Last week I talked about National Treatment in the context of the hearing in the Grand River Enterprises NAFTA Chapter 11 case. Stimulated by re-reading an article by Nicholas DiMascio & Joost Pauwelyn, "Nondiscrimination in Trade and Investment Treaties: Worlds Apart or Two Sides of the Same Coin?" (available from HeinOnline here), I've been thinking a bit more about National Treatment in the area of investment treaty law. In particular, I've been trying to flesh out the possible standards for "less favorable treatment." In this post, I try to articulate as clearly as I can my view of things. (Of course, others might have a completely different perspective on how to approach this issue, which can make a discussion difficult, but there's only so much you can do in a blog post!)
First off, let me note that I'm focusing here on issues of de facto discrimination, not de jure. With de jure, where government actions explicitly discriminate on the basis of foreign nationality (e.g., an official statement by a government agency says that a permit was denied "because the company is foreign"), no one questions that a violation of National Treatment exists. The theory behind why that is so can be more complicated than it appears at first glance, and there are some views of de jure violations that are not what you might expect, but overall it's not too controversial and thus I'm going to focus on de facto cases.
Also worth noting is that my emphasis here is on the "less favorable treatment" aspect of National Treatment, rather than "in like circumstances." (A little more on that later, though).
In my view, the two key elements in thinking about less favorable treatment are: (1) individual vs. group comparisons; and (2) the role of discriminatory effect and intent. (There are many other factors that have been suggested -- people are free to point them out in the comments -- but as a general matter I would characterize these other factors as falling within effect or intent, e.g., if a measure does not reasonably relate to its purported non-protectionist goals, that could be evidence that the true intent is protectionist.) Taking into account both elements, here's how I see the breakdown of the possible approaches to less favorable treatment:
1. Individual comparison and effect
Under this view, less favorable treatment exists where an individual foreign investor is treated worse than an individual domestic investor (in like circumstances, of course -- this applies to all the examples). The effect of discriminating between these two investors leads to a violation. For practical purposes, this standard is a "best treatment" standard, as a violation exists whenever a foreign investor does not get the "best treatment" offered to any domestic investor.
2. Individual comparison and intent
Under this view, less favorable treatment exists where there is evidence of intent to treat an individual foreign investor worse than an individual domestic investor, on the basis of the foreign investor's nationality. (In the alternative, this situation could be characterized as intending the result that the foreign investor is treated less favorably, without any reference to nationality, but I'm not sure this differs much from the effect standard in example 1).
3. Group comparison and effect
Under this view, less favorable treatment exists where, on balance, the group of foreign investors is treated worse than the group of domestic investors. The effect of discriminating between these two groups leads to a violation.
4. Group comparison and intent
Under this view, less favorable treatment exists where there is evidence of intent to treat the group of foreign investors worse than the group of domestic investors, on the basis of the foreign investors' nationality.
**********************************************
In addition, you could require that both intent and effect be demonstrated for the two types of comparisons.
Note that situations 2 and 4 may not be all that different. If there is evidence of intent that an individual foreign investor is to get worse treatment due to its nationality, it's hard to imagine that the group test for intent would not be satisfied as well. That's because if you are discriminating against one investor based on its foreign nationality, it stands to reason that you would do the same to other investors from that country. If you are not discriminating against the other foreign investors, perhaps foreign nationality was not the reason for discriminating against the first investor after all. So it may be that once intent to discriminate due to nationality is proved, individual vs. group distinctions do not matter that much.
The key issue, in my view, is how to deal with example 1, involving individual comparisons and effect. Finding that less favorable treatment exists on this basis is going to lead to a lot of violations and does not always look like what many people intuitively think about as discrimination. Almost any regulatory distinction between investors is at risk under this approach, as long as there is one foreign investor who is adversely affected.
Let me say a little bit now about "like circumstances," which can play an important role here. At various times in GATT/WTO and investment law, some of the concepts from less favorable treatment noted above, such as the intent (policy goal) of the measure, seem to have been brought into the "like circumstances"/"like products" standards. That may be the situation in investment law right now, at least if I read the DiMascio and Pauwelyn article (at p. 73) correctly:
The tribunal in the Pope & Talbot lumber products dispute subsequently refined the analysis from S.D. Myers into a coherent "like circumstances" test. 142 Based upon reasoning similar to that used in the "like circumstances" analysis of S.D. Myers, the tribunal initially presumed that any regulatory treatment of Pope & Talbot that was less favorable than the treatment received by Canadian softwood lumber producers was attributable to the company's nationality. This presumption, which shifts the burden of proof from the complainant to the regulating defendant, obviously makes the NAFTA national treatment test more intrusive. The tribunal then stated, however, that the presumption could be rebutted by showing that the measure has "a reasonable nexus to rational government policies that (1) do not distinguish, on their face or de facto, between foreign-owned and domestic companies, and (2) do not otherwise... undermine the investment liberalizing objectives of NAFTA."143 If the measure distinguished between Pope & Talbot and the Canadian investments based upon such a legitimate policy, those companies could not be said to be in "like circumstances," and the tribunal would conclude that no nationality-based discrimination had taken place even if Pope & Talbot was competitively disadvantaged by Canada's regulation.144
Thus, stated somewhat loosely, the individual comparison effect test creates a presumption of less favorable treatment. But this presumption can be rebutted under the like circumstances language. When determining whether investors are in "like circumstances," a relevant factor is the existence of policies that are not protectionist, in the sense of "a reasonable nexus to rational government policies that (1) do not distinguish, on their face or de facto, between foreign-owned and domestic companies, and (2) do not otherwise... undermine the investment liberalizing objectives of NAFTA." Where such policies form the basis of the measure, the circumstances are not "like."
Arguably, a similar approach was used under the GATT in some of the early 1990s cases, such as U.S. - Malt Beverages and U.S. - Auto Taxes. Here's a quote from from Malt Beverages:
5.74 The Panel then turned to a consideration of the policy goals and legislative background of the laws regulating the alcohol content of beer. In this regard, the Panel recalled the United States argument that states encouraged the consumption of low alcohol beer over beer with a higher alcohol content specifically for the purposes of protecting human life and health and upholding public morals. The Panel also recalled the Canadian position that the legislative background of laws regulating the alcohol content of beer showed that the federal and state legislatures were more concerned with raising tax revenue than with protecting human health and public morals. On the basis of the evidence submitted, the Panel noted that the relevant laws were passed against the background of the Temperance movement in the United States. It noted further that prior to the repeal of the Eighteenth Amendment of the United States Constitution authorizing Prohibition, amendments to the federal Volstead Act -- the Act which implemented the Eighteenth Amendment -- authorized the sale of low alcohol beer, and that the primary focus of the drafters of these amendments may have been the establishment of a brewing industry which could serve as a new source of tax revenue. However, irrespective of whether the policy background to the laws distinguishing alcohol content of beer was the protection of human health and public morals or the promotion of a new source of government revenue, both the statements of the parties and the legislative history suggest that the alcohol content of beer has not been singled out as a means of favouring domestic producers over foreign producers. The Panel recognized that the level at which the state measures distinguished between low and high alcohol content could arguably have been other than 3.2 per cent by weight. Indeed, as the Panel previously noted, Alabama and Oregon make the distinction at slightly different levels. However, there was no evidence submitted to the Panel that the choice of the particular level has the purpose or effect of affording protection to domestic production.
5.75 Thus, for the purposes of its examination under Article III, and in the context of the state legislation at issue in Alabama, Colorado, Florida, Kansas, Minnesota, Missouri, Oklahoma, Oregon and Utah, the Panel considered that low alcohol content beer and high alcohol content beer need not be considered as like products in terms of Article III:4. ...
(emphasis added) So, the consideration of the non-protectionist policy purposes (and effects) led to a conclusion that the products were not "like." (See also paras. 5.11-15 of Luxury Taxes)
Personally, I prefer to keep the two standards clearly separated, with the "likeness" issues a bit more narrow. I fear that blurring the lines can make the issues more complicated than they need to be. But I think that's a subject for another post.
This all leads me back to a question I asked the other day in the context of Grand River Enterprises. The U.S. has argued that NAFTA Article 1102 National Treatment is about "nationality" based discrimination. So what exactly is "nationality" based discrimination? I assume they have in mind something along the lines of 3 or 4, but which of these is it? Or are there other options I have left out?
