Skip to content

Red states, blue states, and affordable family formation

A couple of colleagues sent me a copy of an article by Steve Sailer in The American Conservative called Value Voters and subtitled, “The best indicator of whether a state will swing Red or Blue? The cost of buying a home and raising a family.” It’s here online, and an excerpt is on Sailer’s blog.

The article gives a new (to me) take on the red-state, blue-state paradox: the point that Republicans do well in lower-income states, even though they do better among richer voters. Sailer points out that the strong Republican states in the south and middle of the country have lower cost of living (which of course goes along with them being low income). But, more to the point, many of the rich, Democratic-leaning metropolitan have housing costs that are even more expensive. Sailer attributes some of this to what he calls the Dirt Gap–coastal cities such as NY, Boston, and LA are bounded by water which limits their potential for growth, as compared to inland cities such as Dallas or St. Louis: “The supply of suburban land available for development is larger in Red State cities, so the price is lower.”

Sailer notes that Republicans do better among married voters, and he has the following impressive graph:


Even excluding D.C., the correlation is high. (Sailer has some discussion about why he’s only looking at white women here. I don’t follow all his reasoning here; you can read his article to get the details.)

We were talking about Sailer’s article around the office today, and the point was made that the above graph illustrates that Bush did better in states where people are more culturally conservative (in this case, as measured by the chance of women marrying at a young age).

I’m not quite sure how this relates to our finding that the Republican advantage among high-income voters is large in poor states and small in rich states. One suggestion was that, in poor states, your high income gets you a larger house, whereas in a rich state, even with a high income you’re not living in a palace. Sailer might say that cultural conservatives will move out of rich states, even if they’re high income, because they want a nice house with a yard for their kids to play in. We’ve tried to look into some of these things, but it’s a challenge to analyze data on moving. It’s easier to analyze cross-sectional surveys, so that’s what we spend most of our time doing.

To get back to the main point, Sailer is making a geographic argument, that Democrats do better in coastal states because families are less likely to live in coastal metropolitan areas, because housing there is so expensive, because of the geography: less nearby land for suburbs. This makes a lot of sense, although it doesn’t really explain why the people without kids want to vote for Democrats and people with kids want to vote for Republicans. I can see that more culturally conservative people are voting Republican, and these people are more likely to marry and have kids at younger ages–but in that sense the key driving variable is the conservatism, not the marriage or the kids.

I think Sailer’s arguments are interesting but I can’t quite follow him all the way to his conclusion, where he says that “the late housing bubble . . . reduced the affordability of family formation, which should help the Democrats in the long run.” I just don’t see where the data are showing this. I’m not saying he’s wrong–I’ve certainly heard it said, for example, that the postwar boom helped the Republicans, and conservative causes generally, by moving millions of people up into the middle class–but it just seems like a stretch. I also don’t follow his claim that if the Republican party should move to restrict immigration–that it “could then position itself as the party of more weddings and more babies.” Immigrants have weddings and babies too, right?

That said, the point about affordability of housing seems important, and it’s not always captured in standard cost-of-living measures (see here). And it’s interesting to see these correlations between demography and voting.

P.S. Some of the comments to Sailer’s blog entry are pretty amazing; for example, “For white engineers like me this is a hard call. Engineering jobs outside of these deep blue areas seem to be very rare. High-tech companies tend to get rich and attract immigrants of other ethnicities and races who happily raise families in apartments. You will not be able to find a white woman who shares this preference” and “Expensive housing is merely an attribute of the kinds of lives which are led by the people who choose Death over Life.” I suppose it’s good for a Manhattan resident to know that there are people out there who think this sort of thing. I mean, I know that people say these things, but there’s something weird about to seeing them in writing.

P.P.S. There is further discussion of Sailer’s “affordable family formation” idea here.


  1. aem says:

    "Eighty-eight percent of the increase in the median real price of a house in Seattle since 1989 is the result of land-use restrictions. So finds University of Washington economist Theo Eicher."(from

    So, once your life has been made miserable by central planners, how do you vote?

  2. Bertil says:

    What about saying that urban voters tend to consider that collective investment is more necessary, like seaports & public transport, because they see the need at their doorstep; while country-dwellers think that someone who can't provide for himself is lazy, and don't understand the need for town-hall funded shelters? If life is more expensive in the cities, one can understand that people there are more keen on raising the minimum wage, no? Same goes with attitude toward international intervention: more foreigners in capitals; same with guns: noise in your flat is more likely to be your drunk room-mate then in your house.

  3. Old Crone says:

    I would imagine Sailer tracked white women because it's whites who typically move Republican later in life. As far as I know, most other groups continue to vote Democrat regardless, but I'm open to correction on that point.

    Which is why, though immigrants do have babies and weddings, those babies and weddings aren't going to help the Pubs remain politically relevant in this century.

    Anyway, I'd speculate that, till recently, the Democrats were viewed as the party of minority racial-spoils politics, and the Republicans were viewed as being the party that defended the Great White Status Quo.

    After you have a couple of white kids, affirmative action looks a lot less like a noble social program and a lot more like a threat to your own offspring.

    Ditto for a lot of other programs designed to win (Democratic) minority votes… you don't really want to move to the suburbs for the "good schools" and then vote in a guy who will swamp your investment with, say, inner-city Hispanics. If you can't afford private schooling for your increasing brood, you're going to watch the demographics and quality of that public school like a hawk.

    I would also suspect that abortion rights play less of a role in women's decisions about voting after marriage and children than they do before. Certainly I remember being aghast that any woman would EVER vote for an anti-abortion candidate.

    Five years of marriage and three kids later, I filled in the bubble for Ron Paul. It just isn't a huge issue to me anymore. Raising and educating white kids in a country rather politically hostile to them… that concerns me a little more.

  4. Steve Sailer says:

    Thanks for the posting.

    As for why states with more families tend to vote more for family values candidates, let me cite one example from a 2005 article of mine on the underlying causes of red and blue states:

    "Why the correlations? Consider how differently one well-known issue can seem depending on your family structure: Should the government let the Boy Scouts ban gay men from becoming scoutmasters? To voters who are single, or married but childless, or have only daughters, this often appears as a purely abstract question of justice: of course, everybody should be guaranteed equal opportunity to be a scoutmaster. Yet, to citizens with sons, a ban may seem like a common sense precaution against temptation: of course, homosexuals shouldn't be allowed to lead their boys into the woods overnight."

  5. Steve Sailer says:

    As for why restricting immigration would help Republicans electorally, the most obvious reason is that both major immigrant groups, Hispanics and Asians, vote consistently Democratic.

    More subtly, there is some evidence that immigrant ethnic groups, or at least Hispanics, are less sensitive to the affordability of family formation. For example, in crowded, expensive California, immigrant Latinas average 3.7 babies per lifetime as of 2005, while American-born white women were having babies at the rate of 1.6 per lifetime. "I need my space!" is the kind of thing you hear more from American-born whites than from immigrants. But California Hispanics still vote 70-30 Democratic.

    Finally, mass immigration tends to make family formation less affordable for white voters — due to supply and demand, housing prices go up and wages go down. And illegal immigration (but not legal immigration) makes costly private schools look more like a necessity to whites. (See Sandra Tsing Loh's various articles in The Atlantic Monthly for what white liberals in the Hollywood Hills think of sending their children to public schools with the children of illegal immigrants. White conservatives hold similar opinions.)

    Thus, the state with the highest percentage of immigrants, California, which voted for the GOP candidate 9 out of 10 Presidential elections from 1952-1988 has gone solidly Democratic four times in a row. That's partly due to immigrants voting Democratic, but it's largely due to the white vote shifting to the left in California.

  6. Anonymous says:

    I would think that being in a poorer state, thus having access to buy a bigger house as a rich person, increases your chances of being that much more isolated from the rest of the world. Everything is in your own house, so there is little incentive for you to share anything and it therefore makes you less "used to" doing so. (I'm assuming this has been previously studied, the idea of proximity and physical isolation having an effect of people's behavior…?) That makes people more wrapped up in themselves and more interested in policies that don't take money from their own good to put towards the collective good or others' good(s?), which leads to more Republican views.

    However, I also agree with the other commenters that this probably applies much more to white voters than others.

  7. LemmusLemmus says:

    "We were talking about Sailer's article around the office today, and the point was made that the above graph illustrates that Bush did better in states where people are more culturally conservative (in this case, as measured by the chance of women marrying at a young age)."

    Very good point. A better test would be to correlate data on voting with data on housing prices.

  8. Jacob Felson says:

    The higher fertility among Republicans is fascinating to me. Given that political party is often inherited from parents, what is (or what will be) the fertility advantage of Republicans in the voting booth? Is the Republican fertility advantage one of the reasons that Republican party registrations have been increasing (at least until the last several years)? Maybe I'm wrong on the numbers here. But if the fertility differential between Republicans and Democrats is not fleeting, there has to be a real effect on voting, and it might be interesting to know the size of that effect. To what extent can the Republican fertility advantage counter the Democratic advantage among the growing Hispanic population?

    That brings me to the following — does the correlation between size of family and Republican voting hold among all ethnic groups to the same extent? I might expect an interaction between ethnicity and number of children..

  9. Matt Stevens says:

    Sailer is a somewhat infamous commenter on several blogs. I know him from the Matthew Yglesias site; commenters there regularly condemn him as a racist.

    In one exchange, Sailer claimed that statisticians largely "ignore the work of R.A. Fischer" because he was a eugenicist. He was more or less laughed off the boards for that one.

  10. Patrick H says:

    What's the surprise here? Single women want security. Dems promise them support that makes their lives easier. Married women have security. They want government to stay out of the way and keep it's hands off their money. Dem policies undermine the importance of marriage and family and this helps bring about their statist goals by making ever more people dependent on the state. This isn't new.

  11. Andrew says:


    What's new is that the rich states go for the Democrats and the poor states go for the Republicans. 30 or 60 years ago this wasn't the case, and 100 years ago the pattern went in the opposite direction. But the Democrats have been, by and large, to the left of the Republicans on economic issues during all these periods.

  12. TGGP says:

    Steve's comment sections are indeed a wretched hive of scum and villainy, but they're a lot better than they used to be!

    Obtaining property increases support for property rights.

    The Inductivist has an unusually data-free post on the psychology that keeps upwardly-mobile immigrants from moving to the political right. Some posts with data backing that point are here and here.

  13. Matt Stevens says:

    Sorry for misspelling R.A. Fisher's name, by the way.

  14. Steve Sailer says:

    My entire article "Value Voters" is now posted at


  15. blah says:

    Matt Stevens:

    Great, but do any of the Yglesias readers have a graph like that? I think you'll find that those who are first to laugh and label other people "racists" or "communists" or whatever are really not particularly interested in truth.

    Sailer is definitely onto something here, onto a political truth more deep than most of the posts by Matt Yglesias' commenters (or Yglesias himself, insightful though he may be).

  16. blah says:

    Matt Stevens:

    You said:

    In one exchange, Sailer claimed that statisticians largely "ignore the work of R.A. Fischer" because he was a eugenicist. He was more or less laughed off the boards for that one.

    Out of curiosity I looked for that thread and found this quote:

    There are a number of scientific geniuses, such as Ronald A. Fisher, who aren't publicized today as widely as their contribution to the human sciences would suggest because political correctness has such a hammerlock on intellectual discourse today. Weyl was hardly in Fisher's class (who is?), nor in that of, say, Sir Francis Galton or Carleton Coon, but to say that politically incorrect thinkers of the past aren't widely remembered today is merely to say that we live in an age distorted by political correctness.

    Not quite the same thing. Your paraphrase is something of a distortion; Steven Sailer's statement, arguable while it might be, is about the *publicization* of Fisher's achievements. He never said that *statisticians* ignored Fisher.

  17. Steve Sailer says:

    This is a minor matter, but "blah" is correct: when I said that Sir Ronald A. Fisher is not "publicized" as much as his accomplishments deserve, I was talking about publicized in the mainstream media, such as, say, on documentaries on PBS. Obviously, Fisher's name comes up all the time in college classes on statistics, genetics, and evolutionary theory, but that's not what the word "publicized" normally means. Fore example, people don't hire a "publicity agent" to get their names mentioned more in Biology 401 classes.

  18. Mary says:

    Conservatives like to think that it is families and marriage that make people Republicans. I think this is correlative rather than causal. You can make a case for the real causal factors being education alone. Women with higher levels of education marry later and live in the coastal cities. That explains the marriage graph. The expensive coastal cities have higher average levels of education. Of course, education and income also link back to g as well.

    It's hard to ignore the more negative social stats in the Red states vs. Blue – for unwed mothers, divorce, etc. These are not necessarily more "moral" locales.

  19. blah says:

    It's hard to ignore the more negative social stats in the Red states vs. Blue – for unwed mothers, divorce, etc. These are not necessarily more "moral" locales.

    You need to look at the actual ethnic breakdown. Many red states in the south have large numbers of blacks, who do not tend to be conservative. Conservative whites in those states have lower divorce rates and out-of-wedlock birth rates than liberal whites.