I used to work for Orbital as a NASA contractor. I never dealt with the business end of things, but the margins were small and very stable so long as there was work to be done (i.e. a giant (sub)contract on some huge NASA project). It wasn't exactly a pressure cooker of a work environment. The government stuff was a very different atmosphere than the commercial side.

5.5% profit, if reliably realized, is sustainable. It's comparable to a conservative investment return. That said, most businesses strive for more than that.

"We're not losing money AND we get to shoot things into space" sounds like a winning business model to me.

The R&D is baked into the contract itself.

It shouldn't be.

I mean, things go wrong. It's the one universal truth of any engineering process. Finding out why the thing that went wrong went wrong, and making sure that it doesn't go wrong in the future is just part of the job.

I work on an on-call rotation --- it's not as exciting as rocket science, which is both unfortunate and fortunate --- and yes, things go wrong; and one of the biggest rules of our process for when things do go wrong is 'Is it the end of your shift? Then hand off and go home. NO THIS IS NOT NEGOTIABLE.'

If the process itself is unable to cope smoothly with something going wrong, then the process has gone wrong. (And someone should find out why it went wrong, and make sure that it doesn't go wrong in the future.)

Indeed; I have an uncle who was on one of the 3 NASA Space Shuttle safety teams when the Challenger failed, although fortunately not the one who's remit included the SRBs. It triggered a stress related disability that seems to be a partly genetic thing, the same thing happened to me much more slowly a couple of decades later.