« Power Outage | Main | Remapping SUSA »

How Southern is the South?

07 Mar 2008 01:12 pm

obamawsb%201.png

Brendan Nyhan has a bunch of interesting charts about the Democratic race of which this one, showing the negative correlation between Obama's level of white support and the size of the state's southern baptist population. If you ignore all the states with really small southern baptist populations and just focus on the South you see a correlation that holds up really well. Basically, if you think of the number of southern baptists as a proxy for how "southern" a given southern state is, you see that the more culturally Dixie states are the ones where whites are most hostile to Obama's message. They're not, however, necessarily bad states for Obama overall, since some (like Alabama and, soon, Mississippi) have African-Americans as such a large share of the Democratic vote that he wins anyway.

Share This

Comments (28)

I would like to see african american population % as a proxy for southernness - Is he doing worse among whites in areas with more blacks?

This seems a reasonable component to the analysis. However, I think it would be a good idea to toss out any caucuses in which the turnout was only 5% or less of the Democratic electorate as being just too demographically unrepresentative to be meaningful.

Also, I think that Obama did very badly among Catholic whites as well, e.g. in NJ, NY, and RI, states with almost no Southern Baptists.

Anyway, I think a "non-black" analysis is much more important than a "white" analysis. If Obama were doing as well among non-white non-blacks (e.g. mostly Hispanics and Asians) as among blacks, he'd be an extremely formidable November candidate, and would have already won the primary in a huge landslide. Instead, he's doing *worse* among these important groups than among whites.

Fewer whites voted for Obama (compared with Hillary) in Ohio than in South Carolina? Even factoring in that Edwards obviously took more white votes away from Clinton in SC than from Obama, that's astonishing.

To the first commenter: I think all Democrats do worse among whites in the southern states with more blacks. In places like South Carolina, Alabama, and Mississippi, there just aren't that many white Democrats any more outside of college faculties. Politics in those places has always been structured around race. It's a little different in a state like Tennessee, where the mountainous, moderate east gave the state a Republican influence decades before the Dixiecrats got a new name.

Very interesting, and pretty damning.

It would be interesting to compare non-racial demographics like education, income etc., find the group that is the strongest pro-Hillary overall, and then look at the Hillary-Barack split in that demo among African-American voters.

Echoing anonymous, this seems to be a variant of the analysis that Obama does well among whites who live in states with little diversity because whites in those states (while not necessarily being less prone to racism) have less of a history of racial tension.

Missouri pretty much seems to be in the middle for everything, huh?

At the risk of overinterpreting the available data, I would note the Atlantic coastal states (VA, GA, and SC) are above trend, as are the Southwest states (NM and TX). In contrast, the ones at or below trend are Gulf Coast and interior states--and I am willing to bet if we looked at more detailed breakdowns of places like Alabama and Louisiana, the numbers for Obama would get worse as we went away from the coast and farther into interior.

Which I believe suggests the South is not exactly an undifferentiated region, but rather can be subdivided further in politically relevant ways. And another way of putting all this is that when it comes to the South, the closer you are to Arkansas, the higher Clinton's support among white voters.

Which is why in turn I am beginning to believe the regional effects in this race on Clinton's are being driven as much by affinity with Bill as affinity with Hillary. Specifically, I suspect she is the one driving affinity in places like the Northeast Corridor and lower New England, but he is the one driving affinity in these subregions of the South (and I suspect Appalachia, another of her top regions). And I suspect they at least share credit for the last of her important regions (the Latino-heavy sections of the Southern Border).

Is either Democrat supposed to win states like GA, OK, TN, AL in November?

IMHO this is as useless as looking at gender ID chart for HRC. Its not going to change or affect the candidate since they are not going to change their race or gender. Lets cross the Naval Gazing Threshold. Thanks.

Is either Democrat supposed to win states like GA, OK, TN, AL in November?

Probably not, but Bill Clinton won TN twice and GA once. He even won MS in 1996. So, one can dream.

Does Jesus have something against black people?

ChuckE,

What happened in Ohio was an Appalachian effect. Clinton has consistently done very well with Appalachian Democrats, and she ran up the score on Obama in the Appalachian parts of Ohio. Again, at the risk of overinterpreting, I'd suggest that is why Maryland is also a bit below trend (it has a little bit of Appalachia where Clinton did relatively well).

Monster,

I actually think Clinton could have a shot in TN. In a landslide scenario, Obama might have a shot in Georgia, although I would personally suggest concentrating on the Carolinas first.

huh,

Just an FYI, the reason people are looking at Southern Baptists is that they couldn't find a better proxy for "Southerness" in the exit polls--the Baptist part is just coming along for the ride.

So,

But I think the idea is that what is going on here is cultural, not demographic: again, Southern Baptist is just being used as a proxy for cultural attributes, for lack of something more direct in the exit polls. And culture is not so inflexible as demographics: you may not be able to entirely change your own culture, but you can learn about other cultures such that you become better at things like communicating, deal-making, and so on.

So, I don't think exploring the cultural affinities in political contests is pointless. Rather, it may help those politicians figure out what cultures they need to understand better in order to appeal to more people.

Sure, the number of Southern Baptists is a "proxy" for "cultural Southern-ness." It's an even better proxy for, um, Southern Baptist-ness. I grew up in NC, not far from Jesse Helms's Hayes-Barton Baptist Church, and in my view a lot of Southern Baptists do in fact think Jesus has a problem with black people.

Actually, while this graph shows a negative relationship, it doesn't show a clearly strong negative correlation. For that we need a correlation coefficient, and just looking at all those states clustered on the left-hand side of the graph, I'd say that the coefficient is pretty weak. There's a lot of overlap here between Southern Baptist/southern states and the rest, which is obscured by the x-axis separation between the two groups, not to mention that tendentious line. TBS, if you take the ten observations that matter, they do appear to follow a negative regression line, so there may be something to this; but the observations are few, and the dispersion is great. And that's not to mention the curiosity of the notion that white southern/Southern Baptist racism accounts for voting for Hillary Clinton. Jeez--but I guess this is the quality of analysis we can expect from a guy who's on record as saying that everything you need to know about contemporary southern politics can be learned from a book published 59 years ago.

huh,

That may be true, but it is often the case that people interpret Jesus as saying things they already believe. So, we can't really separate out cause and effect in this case.

David,

I just want to underscore that one need not believe certain subsets of white Southerners are voting for Clinton because they are too racist to vote for Obama. Indeed, as I suggested above, I think once you break down in more detail exactly where Clinton does best among white Southerners, it becomes more likely they are voting for her because of her association with Bill Clinton (broadly defined to include her own time as First Lady of Arkansas).

In general, it frustrates me that some people seem to approach these problems with the assumption that the default is that people should like Obama and dislike Clinton, so if they support Clinton, it must be because they found something to dislike about Obama. Usually, the more obvious hypothesis is that they are voting for Clinton because they actually found things to like about Clinton, and indeed polls suggest that most Democrats like them both.

On reflection, the last bit of my 3:14 post may have come across as a criticism of David. So, I just want to note that I understood David was criticizing Matt's original blog post, and I was intending to agree with David.

Southern Baptists are Southern Baptists because they were pro-slavery and founded the Southern Baptist Convention in 1845 as a rejection of the abolitionist movement. "Slavery was the final and most decisive factor which led Southern Baptists to form their own convention." They officially denounced their positions on slavery and racial superiority in 1995 (the same year that Mississippi abolished slavery).

Southern Baptists are Southern Baptists because they were pro-slavery and founded the Southern Baptist Convention in 1845 as a rejection of the abolitionist movement. "Slavery was the final and most decisive factor which led Southern Baptists to form their own convention." They officially denounced their positions on slavery and racial superiority in 1995 (the same year that Mississippi abolished slavery).

On Oklahoma, I'd just note that Edwards was on the Dem ballot here even though he had already suspended his campaign by then. Edwards has plenty of support here, and got something like 20% here, if I recall.

My sense of things from talking to neighbors, etc., is that almost all of that 20% would go to Obama in the generals, all but handful of Hillary's votes would also go to Obama. On the other hand, I'd be surprised if Hillary (against McCain) could get more than half of Edwards' supporters, and many, many Obama supporters would stay home given the tactics that she has used against him.

In Oklahoma, things would likely play out like this (and I know that some of the OK superdelegates agree): in an Obama vs. McCain general, McCain wins around 55-45, but the Dems retake the state legislature, Jim Inhofe is seriously threatened by Rice, and things on the ground change substantially in city governments; in a Hillary vs. McCain race, it would be a McCain blowout, but with low turnout and net losses for Dems around the state.

My guess is that this is not a unique situation in many of the "red" states. Quite honestly, I grew up in Wisconsin and think that McCain has a chance to do really, really well among working class Catholics (i.e., in Milwaukee, Racine, the Fox Valley) in a matchup with Hillary and, depending on how she "wins" the nomination, African Americans in Milwaukee will likely stay home in droves, as will most of Madison. Wisconsin was for Kerry, but not by much, and the Swiftboat campaign was *very* effective. I wouldn't be surprised to see McCain take Wisconsin if for no other reason than low turnout and some Rovian tactics against Hillary. With Obama its high turnout, and swiftboating attempts would backfire pretty quickly if they involved racial or religious innuendo.

At the end of the day, with Hillary Dems would likely win the presidency, Congress would tilt more Dem, and local races would be little affected. She is just not going to attract many voters beyond a traditional Dem base, bringing nothing new to the table. Midterm elections are a net Dem loss, and its difficult for me to see Hillary as anything but a one term president.

With Obama, on the other hand, there is a pretty good chance of shaking things up from top to bottom and shifting a good chunk of the political spectrum in a progressive direction by making politics-by-bullshit more difficult to pull off. He would have a mobilized coalition to put all kinds of public pressure where it might do some good, but one that he would probably have a healthy fear of pissing off.

Even in the worst case scenario (as far as I am concerned) Obama doesn't live up to the hype and we just get the competency of a Clinton government but without the drama. This is much better than even the best case scenario for a Hillary presidency.

Why pick the Southern Baptist religion? Why not pick the Jewish religion or the Catholics? There seems to be a selection bias here.

Which I believe suggests the South is not exactly an undifferentiated region, but rather can be subdivided further in politically relevant ways.

It's the Old South vs the Deep South.

Just Karl,

I don't think that distinction works. First, at least a couple states are usually considered in both (GA and SC). Second, it doesn't clearly address places like Arkansas or Tennesee. I actually think it is more like a Eastern Lowlands versus Appalachia versus Mississippi River versus Western South distinction (or something like that). The Appalachian and Mississippi River white Democrats are the ones really going for Clinton, Obama does better among the Western South white Democrats, and he does even better among the Eastern Lowlands white Democrats.

VA, NC, SC, and GA are Old South. Certainly in places like Charleston and Savannah. The Old South has a cultural heritage that pre-dates the Revolution. The architecture, fine arts, and cuisine are much different than those found in the Deep South agricultural states of MS, AL, AR, LA, and western Tenn. Even the accent is different. (Think John Edwards vs Trent Lott)

Appalachia is a completely different animal altogether and I would argue not even "southern" in the traditional sense. These states include eastern Tenn, KY, WV, eastern OH, and western Penn. The Hatfields and McCoys were from WV and were neither Old nor Deep South.

Why have none of the major left blogs created a nice chart analyzing income, education and economic insecurity and voting patterns?

Class is the reason Clinton is competitive with
Obama. Obama's netizens do not want to deal with implications of pushing the wine-track ascendancy in the party so they attempt to obscure class. They disparage middle and low income white voters as racists or ignorant to avoid the issue. Downscale whites who vote their economic interests are continually disparaged as racists while upscale whites who do the same are allowed to cover their economic royalism with post-partisan and post-racial bows.

The vanguard always turns its back on the proletariat.

Thinking about it further, the New South focuses mainly on Atlanta, but also includes Raleigh-Durham, Greenville-Spartanburg, Richmond, Nashville, and Charlotte.

One BIG factor you are ignoring here is campaigning. I live in East Tennessee and Barack Obama did not set foot inside the Volunteer State in the last year. As a result, voters in this state had no idea who he was. And by that, I mean they had no idea how we respond interacting with local voters about issues here. Where he doesn't show up, he does poorly. Same thing with Oklahoma.

He needs to spend a lot of time in Appalachia, beginning with PA and then on south through WV, KY, VA, TN and NC. Get to know people here and they will give him the benefit of the doubt - but only after they see him in person and not on the TV. People in Appalachia are (rightly) sensitive to perceptions that they are backward. They have a deep distrust of slick-talking politicians who ignore them when it's convenient. But when those seemingly slick politicians show up and acknowledge that Appalachian people are wonderful people with real problems then those slick politicians become trusted leaders. You want an example of this? JFK in West Virginia. He faced a barrier just as deep as Obama does today. But he toured Welch and other areas and showed that the government can help here (though some were understandably embarrassed by the negative and often mocking attention paid the region when Charles Kerault and his "Christmas in Appalachia" crew came in).

Southern Baptist membership means nothing in this primary. Besides, most Southern Baptists are Republican these days. But showing up will make a big difference.

Re: Southern Baptists are Southern Baptists because they were pro-slavery and founded the Southern Baptist Convention in 1845 as a rejection of the abolitionist movement.

Um, those pro-slavery people kinda died a while back. No one currently living was involved in the founding of the SBC.

Re: I grew up in Wisconsin and think that McCain has a chance to do really, really well among working class Catholics

Maybe-- if he abandons the "I'm Bush's third term" shtick. McCain's only real chance for victory is to do what Nicholas Sarkoczy did in France and run against the incumbent president, promising a major overhaul of GOP policy and positions. McCain's someone who could pull that off, but so far there's no evidence he will. If he continues to run on "once more, with feeling" I cannot imagine how he could pull off a victory even if the Dems run Brittney Spears and Paris Hilton.

Re: These states include eastern Tenn, KY, WV, eastern OH

Minor quibble: you mean "southern Ohio", not eastern. Everything south of Columbus-Dayton is Dixie Lite, and has been since the days of the Copperhead Democrats. Eastern Ohio meanwhile includes the triad of Cleveland-Akron-Youngstown, solid Democratic turf.

I'm surprised nobody has analyzed the Democratic primary contest in the light of Robert Putnam's five-year study on diversity and social solidarity, which he describes in a paper presented to the Nordic Political Science Association in 2007. Drawing on empirical evidence from the U.S. and Europe, the paper, titled "E Pluribus Unum: Diversity and Community in the Twenty-First Century," in part concludes that "[i]n the short to medium run. . . immigration and ethnic diversity challenge social solidarity and inhibit social capital." I think we're witnessing this play out in the Democratic primary. Perhaps someone at the Atlantic is up to the challenge?

For anyone interested in further exploring Putnam's analysis, here is a link to his paper:

http://www.blackwell-synergy.com/doi/full/10.1111/j.1467-9477.2007.00176.x


Comments closed March 21, 2008.

Copyright © 2008 by The Atlantic Monthly Group. All rights reserved.