International law has significant limitations. It’s “laws” are non-binding and unenforced. Most states use law as a framework to normalize diplomatic behavior.

Rulings from the GATT and the WTO adjudication courts are not obeyed the majority of the time. These international arbitration courts rule on trade disputes, which is supposed to be one of the more established and respected international legal branches.
In Adjudication without Enforcement Eric Reinhardt gathered empirical evidence in an attempt to solve the “puzzle” in GATT.

From the Abstract:

Disputes under the General Agreement on Tariffs and Trade (GATT) exhibit a puzzling selection effect. Defendants concede more prior to GATT judgments than afterward, despite GATT’s lack of enforcement power.Yet, why would states plea-bargain if they know they can spurn contrary rulings? To find out, the article develops an incomplete information model of trade bargaining with the option of adjudication. The plaintiff has greater resolve prior to a ruling, believing that the defendant might be compelled to concede to an adverse judgment—even if that belief later proves false. Surprisingly, this resolve induces more generous settlements even from defendants who intend not to comply with any ruling. After a ruling, however, this anticipatory effect is irrelevant: adjudication works best when threatened but not realized. The prospect of adjudication thus conditions the behavior of states even when enforcement is not forthcoming but not through mechanisms identified by previous studies.

In a database of 298 GATT disputes, Reinhardt found this:

Before an adjudication panel formed, the full compliance rate of defendants was 30.4% (38 out of 125 cases).

After a panel formed but before it ruled, the full compliance rates increased to 63.3% (19 out of 30 cases).

After a ruling against the defendant, the compliance rate dropped to 41.8% (38 out of 91 cases).

A similar pattern emerged with partial compliance rates.

There seem to be two purposes to GATT’s court system. First, to clarify the parties’ legal standing, even if it does not enforce the ruling. Second, states use the courts to bring attention to violations so that other states can indepedently enforce the treaty through retaliatory tariffs.

It appears to me that some nation-states may be bluffing, but changed their minds when the others threatened with retaliatory tariffs and adjudication courts. On the other hand, if a state waited through the entire ruling, it probably had no intention to obeying in the first place.

“GATT has no teeth – the enforcement, indeed even the legal formality of its rulings is up to the disputant states themselves.”

Even GATT and the WTO are basically non-binding.