At The Atlantic, Dana Goldstein interviewed Suzanne Mettler (of submerged state fame) on her new book about higher education. Here is one of the questions and its answer:

You portray the four-year college degree as a transformative tool in battling inequality. What do you think of the counterargument that our national debate focuses too much on education as a driver of inequality? There are so many reasons why we need to increase our percentage of college graduates. We need more people who are highly skilled to try and create the kind of innovation and creativity that leads to greater economic development in all sorts of ways. And then it’ll help to mitigate social inequality. If we have more highly educated people, it will create more civic engagement and political engagement and leadership for American society.

This answer is very confusing. An argument for how higher education will reduce inequality should go like this:

1. Increase the percentage of college graduates.
2. X results from (1).
3. Inequality reduction results from (2).

Filling in the X -- or what we might call the "link" -- is where all of the action needs to be, though so many who talk about inequality reduction fail to do so with much clarity. Here, a friendly reading of Mettler reveals two proposed links. The first link is that increasing the percentage of college graduates will result in more highly skilled people, which will result in more innovation and creativity, which will result in more economic development (I assume growth), which will result in inequality reduction. It's not at all clear how this link works. The story works well enough all the way until the very end where "more economic development" (which I read to mean growth) is supposed to then lead to inequality reduction. The only way growth can cause inequality to go down is if the distribution of the gains from that growth lift the bottom and middle faster than the top. There is no reason to think that will happen, certainly not if our distributive institutions remain as they are. Indeed, we've just gone through a half century of solid growth where that has not happened. The second link is that increasing the percentage of college graduates will result in more highly educated people, which result in more 1) civic engagement, 2) political engagement and 3) leadership for American society, which will result in reduced inequality. This story is so vague that I don't know what to make of it. It almost sounds like a politician's speech. It's not clear that having more highly educated people will increase (1), (2), or (3). And then from there, it's not clear how (1), (2), or (3) is meant to result in reduced inequality. My best shot at rehabilitating this particular story is to read (1), (2), and (3) to actually mean "more politically engaged people pushing leftist politics." That story would then facially check out, though the empirics of it are still uncertain. What's surprising to me about the higher education and inequality stuff is just how weak the arguments for it actually are. The idea that increasing college completion will reduce inequality is so pervasive that, for a long time, I worried that I was missing something extremely obvious and that one day I'd find myself very embarrassed because of it. But as time has gone on, I have become increasingly convinced that this is just one of those bits of cultural ideology that people repeat because they hear it said so often without anybody ever contesting it. When I read Lane Kenworthy's latest book "Social Democratic America," which spends a great deal of time on inequality reduction, I was very curious to see how he dealt with the question of higher education. He's as sharp as they come on the empirics of what we know about inequality reduction. So I thought, if anyone was going to explain how higher education links to inequality reduction, it was going to be Kenworthy. Here is what he had to say:

Some feel it makes no sense to try to increase college attendance and completion. After all, there is a limited supply of high-skill jobs, so some graduates will end up in jobs that don't require anything near college-level skills. Yet if our aim is to maximize capabilities, including the ability to make informed preferences, we must help more Americans from low-income families into and through college. In addition to providing a vocational skill and a valuable job-market credential, a college education can aid in the development of general skills, such as complex reasoning, critical thinking, and written and verbal communication.

As you can see, Kenworthy concedes away the inequality reduction point as essentially bogus, but then provides a liberal citizenry justification for expanding college completion. That's probably the only way to go really. The inequality argument has no legs.