I mostly agree. He touches on it, but it's based on assumption and entirely superficial: "That doesn’t mean that you can’t build more of the machines, but I w

I mostly agree. He touches on it, but it's based on assumption and entirely superficial: "That doesn’t mean that you can’t build more of the machines, but I would assume that Pfizer, BioNTech, Moderna (and CureVac as well) have largely taken up the production capacity for that sort of expansion as well." The obvious question, then, is whether that production capacity could be increased, and whether we have a problem of throughput (which added resources might parallelize) or latency (and we will have all the devices we want, N months from now).