Before getting into the substance of the Seal Products dispute, let me take this opportunity to thank Rob for all of his reporting on the panel meetings. I don't think I've ever come across such a detailed analysis of this part of the panel process. I would guess that even people who disagreed with Rob's view of the case appreciated the perspective he provided.
Turning to the case itself, what struck me about Rob's description of the proceedings is how bogged down these cases can get with very narrow issues. Sometimes the big picture gets lost.
So, I wanted to very briefly describe how I am thinking about the discrimination issues. I'm not going to cite any cases, or say how I think this case should come out. I'm just going to offer my framework for analyzing these issues. I've said much of this before in various contexts, so it will probably be familiar to some of you. (And no doubt some people disagree with how I look at it -- if so, feel free to say so in the comments!)
To me, there are two important components to the discrimination analysis: effect (or impact), and intent (or aim, purpose, motivation).
With regard to the effect of the measure, in a de facto case like this one, I want to know what the degree of disparate impact is. What percentage of the complaining country's (like) products are facing the bad treatment, as compared to the percentage of the defending country's products or some third country's products? At one extreme, all of the complaining country's products could face the bad treatment, while all of the defending country's products receive the good treatment. But you could also have a situation where about half the products of each country receive the bad treatment, and half receive the good treatment. Or something in between. In my view, the greater the degree of disparate impact, the stronger the discrimination claim would be.
As to the intent of the measure, I know that subjective intent is controversial, and I'm not really focusing on that here (although I don't have a problem with taking it into account as one factor). Instead, I'm thinking of objective intent. It's hard to pin down exactly what that means, but let me give some examples of what might be part of this analysis: Does the design, structure, and architecture of the measure suggest it was intended to protect domestic producers, or is it more consistent with another policy goal? As with effect, this is also a matter of degree. Specific questions in this regard might be (but are not limited to): Is the measure based on science? Is it part of an overall consistent policy in relation to the stated policy goal? To what degree does the measure achieve its stated goals? No single one of these factors is determinative. You have to weigh and balance them all.
To be clear, I'm not even thinking about how exceptions might apply. This is just about whether there is a violation of the non-discrimination obligations. (And yes, I know there might be a distinction between GATT III:4 and TBT 2.1, but I'm ignoring that as well.)