When I was a freshman in undergrad — lo those many years ago — fulfilling my breadth requirements, I had to sit through a Humanities Core lecture in which the professor explained that the platform of the Republican party is racism — which was (1) so obvious to him that it required no support, and (2) so obviously evil that he simply knew that he’d undone all the brainwashing his young charges had endured under their parents’ tyrannical rule.
Like it or not though, Ron Unz notes, today’s Republican Party does indeed constitute the “white party,” drawing almost all of its national votes from whites, while the Democratic Party serves as the “mixed party,” with roughly comparable support from whites and non-whites.
Therefore, white support for Republicans, particularly at the national level, may serve as a reasonable proxy for a state’s apparent degree of “white racial consciousness,” whether implicit or explicit.
Under the “Sailer Hypothesis,” white alignment with the Republicans should be heavily influenced by the white share of the population, with the residents of lily-white states exhibiting little racial consciousness, while those living in states in which whites have slender or non-existent majorities would tilt much more heavily Republican. A second possibility to consider might be called the “Hispanic Hypothesis,” in which the heavy influx of Hispanic immigrants, both legal and illegal, pushes whites toward the harder-line Republicans; since the vast majority of today’s Hispanics come from a relatively recent immigrant background, a state’s overall Hispanic population can be used as a good approximation for this independent variable. Finally, there is the “Black Hypothesis,” in which the long history of black/white racial conflict is assumed to be the primary factor, and the percentage of blacks in the local population is what generally influences white political behavior.
For the sake of simplicity and to minimize the confounding impact of local political issues and personalities, the easiest output variable to examine would be the percentage of the white vote that supported the Republican presidential ticket over the last 20 years. On a population-weighted basis, the correlation results for elections from 1992 through 2008 across the 50 states are as shown in the chart below.
The results seem conclusive. The correlations between the Hispanic percentage of each state and white voter preferences are approximately zero for all presidential elections, implying that the presence of large Hispanic populations appears to have virtually no impact upon white political alignment, either one way or the other.
By contrast, the evidence for apparent black/white racial conflict being the driving force that prompts whites to vote Republican seems very strong: the correlations between the size of the black population and the degree of white GOP support range from 0.43 to 0.70, with a mean of 0.55, being both quite substantial and very consistent over time.