Education polarization is only growing

And it's making everyone mad all the time

A Harvard Crimson survey of the incoming Harvard Class of 2025 revealed that 87% of the class voted for Joe Biden, compared to 6.7% for Howie Hawkins and 6.3% for Donald Trump.

It is well-known that young college graduates skew left. What’s interesting about the Harvard data is that it suggests the skew is sharper at more elite schools. And given that this is a poll of people who hadn’t even started college yet, it should help ease concerns that the progressive politics of young grads reflects some kind of professorial brainwashing. Instead, Harvard (presumably not deliberately) is selecting students whose political views are way to the left of the U.S. national average.

The Yale Daily News did a similar survey in 2022 and found that “Nearly three-fourths of respondents identify as ‘very liberal’ or ‘somewhat liberal’” versus 9% conservative and 2% very conservative.

Dartmouth’s 2019 survey showed a class that’s not quite as left-wing as Harvard or Yale but still far to the left of a national average.

I don’t think these numbers are incredibly surprising; people are aware that elite college campuses are left-wing places. But it is different from how American politics used to operate. And it has potentially important implications for how our political and social institutions operate.

The Ivy League used to be right-wing

Contrast today’s results to a Crimson survey from 1932 that showed Herbert Hoover beating FDR by a 2-1 margin, even as Hoover lost in a landslide.

To an extent, that’s easy to understand. The one-liner on Roosevelt was that he was a “traitor to his class.” And his class was, literally, the kind of people who attended Harvard. Meanwhile, the majority of the electorate at that point didn’t have a high school diploma.

But this dynamic has been changing for a long time. By 1951, there’s concern about communist professors “who cloak themselves in ‘academic freedom’” and William F. Buckley is publishing “God and Man at Yale.”

Note, though, that at this point the conservative critique of academia is that there is too much free speech on campus, with leftist faculty given too much running room to criticize the established order of things.

If we look at precinct data for 20 top universities today, we see that we are dealing with an overwhelmingly progressive milieu.

And again, this appears to simply be a feature of who has the highest SAT scores rather than a dramatic transformation taking place on campus.

This is similar to what we see in the legal field. Lawyers are more liberal than the general public, and that’s especially true at the most prestigious law firms and among law professors. There is a whole contested literature on the relationship between IQ scores and political ideology, and it features a bunch of complications and nuances. But it does seem to be true that the people who are best at school are on the left politically.

Of course, just because the students start out further left than average doesn’t mean nothing happens on campus.

Time in school pushes people to the left

A new paper by Milos Brocic and Andrew Miles tracked a large sample of people through four waves of the National Study of Youth and Religion to see how passage through the education system changes political attitudes.

They also note along the way that the conservative critique of academia keeps changing over the years. From “too much academic freedom” in the 1950s to a sense in the 1980s that universities promoted “moral relativism” rather than clear thinking to the critique today that the academic climate is too morally fastidious and amounts to a “new puritanism.” What they find, basically, is that some version of this latest thesis has support in the data — higher education pushes people to the left and instills them with more moral absolutism, and the effects are stronger for graduate students:

Our results indicate that higher education liberalizes moral concerns for most students, but it also departs from the standard liberal profile by promoting moral absolutism rather than relativism. These effects are strongest for individuals majoring in the humanities, arts, or social sciences, and for students pursuing graduate studies.

I think conservatives often want to look down on the humanities and will find something reassuring about the fact that the effects here are attenuated among students of natural science.

But check out this National Academy of Sciences report from January 2020 looking at campaign contributions by college professors. The authors found an 8.5:1 ratio in favor of Democrats. There’s a fair amount of variation by discipline, but even in the most right-wing discipline (economics), there’s a 3:1 ratio. Something like chemistry is right-wing compared to sociology, but incredibly left-wing compared to the population average.

The same NAS report also finds a sharper faculty skew at institutions with higher U.S. News and World Report rankings — similar to how Harvard’s first-year class is more left-wing than the general set of young college students.

Intriguingly, the NAS report also looked at donations and found a much sharper overall skew of 95:1, with the least-left discipline, again economics, coming out as having a 17-to-one ratio of Democratic to Republican donors.

I don’t think it’s totally clear how to interpret the gap between the donor data and the registration data. It could be that highly politically committed faculty are just more left-leaning than faculty writ large. It could also be the case that party registration is a lagging indicator, and lots of professors who voted for George W. Bush or John McCain are now de facto Democrats but haven’t switched their registrations.

It’s going to be hard for society to function like this

I usually like to offer solutions, but my main thought about all this is that having society sharply polarized around occupational categories and educational attainment is going to make it very difficult for us to function effectively as a country.

Jay Varma has a good piece in the Atlantic titled, “Not Every Question Has a Scientific Answer: The toughest COVID-19 policy questions are matters for politicians—not health experts—to decide.” That’s completely correct. But it’s not unique to Covid-19. Most policy questions have a significant technical dimension but also feature ineradicable elements of political judgment about tradeoffs, priorities, and risk tolerances. That’s true whether you’re talking about a pandemic, macroeconomic policy, climate change, how to build a school system, or basically anything else under the sun.

One downstream consequence of education polarization is that it’s getting increasingly difficult for conservative politicians to equip themselves with adequate technical expertise to make sound policy. You end up both drawing from a shallow bench, but also adopting a kind of anti-credentialism as your worldview where nobody is actually interested in kicking the tires of ideas to see if they make sense.

Examples of this were easy to find back in the George W. Bush administration when well-regarded social scientist John DiIulio went up to head a White House office on faith-based initiatives, only to walk away denouncing the Bush team as a bunch of “Mayberry Machiavellis.” But the Trump team made George W. Bush look like history’s greatest technocrat.

Conversely, if all the people with degrees are on the same side of certain big moral and political questions, it’s going to be very hard for them to draw the line between actual expert knowledge and beliefs they happen to hold. Perhaps the greatest recent example of this was tons and tons of public health academics joining a sign-on letter in favor of the George Floyd protests. Note they didn’t just say something like “outdoor activity is relatively safe and we encourage everyone to do things outdoors that are important to them (church services, social justice protests, gatherings with family).” Instead, they decided to endorse the agenda of the protests:

However, as public health advocates, we do not condemn these gatherings as risky for COVID-19 transmission. We support them as vital to the national public health and to the threatened health specifically of Black people in the United States. We can show that support by facilitating safest protesting practices without detracting from demonstrators' ability to gather and demand change. This should not be confused with a permissive stance on all gatherings, particularly protests against stay-home orders. Those actions not only oppose public health interventions, but are also rooted in white nationalism and run contrary to respect for Black lives. Protests against systemic racism, which fosters the disproportionate burden of COVID-19 on Black communities and also perpetuates police violence, must be supported.

This surely did approximately nothing to actually advance anti-racism causes (who cares what public health academics have to say about this?) but a great deal to burn the credibility of public health academics in the eyes of a huge swathe of the population.

There can be a healthy relationship between expertise and politics, but it’s extremely difficult for that relationship to exist when education is so strongly correlated with partisanship.

Both sides should try to do something about this

As has been explored many times on this blog, stark education polarization is really bad for Democrats’ prospects of winning a Senate majority. I won’t belabor the point here, but mathematically, Democrats cannot govern in the long term without increasing their appeal to less-educated voters.

Instead, I’m going to concern-troll conservatives.

Conservatives are clearly very bothered (seemingly genuinely so) by the low esteem in which they are generally held by cultural elites. They complain about it constantly. At the same time, conservatives are not as policy obsessed as liberals are. Very few conservatives pronounced Donald Trump a “failed president” despite his very meager legislative achievements during 2017-18, and even fewer experienced his signing of left-coded bipartisan legislation in 2019-20 as a huge betrayal.

In other words, the big edge that education polarization gives Republicans in the Senate does not satisfy what they truly crave in life, which is something like respect or a desire to have their putative status as the great defenders of women’s sports taken seriously as a good-faith concern.

Well, just as the political system skews in favor of the old, the non-college, and the rural, the cultural system skews in favor of the young educated professionals and urban tastemakers. If you want to see your values reflected in Marvel movies and beer ads, then you need to try to find topics on which to align yourself with those younger metropolitan audiences. Demagoguery about refugee resettlement, for example, does bring some electoral benefits. But is the juice really worth the squeeze if it codes you as the party of inhumane sociopaths in the eyes of the youngest and most cosmopolitan generation? There used to be an elite consensus around such things. Republicans decided to break with it, and on one level it has worked, but on another level, I feel like it really hasn’t delivered what they actually want.

The best example of this is the fight over voting rights issues. The actual, concrete electoral gains to Republicans from various measures to make it harder to vote are imperceptibly small, if they exist at all. So what is the point of pushing measures that make you look like an authoritarian menace? And would it really be so hard to throw urban libs a bone in the form of some concern for climate change?

For some people, of course, the current system works great. Culture wars and skewed maps help Republicans win elections, after which they cut taxes for rich people and multinational corporations while doing nothing to satisfy their base’s resentments — resentments that fuel the fire for the next campaign. But there is only so much cynicism to go around, and I think the steady growth in education polarization is mostly serving to frustrate both sides of the never-ending war.