If you consider the sheer number of hours I tend to spend on my dissertation work compared to the stipend I get, I'm paid criminally less than minimum wage. But you're not given the leeway to focus "exclusively" on a single research contribution in a focused manner in other higher-paying options.
Even if you're not taking out a loan in the monetary sense, you're taking out a loan from your long term earnings, one that has little chance of being repaid, to put your mental assets/skills to a non-remunerable task, and get a certification that you did so. The problem is that the option to continue doing this (tenure-track academic jobs) are limited and (naturally) highly contested.
However, you make a good point--there's an underlying and insidious opportunity cost that is often unknowingly sacrificed: that of atrophying skill sets. It's easy for a PhD to be a hugely insular experience, if you let it, and if you take the easy way out and don't stretch your engineering skills (speaking in terms of CS here, since that's what I know), you're in for a rude awakening if you determine that academia is not for you. If you're not careful, you'll get good at writing papers, but might actually get /worse/ at writing portable, readable, and maintainable code. And as brilliant as your papers may be, if you can't ship good code, you're going to have trouble in industry.
The good thing, again at least for CS students like me, is that the "fun and enriching" environment of academia means a lot of opportunity for starting companies, creating libraries/frameworks, working on side projects, and doing contract work, so there's no reason you have to atrophy. Which is something that, sadly, the visions of the tenure-track academic job are engineered to beat out of you.