I would think that the backdrop thing is mostly performative (look at me, I'm producing this under suboptimal conditions), but I am really surprised that these

I would think that the backdrop thing is mostly performative (look at me, I'm producing this under suboptimal conditions), but I am really surprised that these broadcasts aren't being produced asynchronously. Both audio and video would be much better if the person was doing a local recording of their side of the call, and afterwards both local recordings were stitched together instead of using the over-the-internet version. Doesn't work for live stuff, but most correspondent stuff is scheduled in advance and could accommodate this kind of thing.