Yes, the inmates are at least partly responsible for the rules of the asylum. There's an answer to a question about the loneliness in pursuing mathematics on m

Yes, the inmates are at least partly responsible for the rules of the asylum. There's an answer to a question about the loneliness in pursuing mathematics on mathoverflow.net that says, paraphrasing, that one should only do mathematics if they can't do anything else [1]. That is, if it is such a burning desire that you can't stop. As long as salaries allow one to live comfortably there will be a supply of people wanting academic jobs. The lifestyle is compelling for a lot of people. I've been led to believe that even after tenure one still needs to be active and that means to publish. There is a lot of variation on salaries within a given department even at the full professor level. The more active/prestigious you are the more salary you make. Those who are active in research and with graduate students don't have to teach as many low level courses or be on as many committees. At least this is true of mathematics departments at large research institutions that I'm familiar with. I'm not a member of a large research institution but this is what I've been told by members of such institutions. My sample size is too small for me to generalize but it sufficiently large for me to hypothesize. [1] https://mathoverflow.net/a/110063 https://mathoverflow.net/a/110063