Obama won the lowest percentage of the white vote in the state with the highest percentage of black voters. That state was the ever-reliable Mississippi, where Romney made his famous “I like grits” comment. Thirty-six percent of the voters in Mississippi are black. Obama won a mere 10 percent of the white vote there.
Conversely, Obama won one of his highest percentages of white voters in the state with the fewest minority voters: Maine. Ninety-five percent of Maine’s voters were white, and 57 percent of them voted for Obama. That ties with one other state for the highest percent of whites voting for Obama: Massachusetts, where 86 percent of the voters are white.
In fact, Obama won the white vote only in states with small minority voting populations. The others Obama won were Iowa (93 percent white), New Hampshire (93 percent white), Oregon (88 percent white), Connecticut (79 percent white) and Washington State (76 percent white).
This is quite a curious phenomenon.
Curious, maybe. But not surprising. That the political behavior of whites depends on the size of the surrounding population of blacks has been well-known in political science for more than 60 years—at least since 1949, when political scientist V.O. Key wrote his magisterial Southern Politics in State and Nation.
Key found that the behavior of Southern whites in the so-called “black belt” was distinctive: in these areas where blacks made up a larger fraction of the population, whites were particularly focused on maintaining their own political power. This notion became known as “racial threat.” Eric Oliver summarizes the theory as it developed after Key:
…superordinate groups become more racially hostile as the size of a proximate subordinate group increases, which putatively threatens the former’s economic and social privilege.
To be clear, my point is not that the dynamics of the Jim Crow South obtain today. Nor am I suggesting that opposition to Obama is solely about racial hostility—although racial hostility does appear to play a role.
Rather, my point is simply that some familiarity with the political science canon would have helped Blow make sense of these data.