A few years ago, stabilization clauses in investment contracts became the subject of increased attention by human rights and development groups. A report on Stabilization Clauses and Human Rights, issued by the UN Secretary-General’s Special Representative for Business and Human Rights, John Ruggie, was the first comprehensive study to draw on a range of heretofore confidential transactions covering a range of investment projects around the world.

For contract negotiators, these clauses continue to pose some perplexing questions on the proper allocation of transaction’s risk between an investor and a host state. For the arbitration community, it’s worth considering how these clauses work, what they do for investors, and how they could—and should—be adjudicated in arbitration.

There are two basic types of stabilization clauses. “Freezing clauses” act exactly as they sound—they “freeze” the law at the time the contract is executed for that particular investor. Any future changes in legislation or regulation are not applied to the contract. A potentially more nuanced varietal of the stabilization clause is the “economic equilibrium clause,” which comes in two versions. Under a rigid economic equilibrium clause, future changes in law would apply to the investor, but the host State would indemnify the investor for its compliance with the new legislation. Under a more flexible equilibrium clause, the host state and the investor would commit to conducting future negotiations with the goal of recalibrating the original allocation of risks or losses/gains, based on the reality of the new legislation.

At the 10,000 foot view, the purpose of stabilization clauses is pretty straightforward. They are risk allocation tools, designed to increase the predictability of the regulatory environment in which the investor will be operating. Stability clauses are used throughout the world and in a variety of industries, and are often used as a means to mitigate risks with respect to a host state’s future fiscal regulations.

What the Ruggie study pointed to, however, was the disparity in the way stability clauses are used, depending on the particular host state involved. The report found a fairly stark difference in the way that stability clauses are used in contracts from OECD-member states and in non-OECD-member states. OECD states tend use the clauses fairly consistently, and tend to limit the investor’s protection from the application of new laws to only those laws that are arbitrary or discriminatory. Investors tend to assume the risk that they will be subject to new laws of general application.

But when it comes to non-OECD states, stability clauses tend to be all over the place. More of them are generic and across-the-board, precluding the application of or providing compensation for compliance with both arbitrary/discriminatory new laws and bona fide new laws across a state’s full regulatory spectrum.

Some of this differential treatment can be explained. Investors are particularly sensitive to risk allocation when it comes to big-scale long-term investments, particularly in developing countries. Obsolescence bargaining can be a justifiable fear, especially when the other party is an emerging economy. No investor wants to make a large up-front investment or rely on non-recourse funding without some reassurance that the host state will maintain a stable, predictable regulatory environment for their investment. This is particularly the case when the host state has an under-developed regulatory environment, where there could be changes in leadership, or where current governments—populist and undemocratic alike—inspire less than full confidence that they will refrain from opportunistic regulatory behavior in the future.

At the same time, stabilization clauses—and particularly freezing clauses—can cut broadly in the investor’s favor. Ruggie notes that there have been a number of cases in which a broadly-worded stabilization clause permitted the investor to avoid compliance with all new laws that might affect the investor—including regulations that promote a host state’s environmental, social or human rights goals.

The most troublesome forms of these clauses have the potential to strip states of their sovereign regulatory power. They could force a state to forego its international human rights or environmental commitments, or to pay a heavy price if they do so. And they potentially provide a windfall to investors eager to take advantage of a lax regulatory environment and shift some costly externalities onto the public.

Public outcry has led some investors to revisit their contracts and reduce the scope of their stabilization clauses. This was the case for the Mittal Steel’s Mineral Development Agreement, which it had originally negotiated with Liberia under the state’s shaky post-conflict transitional government. But the full picture on these clauses—how many investors have insisted on them, and in which states—remains obscured by confidentiality provisions.

As a few academics have argued (here and here), even the less egregious stabilization clauses have the potential to put host states on the line to compensate an investor for changes in regulation, even when those regulatory changes would not rise to the level of an expropriation under international law. In other words, even if an investor would not be able to claim compensation under the prevailing “substantial deprivation” standard for expropriation, that same investor could seek compensation for a host state’s breach of the contract’s stabilization clause. Even if the new regulation was non-discriminatory and of general application, the state could be penalized for applying it to the investor.

This raises some interesting implications for arbitrators. As one study points out, while stabilization clauses involving nationalization have been upheld by tribunals in the past, there is yet to be a publicly-available decision suggesting how a tribunal might respond to a freezing clause that would limit a state’s capacity to regulate for the public good. And even for less egregious freezing clauses, it’s worth considering the propriety of forcing a developing state to pay for new legislation that is nondiscriminatory, of general application, and for the common good. Most likely, the remedy the investor would obtain for a breach of the stabilization clause would be lower than it would be if the investor had been able to persuade a tribunal that expropriation had occurred. Even then, it’s worth questioning whether the penalty—and the regulatory chill it could lead to—makes sense.


To make sure you do not miss out on regular updates from the Kluwer Arbitration Blog, please subscribe here. To submit a proposal for a blog post, please consult our Editorial Guidelines.

Kluwer Arbitration
This page as PDF


  1. Well, my investments are indeed in the deep freeze, but it turns out that this was not just another post in my Facebook feed about the atrophying of American prosperity, but was in fact something interesting, if somewhat arcane. I feel like I need a Dole before weighing in.

    Just an idle thought I had when reading it: How long do stabilization clauses usually last? Is there a standard term of x years?

  2. Hi John! There’s not any standard term that I know of. While certainly the parties could put a time limit on the applicability of the clause, many of them are intended to cover the investment in toto. Ruggie’s paper gives some sample language that’s perhaps helpful:

    “Specific Juridical Stability: The State guarantees the Investors and the Recipient Company that this Investment Contract, the Project Agreements and the State Institution Authorizations, in each case in relation to the Investments and the Project, shall enjoy absolute legal stability in accordance with the Legal Framework in Effect. Accordingly, neither the Investment Contract, nor the Project Agreements nor the State Institution Authorizations may be modified unilaterally by laws or other dispositions from the State of any type that affect them or by changes in the interpretation or application thereof and each thereof in which the State is a party may only be modified by the mutual written agreement of the Parties that expressly evidences such modifications.”

Leave a Reply

Your email address will not be published. Required fields are marked *

This site uses Akismet to reduce spam. Learn how your comment data is processed.