It makes sense that vaccines need more careful assessment before rollout than treatment pharmaceuticals. When the medical infrastructure has reached triage grade saturation, even coin toss treatment candidates could become rational if they can run in parallel to conventional treatment. Exposing the still healthy to a vaccine is an entirely different risk calculation.

A reasonably opportunity to throw money at the problem would be to fund speculative production ramp-up for likely candidates ahead of approval, this should not be left to the invisible hand alone.

They start making the flu vaccine the year before. It takes six to nine months, and lots of chicken eggs.

Here's a Planet Money rerun that talks about how they make the regular flu vaccine: https://www.npr.org/2020/03/06/812943907/episode-977-wheres-...

Transcript: https://www.npr.org/transcripts/812943907

(Warning: very layperson understanding below)

There are many variants of the flu. Most of them are well-enough understood to vaccinate against, and most of those have an already-tested-and-approved vaccine. The yearly flu vaccine is a combination of those vaccines for the strains the public health experts think are going to be an issue that season.

Sometimes they get it right. Sometimes they don't, and we have a bad flu season (see also: this year).