(This is the second of three posts.)
But here is the paradox: Although whites with less formal education tend to be more racially intolerant, racial attitudes have typically had a weaker effect on their political views, compared with better educated Americans. Understanding that paradox, and how it has changed, is crucial to understanding the 2016 election.
At the root of the paradox is that whites with more education followed politics more closely and thus knew about long-standing divisions between the Democratic and Republican parties on civil rights and other policies related to race. Whites with less education, who tended to follow politics less closely, hadn’t fully learned this.
But the election of Barack Obama changed things. Once Obama was in office, even whites with less formal education were able to connect racial issues to partisan politics.
The graph below shows the trend in the percentage of whites who thought that the Democratic presidential candidate was more supportive of federal aid to African Americans than was the Republican candidate.
Across time, a large and stable majority of whites with a college degree believed that the Democrat was more supportive of federal aid to blacks. But among whites with no college degree, there was a substantial 22-point increase in awareness from 2004 to 2012. The election of an African American Democratic president helped shrink a different diploma divide — this time, in awareness of the two parties’ differing positions on race.
Here’s why this matters. During the Obama presidency, racial attitudes became more strongly connected to whether whites identified as Democratic or Republican. But those stronger connections were most visible among whites with less formal education.
The following graphs show the trend in Democratic Party identification among non-college-educated whites using data from the American National Election Study (ANES) and the Pew Values Survey:
The left-hand graph shows that non-college-educated whites who scored high in racial resentment (about 40 percent) were about 15 points less Democratic in 2012 than they had been in the pre-Obama era.
The right-hand graph shows a similarly sharp downturn in Democratic identification among non-college-educated whites who thought that discrimination against African Americans is rare (also about 40 percent) — a belief closely linked with racial resentment. There was no such drop-off among working-class Americans with more progressive racial attitudes; the effects of racial attitudes on college-educated whites’ partisan attachments were more stable as well.
Simply put, racially resentful whites without a college degree were most likely to flee the Democratic Party during Obama’s presidency.
And flee the Democratic Party they did. In an article titled “White Out” that was published just after the 2014 midterm election, Ron Brownstein noted that Democratic losses in the House during Obama’s presidency were most likely to occur in districts with lots of working-class whites.
There were 79 Democrats who represented mostly working-class white districts in 2009, compared with 15 in 2009. Those Democratic defections were also most likely to occur in racially conservative districts and among racially resentful voters.
Trump clearly did not create these patterns. But he effectively exploited them. His explicit appeals to racial resentment and white identity should have made it even clearer where the two parties stood on racial issues.
That partisan clarity over race seems to have accelerated the Obama-era trend of diminished Democratic support from racially resentful whites. In the final post in this series, I will present preliminary data showing that the effects of racial attitudes in the 2016 election were probably greater than their historically strong effect on Obama’s two presidential elections.
Michael Tesler is an associate professor of political science at the University of California at Irvine and the author of “Post-Racial or Most-Racial? Race and Politics in the Obama Era.”