[psi75]

The problem, I think, stems from the fact that a doctoral advisor, because he influences your reputation in all of academia rather than just one company, is even more of a career SPOF than a corporate manager.

The good news is that your advisor is safe (tenured, or at least likely to make tenure) in most circumstances, so he's not going to fuck you over unless he's a truly terrible person, and 90 percent of people aren't, as opposed to the corporate world where the pressure of constantly watching your back turns that middle 80% into bad people as well. The bad news is that he does have this extortive power over you, if he wants to use it.

There's also a conflict of interest when it comes to delayed graduations. If the advisor's getting cheap or free grunt work, and delaying your graduation for another year can get him more of said work, then he has an incentive to do so. Of course, this can't be continued forever, because eventually he'll get a reputation for his students taking too long and not doing well... but in the short term, it is an option for him, and it is sometimes exercised.

[ModernMech]

> If the advisor's getting cheap or free grunt work

This is a huge misconception. Yes, as a grad student you aren’t paid well. That doesn’t mean you are cheap for a PI. They have to pay your tuition, as well as university overhead related to your stipend. You don’t see a lot of that money in your bank account, but it is leaving the PI’s research grant nonetheless.

The most productive students are those who are about 3 years in when they are done with classes. The least productive ones are those about to graduate; they typically have a serious case of “senioritis”, and are often busy making contacts in industry and planning a career. Keeping them around makes them less productive, not more.

Professors want to graduate them to free up resources and to bolster their promotion portfolio.