While I agree with you main argument, I think that it’s easy to equivocate between schooling and education.

The linked study essentially does so in it’s very title.

I don't have access to the full article itself, but isn't it testing the difference between compulsory schooling and a bit more compulsory schooling, rather than some school or none? To be sure, education is wider than school, but I find it difficult to believe that schooling doesn't make a profound difference, even if slightly more schooling shows diminishing returns.

It is. My point was that “compulsory schooling versus a bit more compulsory schooling” is what they are measuring and not “less educated vs more educated”.

I too would expect a difference in outcomes between more and less compulsory schooling. But I would expect it from the non-education related impacts of it (I.e. de-facto childcare and keeping the kids/teens from engaging in as much long-term harmful activities)