Leading off the article with Yglesias shows the guy has little idea what he’s proposing to discuss. Imports can reduce GDP because the import is imported and not domestically produced. The formula identifies specifically: that which is consumed domestically but not produced domestically is not part of domestic production. There is no inconsistency here at all with revised trade policy increasing GDP. It should be tot…
The moon base example I think makes the argument very clearly. If you have an economy which produces nothing then it has a GDP of 0. If the increase imports for whatever reason, their GDP is still 0, which means that imports doesn't subtract from GDP, otherwise their GDP would be negative which is nonsensical.
But all this is sort of beside the point because arguments from accounting identities are almost always nonsense.