Pew Research has released its major study of the 2020 presidential electorate and there are a few interesting results even if the numbers largely confirm what was already known or suspected. For example, we can get a sense of how identity worked differently for Hillary Clinton and Joe Biden.
In 2016, Hillary Clinton lost white women by two points, and in 2020 Biden lost them by seven. By contrast, Biden lost white Catholics by 15 points, which seems disappointing, but Clinton lost them by 33 points in 2016. It’s a safe bet these numbers reflect a good number of voters who made their decisions primarily based on gender or religious affinity, and therefore they’re not indicative of contest-specific issues or long-term trends. This isn’t rocket science; if you want to do better with white women, choose a white woman as your candidate, and if you wan’t to improve with white Catholics, choose a white Catholic. Unfortunately, the identify difference between Trump’s two opponents makes it harder to figure out how issues or Trump’s performance in office may have affected these two groups.
It’s easier to see identity drift in the non-Catholic religious numbers. White evangelicals gave Trump a 61 point edge in 2016 and that jumped all the way to 69 points in 2020. Clearly, they were happy with his performance. Despite this, Trump’s advantage with regular church-goers (of all denominations) actually declined by two points.
Trump’s effort to makes whites vote with racial consciousness looks to have been successful. This is most clearly seen in the rural vote where his 25-point advantage in 2016 ballooned to 32 points in 2020. Despite this, white voters’ share of the Trump electorate declined from 88 percent in 2016 to 85 percent in 2016. White voters’ share of the Democratic vote actually went up by one point to 61 percent. There’s a lot of push and pull in these numbers which makes them difficult to interpret. The white share of the overall electorate was down two points in 2020, but suburban voters made up a much bigger percentage (55 percent) of Biden’s electorate than Clinton’s (48 percent). The overall impression is that the lower the education level, the more successful Trump was in getting white’s to identify with him (and the GOP) as their savior. But even this is complicated because the most straightforward measure doesn’t show it. The percentage of Trump voters with a college degree was actually up two points in 2020 and the share of non-degree white Trump voters was down by five points. That’s pretty much statistical noise that doesn’t indicate a major shift in voter preference. Still, education levels with Latinos were correlated with preference, with college graduates much more likely to prefer Biden.
The big shifts in the electorate show up clearly in the data. Biden won independents by nine points while Clinton lost them by one. Biden did substantially better than Clinton in the suburbs. Trump ramped up his rural and white evangelical support and made some inroads with less educated Latinos. He also did modestly better than in 2016 with voters under thirty.
The overall picture is favorable to Democrats because Trump’s strongest groups are aging out of the electorate and declining as a percentage of the electorate. But Biden saw some slippage with young voters and Latinos, and he also can’t depend on independents to favor him when he’s the incumbent since they tend to favor change.
The most concerning development is increased racial polarization, which is basically more white voters making decisions based on their own perceived racial self-interest. Another troubling development is the Republican Party’s increasing reliance on low education voters. This isn’t a problem in itself, but mainly due to how the Republicans go about winning their support–namely telling gigantic lies that are unconvincing to college graduates.
For Biden’s election prospects, the biggest warning sign is probably his reliance on independents in 2020. If he loses them in 2024, he’ll have to make it up somewhere else, and he could be close to maxed out in the suburbs. For the Republicans, the steady loss of natural voter share is a problem, but they also have to be concerned that a lot of their massive advantage with rural voters won’t transfer to a candidate who isn’t Trump.
It seems like both sides should be interested in eating into the other’s area of strength. For the GOP, this means improving in the suburbs and among low-education minorities. For the Democrats, it means reducing the rural disadvantage back down to pre-2016 levels. This ought to result in a reduction in the heat level of the culture wars, but it’s highly doubtful that either party will seriously pursue that strategy. Organizing and fundraising are highly dependent on boiling cultural issues, and cultural moderates are dead ducks in most partisan primaries.
Even if neither party will pursue an optimized strategy, someone has to win. I’d rather be the incumbent party with an ascendent base and a mostly friendly media than whatever shitshow best describes the Republicans.