> Every property constructed dataset I have seen shows incomes have increased greatly.Sure, but I was talking about Canada, the "most educated nation" according to the OECD, which, importantly, crossed the 50% attainment threshold a long time ago. The USA only did so recently.
In the interest of fairness, it might be too early to see the big jump we're told to expect in the US data. I am a little surprised you reached for it for that reason. If we're not too early, then I am still surprised you reached for said data. The data shows a fairly consistent growth trend over the span of decades, with no instantaneous growth of any substantial amount to account for the point where the degree was obtained. Perhaps you mistakenly linked to the wrong datasets?
That said, it is telling that US median income has risen – quite significantly, really – even while the median income did not possess a degree. Which, frankly, echoes that the relationship trying to be painted earlier isn't actually there. While we can try to be fair, you and I both know there won't be a big jump in the median income to account for the US median income finally attaining a degree. The idea that a degree would provide increased earnings doesn't even make sense.
What does exist is college filtering mechanisms (entrance requirements, academic rigour, etc.), which have traditionally kept out those who aren't destined to earn much in life, but that, again, doesn't mean attaining a college degree is a cure for what ails them. Indeed, in a given cohort, it is statistically likely that top performers who succeed in everything they do in life will also succeed in college and in the economy in general. But what does that tell us? Don't be born into life problems? Not exactly actionable.