What the close decision on Philip Morris tells us about ISDS
PNG - 682.1 kb

Todd N. Tucker (blog) | 19 July 2016

What the close decision on Philip Morris tells us about ISDS

Sometimes corporations don’t get what they want. That was the case earlier this month when a World Bank arbitration tribunal ruled against tobacco giant Philip Morris in its suit against Uruguay. In a split decision, a majority of two arbitrators sided with the Latin American sovereign’s right to regulate, while a dissenting arbitrator sided partly with the corporation (which had appointed him to the panel).

Philip Morris launched the case in 2010, after Uruguay introduced a set of cigarette labeling policies to deter smoking. First, all cigarette packages were required to have graphic warning labels on 80 percent of their surface area. Second, each cigarette brand family could have only one design presentation. (Uruguay argued that tobacco companies used different color and design schemes to suggest certain variants were healthier than others.) These policies were driven by the administration of Tabaré Vázquez, a left-leaning oncologist first elected president in 2005. Although Philip Morris is headquartered in the U.S.—and although the U.S. and Uruguay have a bilateral investment treaty, or BIT—the tobacco company used its Swiss subsidiary and a Switzerland-Uruguay BIT to bring the case, as it is permitted to do.

In recent years, the case had become the poster child for critics of so-called investor-state dispute settlement, or ISDS. This legal procedure is contained in thousands of international trade and investment treaties, and permits foreign investors to sue governments over alleged harms suffered from regulation. Critics maintain that ISDS chills ambitious public interest regulation, while proponents argue that it helps reassure foreign capital that it will be treated fairly overseas. Once an obscure issue, the investment rules catapulted into the national spotlight in the past year, figuring prominently on late-night talk shows and the Democratic Party’s platform fight over the Trans-Pacific Partnership, or TPP.

Despite (or likely because of) critics’ attention to the case, the arbitration tribunal did pretty much what you would want an international court-like body to do. First, the arbitrators clearly understood both the business and regulatory issues that concerned both Philip Morris and Uruguay. Second, the tribunal refused to second-guess the public health experts. Third, the arbitrators refused to second-guess the Uruguayan courts that had already heard the case. Finally, even the dissenting arbitrator raised reasonable points and did not toe the party line of his appointer.

I’ll go into more detail on each of these points below.

Sensitivity to Context

Throughout the 170-page opinion and 46-page dissent, you find phrases like this: “[T]he Challenged Measures were a valid exercise by Uruguay of its police powers for the protection of public health.” Or, “[I] do not question the broad authority of Uruguay, or other states, to regulate in the interest of public health and safety.” Such pro-regulation language is consistent with how the Obama administration describes the TPP—to the sharp objection of many progressives.

This linguistic couching in itself is not novel. Political scientist Cosette Creamer has shown that adjudicators at the World Trade Organization use pro-sovereign rhetoric to soften the blow of adverse decisions.

But the Philip Morris tribunal was especially deferential, noting the stakes involved:

Uruguay has one of Latin America’s highest rate of smokers, being in third place in the region after Chile and Bolivia. As of 2009, more than 5,000 Uruguayans died each year from diseases linked to tobacco consumption, mainly due to cardiovascular diseases and cancer. Consumption of tobacco and exposure to tobacco smoke are responsible for 15% of all deaths of Uruguayans over 30 years of age, which is higher than the world average of 12%.

Smoking also has an economic impact. Uruguayan smokers spent an average of 20% of the national minimum wage to sustain their habit and the health costs linked to smoking in Uruguay are estimated to amount to US$150 million per year.

Against this background, Uruguay has positioned itself in the forefront of States in terms of anti-smoking policy and legislation… the Challenged Measures were introduced as part of a larger scheme of tobacco control, the different components of which it is difficult to disentangle. But the fact remains that the incidence of smoking in Uruguay has declined, notably among young smokers, and that these were public health measures which were directed to this end and were capable of contributing to its achievement. In the Tribunal’s view, that is sufficient for the purposes of defeating [an investment treaty] claim…

Sensitivity to Experts

The arbitrators echoed a point I’ve often raised: that investment treaties do not should not represent “a justiciable standard of good government… How a government requires the acknowledged health risks of products, such as tobacco, to be communicated to the persons at risk, is a matter of public policy, to be left to the appreciation of the regulatory authority…”

Why am I applauding BITs not providing “good governance”? Well, as social scientist Matt Andrews has pointed out, the phrase (popular in the development community) is often evoked opportunistically to slam ideas unpopular with elite researchers. That doesn’t make it necessarily a bad idea, just not a very useful one. So it’s good not to hang a lot of legal consequences on it—especially if you’re an institution-poor developing country.

The arbitrators did show some of the perils of the specific way that lawyers evaluate scientific evidence.* A crucial if not decisive factor for them was that it was the international community—not Uruguay—that had identified the problem and provided the solution.

For a country with limited technical and economic resources, such as Uruguay, adhesion to the [World Health Organization’s Framework Convention on Tobacco Control] FCTC and involvement in the process of scientific and technical cooperation and reporting and of exchange of information represented an important if not indispensable means for acquiring the scientific knowledge and market experience needed for the proper implementation of its obligations under the FCTC and for ensuring the fulfilment of its tobacco control policy…

As for the company’s arguments that Uruguay should have conducted its own research:

In the Tribunal’s view, in these circumstances there was no requirement for Uruguay to perform additional studies or to gather further evidence in support of the Challenged Measures. Such support was amply offered by the evidence-based FCTC provisions and guidelines adopted thereunder.

There’s certainly nothing wrong with relying on highly competent international proceedings as a baseline for evaluating policy. But there is a risk in doing so. We know from the bounded rationality literature that policies sometimes diffuse because of groupthink. To legally privilege only widely emulated policies risks compounding that, and penalizing the truly original solution.

Of course, the majority was sensitive to this problem as well, at least rhetorically. They wrote:

[It is not] a valid objection to a regulation that it breaks new ground. [BITs] do not preclude governments from enacting novel rules, even if these are in advance of international practice, provided these have some rational basis and are not discriminatory. [A BIT provision] does not guarantee that nothing should be done by the host State for the first time. (emphasis added)

Sensitivity to National Courts

As in the parallel Philip Morris v. Australia ISDS case, the company-country dispute had already made its way through national courts—up to and including the Uruguayan Supreme Court. And as with the Australia case, the ISDS tribunal sided with the country. As the majority wrote, “the tribunal is not a court of appeal.”

Dissenter reasonable

One-person dissenting opinions are a fairly common feature of three-member arbitration tribunals. As my research has shown, dissents can be triggered because of intellectual or ideological leanings of the arbitrator, or simply because the presiding arbitrator has poor people skills.

When I sit down to read a dissent, I’m expecting to read a pretty hard line one way or the other.

That was not the case with Gary Born’s dissent in this case. He did not side with the company on all its claims. Instead, he focused on two points that (I’m sure Uruguay’s legal counsel would agree) were the weakest for the country, and for which Piero Bernardini (presiding arbitrator) and James Crawford (Uruguay’s appointee) were clearly at pains to account.

First, Born noted that Uruguay’s legal system suffers from a gaping design flaw. Like many legal systems, Uruguay has courts that are part of the judicial branch of government and court-like administrative tribunals that form part of the executive power. Unlike those systems, certain decisions of Uruguay’s administrative courts cannot be appealed to a higher authority in the judiciary.

In this case, Uruguay’s judicial branch had sided with Philip Morris’s legal claim, finding that the executive branch did not enjoy (and indeed, had not been granted) legislative authorization for increasing the warning labels up to 80 percent of the cigarette carton. But the subsequent decision by the administrative courts ruled in exactly the opposite fashion.

Born rightly noted that:

[T]his Opinion is directed towards a highly unusual aspect of the Uruguayan legal system, which produced a result in this case that has never previously occurred under Uruguayan law… two of the country’s highest civil courts reached directly contradictory interpretations of precisely the same statutory provision, in closely-related proceedings involving claims by the same party against the government, with these contradictory interpretations then being applied, in each case, to deny that party relief. Moreover, that same party was then left with no judicial forum in which to assert otherwise available constitutional challenges to the relevant statutory provision, as it had been authoritatively interpreted and applied to that party. In my view, that unprecedented result plainly constituted a denial of justice…

One can certainly sympathize with this position, and wonder if BITs might not provide some reasonable protection for foreign investors getting put in such unprecedented legal vises.

But think it through a bit more. How exactly would the government have avoided treaty liability? It would have had to intervene in one or both independent courts to “make things right.” As I’ve written, such a move would be good for the company but bad for domestic checks and balances.

Second, Born was also suspicious of the plank of Uruguay’s policy that standardized the color and design schemes. As he shows persuasively, the policy had a quirk that meant that these cigarette packages were prohibited:

PNG - 160.4 kb

… while these were not:

PNG - 118.9 kb

In other words, companies could get around the branding restrictions by simply putting different brand names at the top.

As Born shows, this branding restriction was added at the eleventh hour as a result of an apparent brainstorm, without expert or scientific deliberation.

So, the dissenter raises a compelling point. But should BIT tribunals of ad hoc experts substitute their hindsight wisdom for that of the local government? Would one rather take the risk that a local government (accountable to local people who can vote them out) gets it wrong, or that a company-appointed arbitrator gets it wrong?


So, all in all, a pretty good decision—although its limitations should be acknowledged. First, their reasoning won’t be binding on later tribunals. Will a future tribunal in an obscurer case be so deferential to sovereigns? There are plenty of cases in which they haven’t been. And Uruguay (and its patrons in the Gates Foundation and Bloomberg Philanthropies) still had to spend time and resources arguing the case. Some critics will see that as per se objectionable for regulation already blessed by the public health community and national courts. And the very saving grace of the case—arbitrators that can read political winds—cuts against the supposed depoliticization of the international legal proceedings. (Notably, WTO adjudicators have been more willing to second-guess anti-tobacco regulations.)

Nonetheless, Uruguay’s victory highlights the difficulty of advocating against legal systems that consolidate their influence gradually via highly technical case-by-case expansion. Nearly every time critics tried to paint a pending case in a negative light, arbitration tribunals sided with the state.** Critics are a victim of their own spotlighting success. Stay tuned for if and how this affects the TPP debate.

Read the award here.


* A related problem of “too much lawyering” was evident in the arbitrators’ passing mention that states can “contract” for total stabilization of the regulatory climate (para. 481).

** For the latest example, check out yesterday’s news of the dismissal of a case over Peruvian mining and toxic clean-up policy.