Per a Pew study Brookings cites (https://www.brookings.edu/blog/social-mobility-memos/2014/02...):
> children born in the lowest income quintile who do not earn a four-year degree are four times as likely to wind up in the bottom (47%) as those who earn a four-year degree (10%).
So while upper middle class people are still the ones with the best chance at getting admitted, paying for, and finishing college degrees, lower class people who persist and do the same really are far more likely to join the middle or upper middle class from doing the same. Though it's far from guaranteed, especially since the relative value of a degree does depend on which degree, and sometimes school prestige as well, which is definitely harder for the lower class to prioritize (costs, scholarship availability, and nearness to family if they need to provide income support or other support to them tend to matter more in my experience).
The observation you mention is accurate, but I just wanted to provide additional context to readers who may not realize the life changing effect a degree can still have for poor people. This is as someone who experienced a comfortable middle class upbringing only because of the combination of a state school in the rural south that gave merit scholarships in the 80s to locals, and my mom's efforts to graduate with a degree that would be worth money.
That said, the idea of pursuing liberal arts in college is still incredibly foreign to me. Being able to support oneself after getting that sort of degree is definitely something I can see as mostly a proxy for upper middle class upbringing, and likely feeds into the rising trends see today where the payoff for the lower class is not hugely significant or comparable to well connected peers, even upon degree completion.
While it is true that the school system acts as a filter that removes those who have almost no chance of upward mobility (those with crippling disabilities, for example) from completing higher and higher levels of schooling, we need to be careful to not reverse that observation. Attaining a degree is not going to undo the disability that limits one's economic growth.
With the great push for everyone to have a post-secondary education we've witnessed over the last decade or two, virtually everyone who is capable of attaining a post-secondary education has done so. Those who are not completing those levels of schooling now are those who had no chance in the first place, and they struggle equally in the rest of their life for the same reason they did not succeed in school.
That said, if you were born with what it takes to be top of the class in Harvard, but chose not to pursue that avenue of life, chances are you still have every bit of upward mobility potential as someone who did graduate from Harvard.
Well, minus the Harvard contacts and network, which is a huge part of the success factor of Harvard grads.
So again, without a college degree, there's a 73% of people born poor, will remain lower or low-middle class. Only 3% will make it into the top quintile (vs 9% with degrees) and 8% into the 4th (vs 17% with degrees). So if we define only as 3rd quintile up as middle and up (which is questionable in itself given the shrinking middle class) a degree lands a lower-class person 52% chance of genuine upwards mobility, compared to 27% without. And it nearly guarantees that they at least won't remain in the very worst stratum of poverty. 52% for real mobility may not be great odds, but neither is 27% and if it was my future on the line, I definitely would not want to be staking it on 27%.
I think this is a critical observation that lots of people--including some economists--have missed.