There’s an interesting discussion underway between Pollster.com’s Charles Franklin and fivethirtyeight’s Nate Silver about variable rates of white voter support for Obama, particularly in the South. Franklin plots a graph and notes that there does seem to be an inverse relationship (in the South at least) between the size of the African-American population and Obama’s support levels from white voters. Silver notes the very different rates of white support for Obama within the South, and wonders if it might have to do with voter exposure to Obama; he did a lot better in states where he actually campaigned.
I think Nate’s on to something, based on some very specific numbers. In 2004, Kerry’s share of the white vote in Mississippi and Alabama was 18%, and was 22% in South Carolina. This year Obama’s share of that vote sank to 11% in Alabama, and 10% in Mississippi, yet rose to 26% in South Carolina.
That’s interesting, because these three states traditionally have behaved a lot alike politically. Way back in 1948, the great historian V.O. Key referred to MS and SC as “the super-south”: places where the racial politics of the post-Confederacy trumped all other considerations. Alabama certainly qualified as a fellow member of the “super-south” during and after the racial conflicts of the 1950s and 1960s. And in the post-civil-rights era, SC, AL and MS invariably exhibited high levels of racial polarization between the two parties in ways that weren’t always characteristic of other southern states.
Perhaps SC’s recent economic growth, accompanied by a significant if not overwhelming number of transplants, has separated it a bit from the rest of the “super-south.” But if so, not by much, and there’s nothing there like NoVa’s large and essentially non-southern voting base, or NC’s Research Triangle concentration of “latte class” professionals and students.
So why did the white vote in SC go in exactly the opposite direction in 2008 as MS’s and AL’s? Nate thinks it may be because South Carolinians had a heavy personal exposure to Obama (and his family; his wife’s roots are in the state) during the primary campaign, and thus saw him as less alien than did southern white voters elsewhere in the Deep South.
No other plausible theory comes to mind. And if, as Nate hypothesizes, “familiarity erodes contempt” when it comes to a voting demographic most suspicious of Barack Obama, it’s a very good sign for his presidency.