It still seems relevant to compare within, say, 25-34. The ratio of college unemployment to high school unemployment is definitely higher than it has been previously.

It turns out that once you correct for this error, unemployment is actually rising faster for NON-degree workers - meaning that explanations in terms of AI replacing the college-educated workforce don't make much sense, and this looks a lot more like an across-the-board labor market slowdown.

And Kaine very incorrectly conflates one party rule with a party winning an election and having governing majority, which is how it works everywhere else.