Earlier quoted context omitted.
"That the current salaries of recent college graduates represent their probable average output over their lifetime." It doesn't represent that at all. It is saying that the value due to education can be roughly approximated by the starting salary. The gains from there are then more based on experience/work, not the education. Of course, you ignore the money made while you would have been in school. You also ignore th…
And it [a college degree] is an advantage again if you need to switch careers. (Which is becoming very common.) But where's the educational correlation between a degree in a specific discipline and a change to what is presumably a different discipline? If I have a degree in Literature, I'm an editor at a publishing firm, and I get fired and then get a job in, say, the biological sciences, how is my degree intrinsical…
From the point of view of the employer, a degree indicates that the possessor has, among other things, a certain level of intelligence, literacy (not a small deal in a country where over half the country is functionally illiterate!), experience in learning, general background, work ethic, and socio-economic status. It is therefore a somewhat useful filter to apply for many professions.
Of course most of those traits have nothing to do with what you actually learned in university. And there certainly are plenty of successful people with all of those traits who did not bother with university. So even though I believe that for most people, education is worthwhile to that person, I'm far less certain that society as a whole benefits much from all of the money spent on higher education. (And I become even less certain when I look at how wasteful the process has become. Costs have been going up faster than inflation for decades, with no apparent reason than that everyone in the process knows that they are free to raise costs by that much, and it is always easier to spend more money than it is to spend less.)