Following on from the previous post’s topic, I want to dig a little bit more into the state-by-state breakdown of the demographic factors in the election.
When they’re not claiming that “shy Trumpers” aren’t being represented in the polls and turnout is going to save them, Stormtrumpers are instead arguing that everything’s rigged, and if Trump loses, it’ll be because of shady dealings at the individual-voter level (shady dealings by GOP state officials to suppress turnout among people of color and young people apparently aren’t a problem.) The problem Trump has isn’t vote counting or fraud, however, it’s demographics. As I noted before, the country is about 70% white (it varies depending on whether you include Hispanic identity), and less than half of white women are voting for him. Add in the 30-40% of white men who are queer, have a disability, have a degree other than an MBA, are active service or vets pissed off about his military nitwittery, or just plain aren’t total assgaskets, and there’s just no way he can win on a popular-vote level. There just aren’t enough of his kind of people in the country to outweigh the votes of everyone else. I suppose he could claim that the Census is also rigged, showing that there are fewer white people than there really are, but that starts treading into faked-moon-landing territory. However much his base may be amenable to such nonsense, the rest of us aren’t.
When he was forecasting the D primaries, FiveThirtyEight’s Nate Silver got almost all of the results right because his models were based in part on demographics. Both polls and early results showed that Bernie was primarily getting white men and young white women, and Hillary was getting everyone else. Dig into Census data for each state, add in some tweaks due to the particular style of primary (closed/caucus/w’ev) and predicting results was pretty darn easy. The only ones that didn’t quite line up were ones where the result was close–even within the margin of error. The same is going to be true for the general.
The five main demographic factors that indicate someone is likely to be a Trump fan:
- No college degree
- Evangelical Christian
The gender split is pretty much the same in almost all states except AK and ND (the only two that have more men than women) so there’s no point in doing a breakdown on that axis. Let’s work on the rest, though:
- Any state that is <40% white is automatically out: HI, DC, CA, NM
- Any state that has >30% with a bachelor’s is also out: MA, MD, CO, CT, NJ, VA, VT, NY, NH, MN, WA, IL, RI
- Any state that’s <25% white evangelical is out unless its white population is >80% or it has a disproportionately older population: WI, PA, NV
- Any state that’s >4% gay is out unless it has a high white or Evangelical population, or low education: OR, ME
And let’s also just throw in DE, since it’s borderline on most of those factors, and does whatever its surrounding states do anyway.
Add all those up: That’s 263 EVs. A nice, solid base to build from.
Now let’s look at some of the states that come close on some of those axes:
- AZ: 57.8% white, 25.6% educated, 20% evangelical, 3.9% gay
- FL: 57.9% white, 25.3% educated, 24.6% evangelical, 3.5% gay
- TX: 45.3% white, 25.5% educated, 25% evangelical, 3.3% gay
- GA: 55.9% white, 27.5% educated, 32.7% evangelical, 3.5% gay
- NC: 65.3% white, 26.5% educated, 40% evangelical, 3.3% gay
- MI: 76.6% white, 24.6% educated, 25% evangelical, 3.8% gay
- OH: 81.1% white, 24.1% educated, 25% evangelical. 3.6% gay
While each of those factors on their own aren’t enough to tip those states into a win for Hillary, the combination puts them into play. They’re all roughly about the same level of gay and educated, so whether they’re going to go Trump or Hillary will depend on the other two factors.
FL and AZ look pretty good on those two, but they’ve trended red in the past because of their disproportionately high percentage of seniors, and in FL’s case, because of the tendency of people with Cuban heritage to vote GOP. I’m going to argue that both of those things aren’t a factor this year. Arizona has seen an influx of Boomer retirees from California, which is a whole different ball of wax than the Northern snowbirds who used to set up shop there. Add in more Latinx growth, and it’s probably at least pink at this point. FL’s Cuban issue is a non-starter this year, what with Trump doing business with Cuba while it was under embargo. That’s going to piss off a LOT of people. FL, too has also seen an uptick in overall Latinx population, particularly from Puerto Rico, which makes voter registration for those folks pretty easy.
TX would seem to be ripe for the picking with all of these demographic factors, so why isn’t it more in play? Two things: 1. It’s very hard to register to vote in Texas, resulting in one of the lowest turnout rates in the country. It’s hard for your competitive demographics to be a factor when your turnout percentages are drastically different. They didn’t run exit polls in 2012, but in 2008, the white share of the electorate was 63%, and white evangelicals were 33%–a drastic difference from the overall state population. 2. Oil. And all the other Don’t Mess With Texas stuff. Educated non-evangelicals are a lot more conservative there because of the overall culture, which grows out of its main industries. If the Latinx vote manages to surpass expectations this year, it’s certainly possible to tip, but I think we’re still one general election out from that.
GA and NC have long been on the edge of blue, but aren’t quite there because of the relatively high white evangelical population. However, because Trump’s grossed out a lot of Good Christian women, they’re in play this year, when they might not otherwise be if a more traditional candidate was running. Add in the push to unseat NC’s governor, due to the fiasco with the anti-Trans bathroom law, and I think that’s a solid bet this year.
On demographics alone, MI and OH would seem to be ripe for Trump, what with their relatively high white populations. And indeed, I think OH might be a loss this year because of that. MI, however, is a different story because of two factors: Obama’s auto-industry bailout and the Flint water crisis. Where a lot of white working-class voters—even union members—are leaning right these days, the ones in that state haven’t done so, yet, because Obama has been very responsive to their crises, and HRC will likely be more of the same. Add in a population that’s slightly less white than OH, and I expect MI to stay blue this year. Indeed, Hillary has been polling very high there, so it’s basically guaranteed.
To our 263 EV base then, let’s add MI, which takes us to 279 and a certain win. Everything else is gravy: Add FL and you get 308. Add NC, AZ and GA: 350. If things are going particularly well, and we add OH and TX, it goes up to a whopping 406. Wow!
I certainly don’t expect a 400+ landslide, but it’s pretty clear, between polls and demographics, that that 279 is the absolute minimum that Hillary can expect this year. That’s probably also going to be the case going forward. While the GOP may still pick up more of the white working class in the future, it’s going to be hard to make any gains overall because the percentage of whites and conservative Christians is shrinking, and the percentage of people with a college degree is growing. Moreover, even though Boomers are living longer than previous generations of seniors, they’re rather more liberal than the Silent and GI generations were in their elder years. Millennials and Gen X are also considerably more socially liberal than previous <55 generations were, too. Issues like queer and reproductive rights aren’t going to be wedge issues for much longer. Additionally, if she has at least a couple of years with a relatively friendly Congress, Hillary will push through legislation aimed at helping the working class in general improve their conditions, which is going to stop much of the bleed off there. Strengthen unions, cut middle-class taxes, add a public option to health care, and do some other worker-friendly stuff like requiring paid sick and family leave and people will have a lot less to make them susceptible to the GOP’s frothing up of racism and xenophobia.
The GOP knows this, of course. There’s a reason that they’ve worked so hard to suppress voting in Democrat-friendly populations. Make it so the only people who have the time and transportation to stand in line at the polls are salaried and don’t have disabilities or young children to mind, and it’s easier to stay in power even if the overall demographics of the population change. But absent such blatant disenfranchisement–thanks to the courts, that’s getting curbed–demographic shift means the GOP can no longer expect to win big elections. They still have a lock on 35% of the population, but that’s all they have, and as trying to bring in more people only irritates that 35%, pushing them even farther right, they’re stuck. They’ll keep winning states that have a lot of straight, white religious folks without a college degree, but that’s it.
And speaking as someone outside of that 35%: Thank goodness.