Paul Romer
copied to clipboard


The more I look into this, the more cases I see where the short-run scarcities and bottlenecks are the result of an FDA approval process for tests that is very specific about inputs — use these specific RNA extraction kits sold by these firms; these specific swabs…

As we scale up and hit a capacity constraint, the academic researchers seem to quickly identify workarounds, but it seems to take a long time for these to get throug the approval process - even for an Emergency Use Authorization.

So I keep having conversations where people close the existing testing system tell me that e.g. “swabs are a huge bottleneck” and then conversations with the researchers where they say, “you don’t need swabs, saliva works just fine.”

So we should distinguish two questions about scaling:

a) How hard to scale the FDA approved test, with an EUA amendment for each change?

b) How fast could the veterans from human genome project go if all they had to do was run a lot of tests that work?