expectation need not be based on the legal rights of the investor, so long as it based on specific, unilateral representations made by a government official.[12] This also seems to be the dominant view in academic commentary.[13] This set of decisions also embodies the limitation that expectations must be reasonable in light of the political and economic circumstances of the host state to be protected by the FET standard.[14]

A third strand of decisions suggests that an investor may legitimately expect the regulatory regime in place at the time of the investment to remain in force, even if the government has not promised to retain the regulatory regime and the investor has no legal right under domestic law to its continuance.[15] In these decisions the emphasis is on the protection of expectations that are ‘basic’ to the decision to invest;[16] there is far less emphasis on assessment of whether the basic expectation in question was reasonable in the circumstances.[17]

At the far end of the jurisprudential spectrum are a fourth group of cases in which investors have succeeded in claims for breach of legitimate expectations, despite the identified expectation having no base in the legal rights of the claimant under domestic law, nor in representations made by the host state or the regulatory arrangements in force at the time the investment was made. One such decision is Bau v Thailand.[18] Here the tribunal accepted that the investor had an expectation of a ‘reasonable rate of return’, where the expectation was based solely on the investor’s business plans at the time of making the investment.[19]

What the problem of moral hazard means for the doctrine of legitimate expectations

Extending legal protection to the basic expectations that underpin an investor’s business plan—as was done in Bau—is highly likely to induce moral hazard. A state’s liability under this interpretation of the doctrine of legitimate expectations does not turn on an examination of whether the state’s conduct was efficiency improving (either explicitly or de facto). Rather it provides investors with a degree of insurance against government actions that undermine the profitability of their investments, regardless of whether the government action is efficient. This is precisely the sort of legal rule that is likely to discourage investors from internalising the risk to their business plans posed by future, efficiency-improving government conduct.

Protecting general expectations of regulatory stability is also likely to cause serious problems of moral hazard. Arbitral decisions that have applied this understanding of the doctrine of legitimate expectations have determined liability by assessing whether the altered regulation was ‘basic’ to the investor’s decision to invest. This approach eschews economy-wide judgement of whether the regulatory change was efficiency-improving, in favour of an assessment of whether the change has seriously affected the investor’s interests. Such expansive legal protection is unjustifiable from an economic perspective. An investor should be required to bear the risk of efficient regulatory change, because that risk plays an important role in discouraging investors from initiating socially undesirable investment. This holds true even in cases where regulatory change is ruinous of an investment.


This article argues that two of the broader interpretations of the doctrine of legitimate expectations are likely to reduce economic efficiency on account of inducing moral hazard on the part of foreign investors. This is because broader interpretations of the doctrine provide foreign investors with too much protection from regulatory and policy change. Economic theory shows why leaving foreign investors exposed to the risk of certain types of policy change plays a crucial role both in dissuading foreign investors from undertaking projects that are not in the public interest, and in encouraging foreign investors to structure the projects they do undertake in a way that minimises external costs. This is an important conclusion because it illustrates that stronger legal protections of this type for foreign investment are not necessarily desirable on economic grounds; indeed, sometimes they are profoundly undesirable.

There may be a number of other grounds on which one could criticise (or defend) broad protection of an investor’s expectations. This piece has not attempted to assess these arguments. By putting environmental and human rights arguments to one side, I do not intend to suggest that they are less important than economic arguments, nor to suggest that dumping pollution in a river is only objectionable if it inhibits potentially more lucrative investment downstream. Rather I hope to show that debate about IITs does not necessarily reduce to an argument about the importance of economic development vs. the importance of environmental protection and realisation of human rights. My own view is that narrower interpretation of key IIT provisions would be preferable on economic, human rights, environmental and rule of law grounds.[20]

Author: Jonathan Bonnitcha is a lawyer, Rhodes Scholar and final year DPhil candidate at the University of Oxford. He holds the degrees of MPhil (Dist) and BCL (Dist) from Oxford and LLB (Hons) and B EC (Hons) from the University of Sydney. Readers with comments or criticisms of this article are invited to contact the author at

[1] With the exception of empirical scholarship seeking to determine whether signing a bilateral investment treaty increases foreign direct investment in signatory States; see, UNCTAD, The Role of International Investment Agreements in Attracting Foreign Direct Investment to Developing Countries (United Nations, Geneva 2009)

[2] This is a colloquial definition of Hicks-Kaldor efficiency. When economists evaluate legal rules on the grounds of efficiency they are invariably invoking the notion of Hicks-Kaldor efficiency. R Posner, Economic Analysis of Law (Aspen Publishers, New York 2007)

[3] G Calabresi, ‘The Pointlessness of Pareto’ (1991) 100 (1211) Yale Law Journal ; A Sen, ‘The Impossibility of a Paretian Liberal’ (1970) 78 Journal of Political Economy 152.

[4] L Blume and D Rubinfeld, ‘Compensation for Takings: An Economic Analysis’ (1984) 72 California Law Review 569

[5] A Van Aaken, ‘Perils of Success? The Case of International Investment Protection’ (2008) 9 European Business Organization Law Review 1; E Aisbett, L Karp and C McAusland, ‘Compensation for Regulatory Taking in International Investment Agreements: Implications of National Treatment and Rights to Invest’ (2011, forthcoming) Journal of Globalization and Development

[6] L Blume, D Rubinfeld and P Shapiro, ‘The Taking of Land: When Should Compensation Be Paid?’ (1984) 99 (1) The Quarterly Journal of Economics 71; T Miceli and K Segerson, ‘Regulatory Takings: When Should Compensation Be Paid?’ (1994) 23 Journal of Legal Studies 749

[7] E Aisbett, L Karp and C McAusland, ‘Police powers, regulatory taking and the efficient compensation of domestic and foreign investors’ (2010) 86 The Economic Record 367; J Stiglitz, ‘Regulating Multinational Corporations: Towards principles of cross-border legal frameworks in a globalized world balancing rights with responsibilities ‘ (2008) 23 American University International Law Review 451

[8] C Schreuer and U Kriebaum, ‘At What Time Must Legitimate Expectations Exist?’ in J Werner and AH Ali (eds) A Liber Amicorum: Thomas Wälde Law Beyond Conventional Thought (Cameron May, London 2010); A Von Walter, ‘The Investor’s Expectations in International Investment Arbitration'(2008); E Snodgrass, ‘Protecting Investors’ Legitimate Expectations: Recognizing and Delimiting a General Principle’ (2006) 21 ICSID Review – Foreign Investment Law Journal 1; A Newcombe and L Paradell, Law and Practice of Investment Treaties: Standards of Treatment (Kluwer Law International, Alphen aan den Rijn 2009); R Dolzer and C Schreuer, Principles of International Investment Law (Oxford University Press, Oxford 2008); I Tudor, The Fair and Equitable Treatment Standard in the International Law of Foreign Investment (Oxford University Press, Oxford 2008)

[9] This is a significantly abbreviated presentation of a taxonomy of legitimate expectations decisions that I develop in my doctoral dissertation. J Bonnitcha ‘A normative framework for evaluating interpretations of interpretations of investment treaty protections’  (University of Oxford, forthcoming April 2011)

[10]  LG&E Energy v Argentina ICSID Case No ARB/02/1, Award, 25 July 2007, [130]. Similarly,  MCI Power Group v Republic of Ecuador ICSID Case No ARB/03/6, Award, 31 July 2007, [278];  BG Group v The Republic of Argentina Final Award, 27 December 2007, [305]-[308]; Suez, Sociedad General de Aguas de Barcelona and Vivendi Universal v Argentine Republic ICSID Case No ARB/03/19 and AWG Group v Argentine Republic, Decision on Liability, 30 July 2010, [79]-[80]

[11] According to this interpretation, the doctrine of legitimate expectations operates somewhat akin to an umbrella clause, in that it protects foreign investors’ rights created by contracts with the host state. However, the legal effect of this interpretation differs from a strictly drafted umbrella clause in that only a sufficiently serious breach of expectations based on an investor’s legal rights would constitute a breach of legitimate expectations.

[12]  International Thunderbird Gaming v United Mexican States Arbitral Award 26 January 2006, [147], [196];  National Grid v Argentine Republic Award, 3 November 2008, [173]; Duke Energy Electroquil Partners v Republic of Ecuador ICISD Case No ARB/04/19, Award, 18 August 2008, [340]

[13] R Dolzer, ‘Fair and Equitable Treatment: A Key Standard in Investment Treaties’ (2005) 39 The International Lawyer 87; C Schreuer, ‘Fair and Equitable Treatment in Arbitral Practice’ (2005) 6 The Journal of World Investment and Trade 357; Snodgrass; S Fietta, ‘Expropriation and the « Fair and Equitable Treatment » Standard’ in F Ortinoand others (eds) Investment Treaty Law: Current Issues II (British Institute of International and Comparative Law, London 2007); S Westcott, ‘Recent Practice on the Fair and Equitable Treatment’ (2007) 8 The Journal of World Investment and Trade 409; M Kinnear, ‘The Continuing Development of the Fair and Equitable Treatment’ in A Bjorklund, I Laird and S Ripinsky (eds) Investment Treaty Law: Current Issues III (British Institute of International and Comparative Law, London 2009); J Salacuse, The Law of Investment Treaties (Oxford University Press, Oxford 2010).

[14]  Biwater Gauff v United Republic of Tanzania ICSID Case No ARB/05/22, Award, 24 July 2008, [601]; Saluka Investment BV v Czech Republic Partial Award, 17 March 2006, [305];  Continental Casualty v The Argentine Republic ICSID Case No ARB/03/9, Award, 5 September 2008, [261]

[15] CME v Czech Republic Partial Award, 13 September 2001, [611];  Occidental Exploration and Production Company v the Republic of Ecuador LCIA Case No UN3467, Final Award, 1 July 2004, [184];  Enron Corporation v Argentine Republic ICSID Case No ARB/01/3, Award, 22 May 2007, [267]

[16]  Técnicas Medioambientales Tecmed v United Mexican States ICSID Case No ARB(AF)/00/02, Award, 29 May 2003, [154]

[17] According to this interpretation, the doctrine of legitimate expectations operates somewhat akin to a stabilization clause, in that it protects foreign investors from change to the legal regime governing an investment. However, the legal effect of this interpretation differs from a broadly drafted stabilization clause in that only changes to regulations that were ‘basic’ to the decision to invest would constitute a breach of legitimate expectations.

[18] Bau v The Kingdom of Thailand Award, 1 July 2009, [12.3]. Similarly,  Bogandov v Republic of Moldova Arbitral Award, 22 September 2005, 17;  MTD v Chile ICSID Case No ARB/01/7, Award, 21 May 2004, [163]

[19] The tribunal held that:

In spite of the fact that there was no guarantee by the Respondent of an explicit rate of return, the Tribunal considers that a reasonable rate of return – reasonable in all the circumstances, including the signing of MoA2 – was part of the Claimant’s legitimate expectations and the failure to fulfil such a reasonable expectation was a breach of the Respondent’s obligations.

[20] I develop this argument at length in my doctoral dissertation, J Bonnitcha ‘A normative framework for evaluating interpretations of interpretations of investment treaty protections’  (University of Oxford, forthcoming April 2011)