even if we take the number presented as fact (i'm not sure we should), the articles claim is that:
> "Foreign exporters absorb only about 4% of the tariff burden-the remaining 96% is passed through to US buyers."
so yeah, the exporter does pay some burden. it's not binary. indeed, tariff exports can be designed in a way to dial either direction. certainly, we could dial foreign exporters burden to 0% – and we could dial it back up to 4% (where we're currently at). but, 4% likely isn't a hard ceiling, either. Of course, the 4% number is an aggregate, not the blanket value across indidual goods (or services).
finally, the effect of tariffs is argued to be wealth transfer to the US Treasury. this is worth thinking harder about. but also, exports may change from whom goods are purchased. thus, it's a diplomatic policy, as well.