Here’s my presentation at the early 2013 VDARE.com symposium, transcribed and then translated from spoken Sailerese into actual written English.
Hi, I’m Steve Sailer, and it’s a real pleasure to address our symposium. I’m going to talk about some overlooked aspects of the 2012 election.
As we get to the data, we’re going to focus on voting by state because that is, more or less, how Electoral Votes are counted. For Republicans to ever take back the White House, they will have to figure out more states they can win.
In the interest of simplicity, all the percentages are going to be for Romney’s share of the two-party vote. I’m leaving out Libertarian voters, write-ins, and so forth. I apologize for ignoring non-two party voters (I saw recently that Tom Wolfe wrote in Ron Paul’s name in 2012), but this expedient will allow us to think about just one number at a time: Romney’s share. Thus, if you want to know what Obama got, just subtract Romney’s percentage from 100.
I’m working with a huge poll that almost nobody’s talked about. It was conducted online by Reuters-Ipsos throughout the election year. This particular edition features a sample size of 40,000 two-party voters who responded immediately after voting.
Now, the Reuters-Ipsos panel has advantages and disadvantages versus the better-known Edison exit poll, which had a sample size of only about 25,000. I haven’t noticed any systematic differences in results reported by the two polls, but Reuters-Ipsos has a number of strengths for the serious analyst.
For example, the more celebrated exit poll wasn’t even conducted in 20 states, including Texas. If you want to know something about the future of American politics, you better know something about Texas. The Reuters-Ipsos poll had a sample size of 2,403 respondents in Texas. In summary, we’ve got a decent sample size on almost every state, not just 30 favored states.
Most importantly, Reuters lets anybody make any crosstabs they want of their results, while the Edison exit poll only lets subscribers who pay tens of thousands of dollars get their hands dirty with the data. So, the quality of discussion of the exit poll numbers has been constrained.
Below is something nobody has seen before, a table of Romney’s share of the vote by race in each of the 50 states.
The first column of percentages is Romney’s final share of the actual two-party vote. Nationally, Romney only got 48.0 percent to Obama’s 52.0 percent. (After all the votes were counted, Obama’s victory margin turned out wider than almost all polls had predicted. The Reuters’ poll has Romney at 48.5 percent, so it was a half-point too high.)
National, Romney won 58.1 percent of the white vote which, unsurprisingly, was not enough. He lost 97-3 among blacks and 72-28 among Hispanics.
Unfortunately, Reuters just lumps together American Indians with Asians and whoever else feels like calling themselves “Other.” Romney garnered only 39 percent of the Other, although that’s better than what the exit poll reported for Romney among Asians (26 percent, down a purported 9 points from 2008), and 38 percent among “Other” mostly American Indians (up 7 points from 2008). There was a fair amount of theorizing based upon the exit poll about why Romney did so much worse than McCain among Asians (although none about why he did so much better among American Indians).
The Reuters poll, however, suggests these sharp swings didn’t actually happen.
Which poll is right about the Other? Beats me. Mostly, the exit poll and Reuters are pretty similar, so when they disagree, I’d just recommend taking the average of the two surveys.
The Reuters-Ipsos Polling Explorer interface won’t display any breakdowns where the sample size is less than 100. But I managed to get around that cautious limitation by lumping together in huge California with each small state’s sample, then doing the math. That worked out fairly well. Rather than a minimum sample size of 100, I chose an aggressive minimum of merely 15. That’s quite small, so don’t trust each number above too much. Since it’s so hard to get these numbers, I felt it better to err on the side of giving my readers more rather than less information.
We’ll start our analysis with minority electorates, then give the white vote the careful inspection it requires. Yes, I know that white voters are out fashion, but they are still numerous and much more of a swing vote from state to state than are the trendier minorities.
The black share of the vote is routine almost all the way through. Traditionally, California blacks vote a little more Republican than the national blacks, and, sure enough, Romney hauled in a full 5 percent of California blacks versus 3 percent nationwide.
The one black figure that’s unexpected is Ohio, where Reuters reports that Romney get 13 percent of the black vote. That’s from a moderate sample size of 92 black panelists. A vast amount of money was spent on advertising in the battleground state of Ohio, so maybe Romney’s strategists can pat themselves on the back for buying a few extra black votes. Or maybe this 13 percent figure is just a fluke due to limited sample size.
A few anomalies like this are actually reassuring about the authenticity of the Reuters poll. The results fit my model of how the world works, of how various factors interact so well that occasionally I break into a cold sweat over the thought that maybe Reuters just made up the results! I mean, if you hired me to create a model of how demographic and regional factors work together, it would spit out numbers very much like these. But, the occasional unpredictable result, like Romney supposedly getting 13 percent of the black vote in crucial Ohio, is, in a way, confidence-inducing.
With Hispanics, you can see that Puerto Rican Hispanic states like New York (Romney got 18 percent of New York’s Hispanic vote) and Pennsylvania (13 percent) are a little bit further to the left than Mexican Hispanic states such as California (25 percent). But, most of the Hispanic vote falls within a relatively narrow band. Rather than swing voters, these look like solid Democrats who drift a little right if their white neighbors are conservative..
Ever since the election, we’ve been told constantly that the main thing Hispanic voters care about is amnesty for illegal aliens, and the only way for Republicans to ever win the White House again is to grant amnesty (and, while you’re at it, throw in “a path to citizenship”). If you doubt this is the right course for the GOP, just ask any Democrat and they’ll tell you.
If there is any state where this logic shouldn’t apply, it ought to be Florida, which Obama won by a hair. The two main groups of Hispanic voters in Florida are Cubans and Puerto Ricans, neither of whom care about “immigration reform.” The Puerto Ricans are born citizens, and yet they still vote overwhelmingly Democratic. You might almost think Democrats are pulling Republicans’ legs over amnesty …
The Cubans, as described in Tom Wolfe’s Back to Blood, have their own special immigration law that applies to any Cuban who can set foot on American soil. The Cubans used to vote heavily Republican, but Florida Hispanics now went overall 65-35 for Obama, suggesting younger Cubans are trending Democratic. In Wolfe’s novel, even the conservative cops among the Miami Cubans resent the Anglos as competitors who get on their nerves by thinking of Florida as part of America. And the Democrats are the natural home for the resentful.
There is a small difference between the Mexican American voters in California (25 percent for Romney) and Texas Hispanics (37 percent). That 37 percent sounds pretty good – it must be the pro-amnesty role models of the Bush family, while, as we all know, California Latinos were alienated by Proposition 187 — until you notice that Romney got an astonishing 76 percent of the white vote in Texas versus only 49 percent in California. So, relative to whites, Romney may have performed better with Hispanics in California where there is only a 24-point gap, not the 39-point gap in Texas. Or if you look at it proportionally, California’s 25/49 is almost identical to Texas’s 37/76. So maybe the Bushes and Prop. 187 don’t really matter, and what really matters is that Mexican Americans mostly vote Democratic because they find it to be in their self-interest for old-fashioned tax-and-spend reasons?
What about the white vote?
This graph below shows Romney’s share of both the total vote (in dark) and white vote (in red). The states are sorted in order of how well Romney did overall, with Utah at the top and Hawaii at the bottom.
It started out as a bar graph, but I had 100 bars (50 states times two), which seemed excessive, so I made the bars invisible and just left the values of the bars. If you look at Utah, you can see that Romney got 75 percent of the total vote and 75 percent of the white vote in the state. In Wyoming, 71 percent of the total vote and 74 percent of the white vote.
So, for Romney to do really well, he needed two things: states that are almost all white and whites that are almost all Republican.
Now, as you get further down, you see outliers where the GOP’s share of the white vote is far higher than the GOP’s overall performance, such as Alabama, Louisiana, and Mississippi. These are states typically in the deep south with large black populations where there’s a strong degree of white solidarity to keep blacks from taking over the state. For example, the state of Mississippi went for Romney 56-44, and the way he won was by getting 88 percent of the white vote. Why did he get 88 percent of the white vote? Well, Mississippi has the largest black population of any state and according to this Reuters-Ipsos poll, blacks in Mississippi voted 100 percent for Obama (sample size = 38)
So that’s kind of what diversity gets you in the long run. As Lee Kwan Yew of Singapore says, in a multicultural democracy, everybody ends up voting on race.
Probably the two most interesting states that Romney won are not in the deep south: Texas (76 percent of white vote) and Arizona (66 percent).
Texas is not really an old deep south state by any means. It has had a huge influx of Americans since oil was first discovered in 1901, and it has its own culture. It shows the possibilities of what a state could do in terms of going heavily toward Republicans as a bloc vote: 76 percent is a pretty amazing number, but that’s what it took to keep rapidly-Hispanicizing Texas handily Republican. If whites in Texas don’t vote consistently Republican, then the state, with its 38 Electoral Votes, will go Democratic in some future presidential election. And that would end the chances of the Republican Party as we know it ever regaining the White House. So, GOP, you better hurry up and put all those illegal aliens in Texas on the path to citizenship!
One thing to keep in mind about Texas is that its formidable degree of white solidarity is the result of generations of white Texans indoctrinating each other in the superiority of Texas over the rest of the country (as I noticed while a student at Rice U. in Houston). This solidarity has some real payoffs. For example, back in the 1980s Texas had a hugely successful anti-littering campaign featuring the slogan “Don’t Mess with Texas.” Politically, it turns out that Texas pride among whites keeps Mexicans discouraged. (Mexicans are not terribly hard to discourage.) On the other hand, the braggadocio of Texans has not necessarily endeared themselves to the rest of the country.
As you may have observed, the demonization of Arizona in the national press over the last few years has been virulent. The front page of the New York Times routinely featured articles about horribleness of white people in Arizona and how something needs to be done about them.
That’s because by the standards of Western states without many blacks, there was strong solidarity among Arizona whites, with 66 percent voting Republican. That frustrated Democratic efforts to register and turnout as many Mexican Americans as possible.
The most interesting states on the graph are the ones where Romney came close to 50 percent. These are the states future Republican candidates must improve in to have a shot at the White House.
The message you’ve heard ever since the election is that the Republicans lost because of the amnesty issue and therefore they must agree to amnesty and a path to citizenship. You know, the New York Times and the POTUS have all been explaining to the Republican Party how they need to pass amnesty right now for their own good. And if Republicans can’t trust the leadership of the Democratic Party to look out for their partisan interests, who can they trust?
Yet, the states in which Romney came close to winning are typically ones where he just did not get enough of the white vote. Consider Ohio, where Romney lost 52-48 overall by only getting a grand total of 54 percent of the white vote. Almost anywhere in modern American, Republicans have to win more than 54 percent of whites to win.
Here are some other north central states where Romney came fairly close:
Pennsylvania: 54 percent of the white vote
Iowa: 48 percent
WI 49 percent
Minnesota 47 percent
Michigan 53 percent
Romney couldn’t get the job done in these northern states not because of the tidal wave of Hispanics, but because he just didn’t get enough whites to show up and vote for him.
Let’s see where we could make the amnesty argument. Florida was close. And, as we know ever since the infamous 2000 election, Florida has been ripe for people with an ax to grind to claim that their particular panacea would have determined who won the Presidency. For example, I got a press release during the 2000 vote counting in Florida from a Sikh lobby. The Sikhs hate laws requiring motorcyclists to wear helmets because they muss up their turbans. Traditionally, helmet laws are the Sikhs’ hot-button issue. The press release announced that if Al Gore had come out against helmet laws, the Sikhs of Florida would have made him President. I checked their math, and, yeah, they had a point.
But the larger point is that this logic is mostly nuts.
But the Republicans don’t get it. At the moment, they think that all they have to do to get back to the White House is turn the party over completely to Marco Rubio. Let him negotiate amnesty with the Democrats. (What could possibly go wrong?) Mexicans must love the guy, right? After all, both his name ends in vowels.
Yet, do Mexican Americans even like Cubans, such as Sen. Rubio? (One of the hidden messages of Back to Blood is that Cubans don’t care at all about Mexicans.) Nobody seems to have checked.
Virginia is another interesting state. It’s an example of how the Republicans are beginning to shoot themselves in the foot with legal immigration. The Washington DC suburbs are home a large number of well-educated legal immigrants, and, it turns out, they like to vote Democratic. Even if they’re making a lot of money and it’s going to cost them in taxes, these legal immigrants just find the Democrats more to their taste.
Then there are what I call the Clean Green states such as Colorado (where Romney won 52 percent of whites), New Hampshire (48 percent), Oregon (48 percent), and Washington (46 percent). Amnesty isn’t going to win them those states.
There’s New Mexico, with its large Hispanic population, but once again the GOP lost there because they only won 52 percent of the white vote. New Mexico is interesting as a view into the future of Hispanicized America. Hispanics have been in the Upper Rio Grande Valley for 400 years, yet the state that does not attract many illegal immigrants. How come? Because there aren’t many jobs in New Mexico. Why not? Because it has been filled up with Hispanics for its entire history, and they don’t create a lot of jobs.
What about California? Surely, that’s a state where whites have been crushed under the rising tide of Hispanics? Actually, Romney only won 49 percent of the white vote there. Kind of hard for a Republican to win that way.
As we all know from having heard it over and over that Republicans were doing fine in California until they shot themselves in the foot with Proposition 187 in 1994. What they don’t tell you is that George H.W. Bush won less than 33 percent of the total vote in California in 1992, two years before Proposition 187. But who has time to fact-check The Narrative?
Nevada might be the closest thing to an example supporting the amnesty-uber-alles narrative. Romney won a mediocre but not terrible 57 percent of white votes there, but lost due to Hispanics (and Filipinos) voting heavily Democratic. Unfortunately, the Reuters-Ipsos poll only has a Nevada sample of 14 Hispanics, so we’re flying kind of blind here.
My impression of Nevada Hispanic voters is that the big issue for them is not amnesty, it’s that they were just hammered by the mortgage meltdown of 2007-2008. Nevada long led the country in foreclosures. Nevada Latinos were flying high during the Bush Bubble, but haven’t forgiven Republicans since for their defaulting. How amnesty will cure that for Republicans is a mystery.
Let’s briefly look at the national level. A one-word characterization of Mitt Romney’s campaign would be bloodless. He stressed serious, respectable issues involving entitlements and taxes. He avoided any mention of anything ungentlemanly. Unfortunately for Romney, he’s living in a time that our leading man of letters calls the age of Back to Blood.
In contrast, coming out of the 2010-midterm elections, Obama saw he had a real problem. The Obamamania of 2008 had carried him to a large victory over a wounded and already flawed Republican candidate. But how was he going to re-mobilize his base, which largely consists of the margins of American society, without the Hope and Change piffle of 2008?
The Obama base is, to be blunt, the fringes. The epitome of Romney’s base is the married white father, while the essence of Obama’s base is the single black mother. Obama’s base hadn’t bothered to show up to vote in 2010, so how was he going to motivate them in 2012? The former are a lot more likely to vote out of a sense of civic duty, while the latter need some emotional motivation.
Here’s a table of data I published on VDARE.com just after the election that clearly shows the Core v. Fringe distinction:
|Reuters-Ipsos Exit Poll||Romney’s Share||Sample Size|
|Married white Prot.||74 percent||11,761|
|White Protestants||70 percent||15,732|
|Married white men||65 percent||7,001|
|Married whites||63 percent||24,176|
|Married white women||62 percent||17,175|
|White Catholics||57 percent||8,173|
|Married men||58 percent||7,910|
|Married women||55 percent||19,196|
|Single white men||51 percent||3,383|
|Married other races||48 percent||958|
|All Voters (2 candidate)||48 percent||40,000|
|Single whites||48 percent||10,270|
|Single white women||44 percent||6,886|
|Other races||39 percent||1,642|
|Married Hispanics||35 percent||928|
|Single men||39 percent||4,092|
|Married Jewish men||40 percent||419|
|Single Jewish men||30 percent||163|
|Married Jewish women||34 percent||652|
|“Other orientations”||31 percent||229|
|Single other races||28 percent||684|
|Single women||31 percent||8,801|
|Single Hispanics||21 percent||656|
|Single Jewish women||23 percent||328|
|Black single women||2 percent||925|
At the top are Mormons at 86 percent for Romney. Now, obviously, Mormons are a minority, but they’re increasingly the only minority group in modern American that still tries to act like they’re part of the core.
Then come married white Protestants (74 percent), then white Protestants, married white men, married whites, married white women, white Catholics, whites, married men, marrieds of both sexes, homeowners, married women, single white men, married other races and men in general.
At the bottom are black single women at 2 percent for Romney. Then blacks, gays and lesbians, single Jewish women, Hindus, single Hispanics, single women, single other races, other orientations. I’m going to stop there. “Other orientations” comes from the sexual orientation question. They gave you four choices: heterosexual, homosexual, bisexual; and for those who didn’t find those adequate, “other” was a choice. The Other Orientation folks went strongly for Obama.
Obviously, this turned into an election based on identity, on whether people felt themselves in the core of America or in the fringe of America. The core versus fringe can be defined in a couple of ways. For example, over multi-generational periods, do you come from people who settled this country a long time ago, or are you, say, an immigrant from Somalia who is now going to gift us with all the lessons that Somalis have developed over the eons on how to run a successful country?
Or, on a personal level, are you somebody who is married, has stayed married, has children, owns a home, and is employed? Or are you somebody who’s single, renting, who basically doesn’t find your life satisfactory and is looking for somebody to blame?
The way the Obama campaign turned out their base was to whip up feelings of resentment toward core Americans, toward those people whose ancestors had built the country, who largely keep it running today and who in their personal lives have done a pretty good job of keeping their act together.
Obama did a spectacular job of taking those two kinds of people from the fringe, and telling them that they should resent the white married people of America, the ones who own their homes, the ones whose grandparents helped make this country, and that there’s something shameful, unfair, or at least uncool, about coming from the core of America.
It was a brilliant strategy. Obama ran a really ugly, nasty campaign full of subliminal hatred. The Obama campaign did a good job keeping the stew of ill will they were brewing somewhat under wraps until after the votes were counted. But in the days following the election, out came pouring the chest-beating Suck-It-White-Boy exultation, the mindless fury at the losing white male bogeyman for being old and white, but, mostly, for losing.
The Republican Brain Trust now assumes that the way to solve this problem is via amnesty, just like their good friends the Democrats keep telling them. Amnesty, however, will be seen as white America’s surrender declaration, as an official invitation to kick the former top dogs while they’re down. And who can be expected to resist that?