Nate Silver discovers the educated poor

Silver’s synchronic stats are useful, even if the underlying social forces he identifies are not novel. It’s part of the pure quant’s weakness: they look at what their dataset tells them, and what it says hasn’t been said in the same terms as an earlier quantitative dataset told them, then it must be novel. And since the Democratic Socialists of America are more popular than they’ve ever been, any finding that correlates their current popularity with anything at all is, to the quant, very very new.

My biggest challenge with class-based analyses of anything in the Global North is that it feels like it could use a huge shot in the arm of Immanuel Wallerstein.

You can’t understand the social position of a New Yorker without understanding the existence of open pit mines in the eastern DRC where materials for producing tech components are extracted by non-unionized, informally employed workers with little recourse to social benefit programs, and living in a country with a national GDP per capita just a few hundred dollars above extreme poverty. With a few exceptions, most radically exploitative work producing value-surpluses captured in the Global North has been offshored.

Yet all these analysts are trying to reason from definitions of “blue collar” and “working class” established as though the US (or the UK, or the EU) were some economic autarky. A simple understanding of Global Value Chain Theory would make the political currents of the Global North a lot less mysterious. This includes the fact that, among people making more than the national poverty line but less than $60k/year, those who throw their lot in with a protectionist anti-poor platform tend to outnumber those who throw their lot in with a social justice platform, while the social-justice platform’s supporters tend to have advanced farther through the educational system.

I was a member of DSA in the 1970’s and ’80’s. Like quite a number of members, I was actively engaged in the labor movement, and with a specific labor union for whom I worked. There was a large contingent representing the more progressive wings of the AFL-CIO unions. That presence is (maybe?) still present, but obviously much diminshed.

JPL,

There are different levels of understanding. I may not understand at an emotional or instinctive level why somebody drove their car into a shop at 80 km/h in the sense that it is inconceivable for me do such a thing, but if it is explained to me, I can ‘understand’ the mechanistic chain of events of that person being angry, getting drunk, and then driving their car into a shop at 80 km/h.

The same applies here. As I ‘understand’ it, the salient factors appear to be:

First, plain and simple racism. As my mother once explained to me with reference to the time when we lived in South Africa (and I was much too young to understand), the dumbest and laziest white person, who in an egalitarian system would simply be very dumb and lazy, can still see themselves as a superior kind of human relative to the most accomplished black neighbour. And if the racism is institutionalised, as it was in South Africa at the time and as MAGA is attempting to have it again in the USA, that dumbest and laziest white person can secure a job that a much more qualified black person is excluded from, and because.

Second, this is supercharged by defensiveness about slavery. Again, I do not understand this at any emotional or personal level, because one of my grandfathers was a Nazi, yet here I am, finding it extremely easy to say that one of my grandfathers was a terrible person. But watching clips from the USA, it is very obvious that for some inexplicable reason a lot of white Southerners cannot face up to what their ancestors and their states have historically been. Despite the atrocities Europeans committed in colonialism, they are not confronted with that history in the way the USA are. At most, countries like Canada, New Zealand, and Australia are somewhat comparable.

Third, the conservative movement in the USA has become a cult. It is not hard to see the signs everywhere: inability to admit that the leader or the movement’s core ideas are flawed even if the member is being directly harmed themselves, statements such as “I am confused by I cannot discuss this with my liberal family, because they are obviously biased”, and patent falsehoods and incoherent statements used as markers of fealty. I do not really understand why there are now cults everywhere from politics to ‘cryptocurrency’ to Musk fandom, especially in the USA. The lack of a decent welfare/social security system is a possible factor, as is the greater religiosity of the USA, which in turn may also result form the lack of a decent welfare system. But at a purely mechanistic level, being part of a cult makes one vote as a block.

Fourth, the two-party system and first past the post election system do not help either. The entire idea of voters ‘registering’ as Democrat, Independent, or Republican is bizarre from the perspective of every other country on the planet, not least because elections are meant to be confidential, but the vast majority of US citizens has been conditioned to be unable to think outside of those boxes (and for the most part, unable to think outside of D vs. R, and liberal vs. conservative).

I suspect that the 2nd factor which ASL observes has some, or much, to do with the fact that several years after the American Civil War ended, the Southern oligarchs were able to regain control of the former Confederacy, at least at the state and local levels. Hence, they and their minions were able to control most of the instruments for shaping public opinion among the non-elite White Southerners, effectively propagating the myth of the Noble Lost Cause.

The Nazis were not able to regain control of Germany after WW2. If they had been, perhaps they could have conditioned young Germans to tolerate, or even celebrate, the Nazi portion of their national heritage, as White Southerners (disclosure: I resemble that remark) have been conditioned to tolerate, or even celebrate, our Confederate heritage.

As a conceptual tool for analysing voting patterns, the term “working class” is too ambiguous to be useful. It can mean anything from “wage workers in labouring and manual trade occupations” (perhaps 20 per cent of the population) to “below median income” (50 per cent by definition) to “did not complete 4 years of college” (about 60 per cent) to “income not primarily derived from capital ownership” (95 per cent), with only partial overlap between these categories. All of these definitions are useful in some contexts, but in debates over voting patterns they form a hopeless jumble.

A definition of ‘working class’ which takes in 95% of the population (‘income not primarily derived from capital ownership’) is going (if 95% is the correct figure) to cover a heterogeneous group of people and is not going to explain much about their voting patterns (or anything else about them). But it also implicitly defines a category of 5% of people (if the figures are correct) made up of those whose income is primarily derived from capital ownership, who may be a much more homogeneous group and whose membership of that group may explain a lot about them, including their voting patterns.

Interesting observation thank you for posting. I was served malware at the first load of this page – a captcha that asked me to paste a command into a run dialog.

J-D @9:

A definition of ‘working class’ which takes in 95% of the population [] is going [] to cover a heterogeneous group of people and is not going to explain much about their voting patterns []. But it also implicitly defines a category of 5% of people [] made up of those whose income is primarily derived from capital ownership, who may be a much more homogeneous group and whose membership of that group may explain a lot about them, including their voting patterns.

I do believe we may have just summoned the ghost of the Eleatic Stranger from Plato’s Statesman (263c):

Stranger
I think it started when you were asked how the art of herding should be divided and said with great readiness that there were two kinds of living beings, the human race and a second one, a single class, comprising all the beasts.

Y. Socrates
True.

Stranger
And it was clear to me at the time that you removed a part and then thought that the remainder was one class.

Harvard alumni aren’t more numerous or poorer than in the past but iirc they switched from R to D in large numbers…

Iirc it’s well attested that education tends to make people more liberal and downward mobility tends to radicalise. Establishment “left” parties in US/Europe have become more liberal and less left-wing over time, which made them more appealing to educated professionals and less appealing to their traditional working class bases. Downwardly mobile grads were responsible for much of the noisy cultural politics of the last decade but much of this wasn’t really “left” but radical liberal afaics.

If education made people more left-wing I’d have expected societies to have become more left-wing as they became more educated but I don’t think this has happened.

Happy to look at a statistical argument but I’m not sure what is being referred to.

engels @15:

Perhaps the best articulation of the finding to which John Q is referring is Gethin, Martínez-Toledano, & Piketty 2021, whose findings, to the best of my awareness, have been amply reproduced.

Your best counterarguments would rest on two foundations, neither of which are the strength of the statistical relationship.

First, the length of the time series (going back only to the 1950s, so excluding the pre-WWII radical parties that drew working-class support and such events as the Paris Commune and the movements of 1848, not to mention Levellers and all that fun and funky history) and second, geography (almost exclusively Global North). John’s “nearly always” may best be understood as “during the Pax Americana,” and his “everywhere” as “Western democracies in the Global North.” See Gethin & Martínez-Toledano 2025 (PDF):

Unlike Western democracies, which have undergone many common changes, non-Western democracies primarily stand out by their large diversity of [. . .] cleavage structures.

Is there any evidence that Harvard alumni are tending Democratic? Don’t forget that graduates of elite institutions are only a tiny fraction of the educated “class”, and many many prominent right wingers/ fascists (including Vance and Trump) are in fact elite educated.

The media of course work hard to uphold the opposite impression that progressive politics is somehow related to elite education so they can portray the left as out of touch and the fascists as anti-elite. Remember how GW Bush was magically absolved from his Yale and Harvard background? He only needed to express disdain for education, and was immediately reborn as a folksy anti-elite outsider. And the trick works extremely well to this day.

Tm @19
Alumni of elite universities in the U.S. today generally vote Democratic by substantial margins (with perhaps some allowance for regional variation). I’d be v surprised if the extant research showed otherwise.

Btw the fact that some prominent right wingers went to elite schools is beside the point when it comes to aggregate voting behavior. For that matter some prominent Dem liberals (and people to the left of that) also went to elite schools. Bernie Sanders went to Univ of Chicago, an elite school by any definition. Cory Booker, iirc, went to Yale Law School. Jamie Raskin went to Harvard (undergrad and law school). Pete Buttigieg went to Harvard and Oxford. Obama went to Columbia (after transferring from Occidental College) and then Harvard Law School. Etc.

IMHO, it’s all Lyndon Johnson’s fault.

He desegregated the schools and thought the working class would come back to the Dems a generation later. He.Was.Wrong.

(I grew up on Beacon Hill, and would walk across said hill (we lived on the funkier north slope) to get to school (Prince School and then Boston Public Latin School), and in the days when “bussing” was a dirty dirty word, my fellow rich white kids were all being bussed out to private schools in the suburbs.)

“have the resources and affluence to avoid the negative outcomes of their own policy positions”

Oh, hey, that sounds like conservative small business owners and well-off retirees voting for cuts to education, cuts to the social safety net, cuts to social housing, or Brexit, thereby destroying the life plans, opportunities, and freedoms of those who are currently young adults or teens after enjoying them themselves when they were that age. The “white graduates of elite (meaning selective) universities”, on the other hand, are so small an electoral constituency as not to matter for practical purposes.

MisterMr,

Probably a combination, but potential cause 4 is, understandably, underappreciated, especially at the level of the individual voter. Whether you think it has become more difficult (I am not necessarily convinced that capital really will flee if taxes are raised but business remains profitable, in part thanks to how those taxes are then invested), it certainly is more difficult to get the parliament of one’s country to tax and regulate and enact environmental protections than it is to call somebody on social media ableist because they called somebody else’s argument insane or to insist that names for professions be gender neutral.

Whatever the merit of this or that in the second set of activities (and there are lots on which I agree!), the required investment at an individual level is effectively zero, all that is needed for a feeling of success is to build a network or bubble of a handful of other people to enforce the language restrictions together by banning others, and the pay-off is to feel virtuous. For the first set of activities, the individual investment starts with giving money to a political party or lobbying group and balloons from there while it turns out that actual success may require a few decades of struggle against other interest groups with much deeper pockets. No wonder, then, that a lot of young people despair of even voting despite being passionate about social justice in the narrowly circumscribed areas where they have a feeling of control.

Causal mechanisms here are complex and context-dependent (varying by time, place, party type, issues of the day, etc.) so anyone building a strategy to capture a given voting bloc will likely have to work case-by-case. That said, a few signposts are worth noting.

First, the “Brahmin left” isn’t one bloc. As Oesch’s occupational-class work suggests (“Redrawing the Class Map: Stratification and Institutions in Britain, Germany, Sweden and Switzerland,” Oesch), it splits into a stable, economically-left core of socio-cultural professionals (teachers, precarious academics, public-sector/healthcare workers, etc.) and a more contingent, economically-right-leaning periphery of technical/managerial professionals (engineers, IT, accountants, corporate managers, etc.) held together by shared cultural liberalism rather than shared economic interest, and open to fracturing when a party’s platform shifts sharply left on economics. Moving economically left would likely harden the first group’s support while risking the second.

On the working-class side, the picture is more encouraging for the left than the standard narrative suggests. Working-class support for left parties rises with greater emphasis on cultural issues, largely regardless of the position taken – and economically left positioning and emphasis independently help too, especially for social democratic parties specifically, which turn out to be far more sensitive to programmatic strategy than radical left parties (“Left Parties’ Strategies and Working-Class Vote in Contemporary Western Europe (2002–2020),” Trastulli). Consistent with this, the standard “Third Way” story (that social democrats lost workers by moving right on economics while moving left on culture, with the implied fix being to reverse both) doesn’t hold up well against the data (“Abandoned by the Working Class?,” Abou-Chadi et al.).

That said, this shouldn’t be overstated – a sub-literature on cross-pressured voters (economically left, culturally authoritarian) finds that there is a trade-off between economically left, culturally authoritarian and economically left, culturally libertarian voters within radical-left parties (“Class Voting for Radical-Left Parties in Western Europe: The Libertarian Versus Authoritarian Class Trade-Off,” Steiner et al.). But the opposite fix carries its own risk: shifting right on immigration specifically alienates highly educated voters and socio-cultural/self-employed professionals, so “move right to win workers back” risks losing socio-cultural/self-employed professionals without a clearly demonstrated working-class payoff (“Electoral Fortunes of Social Democratic Parties: Do Second Dimension Positions Matter?,” Abou-Chadi and Wagner).

Put together, it seems that the question for the left isn’t “economics or culture” but what combination of the two produces a stable, winning coalition. I suspect the answer involves both – much though that may disappoint our class-reductionist friends…

“The left”:
https://www.theguardian.com/politics/2026/aug/04/labour-members-view-welfare-health-economy-poll-survey

47% of [Labour] members surveyed by ThinkLabour believe the government is spending too much on welfare… 68% of members wanted Britain to spend more on defence… Three-quarters of Labour’s members said they had backed Keir Starmer in the 2020 party leadership contest…

What an embarrassment.