Or they are just underfunded relative to the amount of work they have to do.
http://www.npr.org/sections/health-shots/2016/09/01/49223579...The Slate Star Codex article fails to make the case that lowering the general regulatory thresholds for medical devices or drugs improves societal outcomes. It makes a compelling, if possibly flawed, case for the EpiPen, but assumes generalizability rather than showing it.
We know the costs of not having something like the FDA. All one has to do is look at the supplements industry and imagine if real drugs were like that.
Oh, you thought you were getting what was on the label? Silly you.
https://healthyfoodusa.com/fda-finds-majority-of-herbal-supp...
You thought you'd get something that would help? Silly you.
http://www.health.harvard.edu/staying-healthy/dietary-supple...
You thought you'd get something that didn't have major unknown side effects? Silly you.
http://emedicine.medscape.com/article/817427-overview
Edit:
Also ignored is how the FDA is a major incentive to create something new that both helps with the problem and doesn't generate an unreasonable amount of other ones. If the barriers to market entry are removed, you are putting the informational cost on the consumer. If it's cheaper to fool them than to do the due diligence in development, that's what will be done. And it's pretty clear from the supplement industry that it is cheaper.
Scott Alexander likes to make the future costs of a lack of new medications argument a lot. He has to account for the costs of a lack of new effective medications as well.
Edit 2:
Scott Alexander doesn't try to make the case the FDA is corrupt. He talks about how Mylan uses the court system and lobbies congress. And his point about costs is undercut by a later post he made about coming up with the chairs statistic: http://slatestarcodex.com/2016/08/31/terrorists-vs-chairs-an...