It may be significantly worse in research:

1. PIs are under incredible pressure themselves

2. PIs are as a rule not formally trained in management, nor have they tended to "apprentice" in middle-management positions

3. Research is heavily reputation-based, which creates huge power disparities between employers and employees: lab employees need references from their employers, who can withhold them for capricious reasons

4. All of these factors also applied to the PI themselves, when they were starting their career, so it's normed; PIs may see it as a form of "dues-paying"

These dynamics are not in fact present in most ordinary market jobs. The jobs where the dynamics are at play --- say, as an assistant to a producer at a media company --- are notorious for it.