Recall this post on the EU position in the Philippines - Distilled Spirits (DS396,403) case in relation to Article III:2, first sentence, where I characterized the EU argument on the discrimination aspect of this standard as follows: "the discriminatory effect (or disparate impact or disproportionate disadvantage or asymmetric impact, or whatever you want to call it) approach is mixed together with the individual product (or diagonal test or best treatment, or whatever you want to call it) approach."
Here's the U.S., in its second written submission, on the same issue:
13. To find that the Philippine measures are inconsistent with the first sentence of Article III:2, the second element simply requires showing that the imported products are taxed in excess of domestic products. This conclusion is evident from the face of the measures – including the implementing annexes, which separate “local” from other brands. These documents show that nearly every imported product is taxed at one of the high rates – from ten to 40 times the rate applied to local products on a proof liter basis.
So the U.S. submission emphasizes that "nearly every" imported product is taxed at one of the high rates, which are much higher than those applied to domestic products.
The "nearly every" language seems to reflect a discriminatory effect view. (The EU submission used similar language, but then mixed in the individual product approach as well.) It's important to note, though, that the U.S. does not put this forward as the definitive legal standard which must always be applied. It is just saying that this is the case on the facts here. At the same time, it is perhaps relevant that the U.S. does not include the individual products language.