I view sharing my hard-earned wisdom on the world with students as part my responsibility as an educator, and every semester I encounter students with ill-formed plans to attend graduate school in the social sciences, humanities, or law. When I ask them “why graduate study?” very few respond coherently, and then I start laying out the facts.
I send these students a list of articles and op-ed columns that explain why people should not get sucked into something that has such a high opportunity cost. This Al Jazeera editorial by Sarah Kendzior on the current state of anthropology is particularly illustrative. She writes:
“I received my PhD, but I still do not know what to do with it. I struggle with the closed off nature of academic work, which I think should be accessible to everyone, but most of all I struggle with the limited opportunities in academia for Americans like me, people for whom education was once a path out of poverty, and not a way into it.”
I recently stumbled upon the motherlode of such advice, which I will add to my list of what I give students:
100 Reasons Not To Go To Grad School
There are a few reasons why we, as academics, continue to encourage our impressionable students to try to join us, even though about 70 percent of faculty positions (at least in the USA) are now part-time, with low salaries, no health insurance or retirement benefits, and no job security. First, hearing that some of our brightest students want to be just like us assuages our egos and reassures us that we, at least, made the right decisions about our lives. The urge to imitate is the most effective form of flattery. Or something like that.
Second, both sides engage in biased thinking. Students see a tenured professor, remember certain films depicting the quirky but beloved professor strolling about an idyllic campus, and conclude “this person made it, there’s a [good] chance I will too, because I’m special!” Meanwhile the faculty mindset is “I made it, maybe you can too!” As cognitive psychologists and financial markets tell us, people are remarkably good at ignoring statistical probabilities. Few people say “the real chance of you becoming a professor is so small that it is effectively zero.”
Third, many of us continue to assume students are younger replicas of ourselves, with the same qualities and interests that we had when we were in college. 
We still use the tried but not necessarily true instructional methods that we encountered when we were in college, like the research paper, despite lack of evidence on their effectiveness. While problem-based learning, entrepreneurial training, and project design have become more popular, especially in engineering, natural science, and medical training programs, many curricula still emphasize understanding the abstract over solving specific, real-world problems. And when one of our top students comes to us to ask for a recommendation letter for his or her graduate school applications, we see that student as validation of our hard work and methods while at the same time we lament the majority of students who don’t meet or have an interest in our definition of success.
Nevertheless, there is something intrinsically valuable about enjoying solving math problems, regardless of our illusion of its usefulness, even though there may not be any immediate usefulness of that problem in the real world. When I was a student in the late 60’s, the word we threw around to condemn academia was its lack of “relevance.” Maybe all that fumbling for academic career, (which I recall in a positive way), despite the poor odds of succeeding, taught me about myself, my strengths and weaknesses, where my desires can lead me astray, and gave me a unique set of tools to apply to real world problems. Maybe your right, and many who strive to emulate professors have little chance of succeeding now, but striving to be successful can be like solving math problems for the fun of it.