If it is true that a college degree can pay for itself in a few years after graduation, then the amount of demand for college degrees would rise (it's a no brainer to do it right?).

But the production of college degrees is somewhat limited by the availability of positions, and is not infinite. Therefore, if demand rises but not the supply, the cost _must_ rise to match. Therefore, after some time, the cost of a college degree slowly grows to match the potential future value of having that degree (fuelled by free debt).

I think at some point in the last couple years, the cost of a college degree has, on average, grown to almost match the future value in earnings accounting for the debt repayment.

I only see this with people trying to check the “college education required” box, rather than trying to acquire a marketable skill. From my perspective, this was always the case, but there doesn’t seem to be any “cultural knowledge”. My naive assumption is that college recruiters/counselors are pushing people, who don’t have marketable interests, into debt, and the students don’t know any better, for whatever reason.

I agree that more people think that way than 10 years ago. However I still think it's a tiny portion of the college-age population.

The problem with that conclusion, even if true (I'm sure if you dig into the study you'll find all kinds of assumptions that you may or may not agree with) is that it's based on average earnings. And as we all know, everyone believes they are above-average, or their outcomes will be above-average. It's easy to read a study such as yours then dismiss it because it includes graduates with degrees that many people assume have little earning power.