Peer review incentivizes the worst features of humanity: it affords peers in your field the ability to reject your paper because they want to take the idea for themselves and then publish it.
In the corporate university, the usual game is to get the patent approved first, and hold off on the publication until that happens. At least that's the case in applied technical science - the social science world seems more like a club where the involved players just rubber-stamp each other's non-reproducible work in order to up their publication count and pack their CV. The actual work has no value, so why would anyone bother stealing it?
If that were true we'd see more patent applications in, say, engineering, than journal papers in the same field. (Or at least around the same order of magnitude). Do the numbers check out?
My experience was: Most papers are not useful for patents. Corporate funding might require a pre-publication review for patent worthyness, and if it is there, it is applied for before publication.
Most researchers I knew had little interest in patents, so non-corporate funded papers got published without patent considerations. When it came up, it was a before-you-publish-think-about-it topic.
