The amount of time it takes to get a Ph.D. in the humanities varies, but 4 years is very rare. (It's much more common for STEM/social sciences.) Unless you go to one of those dinky oxbridge places, of course.
Getting a job is about going to a top program and publishing... and about what area you study, and who's hiring in that area, and whether they happen to hire during the window when you're competitive for jobs (done/close enough to done, but not done so long ago you look stale), and whether they have any money to hire a tenure-track position as opposed to just adjuncts. And whether you're in a larger market with lots of universities is absolutely irrelevant - more universities = more people looking for jobs. Even if you have a Ph.D. from Yale with a published book, you may end up taking a job in rural Montana. You absolutely can't say you'll get a job in a larger market just because it's a larger market. And you absolutely can't say someone who's at a top program and publishes will get a TT job the way you can say someone who goes to Harvard and finishes in the top 10% will get a job. You just can't. They have a better shot than people at worse programs, sure, but they're still going to have to go where the jobs are, if they get one, and there just might not be one.
In some fields, there are decent jobs for Ph.D.s outside the academy. In many (especially the humanities), there are not. Teaching in private schools is an option, but private schools are rarely hurting for candidates to teach in the humanities. Teaching in public schools isn't that straightforward. The union issue is correct. And a friend of mine who has a Ph.D. in history teaches at a community college. Her degree is in European history, but she teaches all kinds of history (because it's a CC), including Intro to US. She looked into getting teacher certification so she could teach in a public high school (which includes charter schools), and the state would require her to take the exact same Intro to US class that she teaches (along with other stuff). Plus, even K-12 teaching jobs aren't exactly thick on the ground in a lot of places. And you didn't need the PhD to get the job in the first place.
Ditto most other options for Ph.D.s outside of STEM/some applied social sciences - you can get work (Ph.D.s actually have a fairly high rate of employment), usually, eventually - but it's going to be work you didn't need a Ph.D. to get. And that's after you convince employers that you're not going to run back to an academic job at the first opportunity - so in that respect, it's exactly like JD hiring for non-JD required jobs.
I also know lots of people who have debt coming out of Ph.D. programs (these are people who went to programs with funding). It is, admittedly, nothing like law school debt, but still, the opportunity costs of grad school are longer, and even the best outcome, the salaries are low.
you'rethemannowdawg, sorry, I get what you're saying now about risk - I didn't grok that earlier. Yes, I would say today, most Ph.D. students are willing to take on more risk than JD students are (although think of all the people who come here and say they're going to work really really hard and end up in the top 10% at their TT and they'll do great or they'll just transfer to a better school... that mindset affects Ph.D. students, too, and there's still some bad advice out there).
Again, this is all directed specifically at Ph.D.s in the humanities - I can't speak to other fields, many of which do have much better job opportunities. (A friend of mine is a prof of math education and they can't get enough candidates to fill positions in that field - in part because grads can get really good jobs in school systems/state departments of ed, that kind of thing.)