Well, I could be just wrong:
H/T to Chris Blattman for re-tweeting this figure.
Seems as if getting a PhD is the right idea--low unemployment and high wages. The question would be: how do folks who do not finish compare? If you just end up with an MA, you are still better off, but not all folks who fail to finish their PhD program leave with an MA. They often leave with nothing but a few years of lost time.
Still, students who seek this life may not be making as bad bets as I thought. However, the academic job market varies widely among fields. Getting a PhD in English or Philosophy is unlikely to lead to heaps of wages or certain employment, whereas a PhD in engineering or bio-something is likely to lead to a very good paying position. Poli sci? Somewhere in between. Not every student finishes, not every student who finishes gets a good job. Even students who get good jobs have to face big problems as one usually has to surrender control over where one lives (six years in Lubbock!) and is especially problematic if one marries another academic.
Anyhow, I will dial done my anti-grad school rhetoric, just in time for my new job where I will be teaching only graduate students.
Update: Lukas Neville, a former student, provided a link to yet more math. Most job growth will not be for those with PhDs--only 3% of the new jobs will be for those with too much learnin'. Bad news. The good news is that there will be more jobs for PhD's in 2020 than today: 1.7 million. So, it depends on whether you see the glass as half full or half empty. More likely, it depends on whether you see yourself as employed or not down the road. 21 year olds will be optimistic--my guess.