I'm curious about what exactly that figure is too. It doesn't match my expectations from studying labor economics in grad school.

Typically, you restrict it to "working age men" and omit those in school. My recollection is that you see a very flat line, until quite recently when it starts declining.