Polls Aside, New Research Shows the Parties Are Deadlocked
Despite alarms of realignments set off by random polls, a long-term Pew study shows Democrats and Republican in constant balance with some exchanges of supporters and more polarization.
If you listen to a lot of the noise in this raucous election year, the two major parties are in a state of extreme volatility that could lead to strange results. An uprising of interest in third-party or independent presidential candidacies is based in part on the assumption that big percentages of Americans are sick of the Donkey-Elephant duopoly and crave something different. And there’s also a lot of talk about one or both of the big parties falling apart or losing a big chunk of their past constituencies. That’s been particularly true with respect to fearful Democrats who are regularly freaked out by polls showing Black or Latino or even young voters stampeding away from them and perhaps even into the arms of Donald Trump. Some Republicans also fear Trump is in the process of blowing up their party with his turbulent personality and heterodox policy views, abandoning more traditional conservatives who may stray into the Democratic coalition.
It’s a good time to get yourself grounded with a new Pew Research Center study of partisan identification over three decades (from 1994 to 2023). While this huge trove of data and analysis richly documents some of the demographic shifts that have been occurring just under the surface during the period in question, it also confirms the more basic reality that the two parties have been in relative equipoise for a long time. Yes, there has been a steady uptick in self-identification as “independent” over these years, but once you sort them into leaners (which isn’t hard since most so-called indies regularly vote for just one party), the stability is striking. In 1994, a big GOP midterm-election year, Republicans plus leaners were at 51 percent of the electorate and Democrats plus leaners were at 47 percent. In 2012, when Barack Obama was reelected, those numbers were precisely reversed. And in 2023, the Democratic coalition was at 49 percent and the Republicans were at 47 percent. Not a lot of movement at all.
There has, however, been some movement in the components of the two major-party coalitions, but typically it hasn’t affected the overall balance for any length of time. Pew notes a small drift away from Democratic affiliation among Black and Latino voters in recent years. But among the latter, Democratic percentage losses have been offset by growth in the size of the group. And as a separate Pew study of validated voters shows, Republican gains among Black voters shown in the 2020 exit polls (and more recently by all sorts of polls) may have simply represented a polling error (Pew shows Trump winning only 8 percent of the Black vote in 2020).
The more famous trends affecting the partisan affiliation of different education categories of white voters are validated by Pew, but they represent more or less a wash overall:
In the last years of George W. Bush’s presidency and the first year of Barack Obama’s, Democrats had a double-digit advantage in affiliation over Republicans among voters without a college degree. For example, in 2007, 56% of voters without a degree were Democrats or leaned Democratic, while 42% were Republicans or GOP leaners. This group was narrowly divided between the two parties for most of the next 15 years, but in the last few years it has tilted more Republican.
College graduates moved in the opposite direction, becoming more Democratic over this same period.
And while the “gender gap” between Republican-leaning men and Democratic-leaning women has been a regular feature of politics over the last three decades, it has changed in where it appears:
As recently as 15 years ago, there were sizable gender gaps in partisanship among both college and non-college White voters. In both cases, men were substantially more likely than women to associate with the Republican Party …
Today, there is no gender gap in partisanship among non-college White voters, while there is a gender gap among college graduate White voters. The difference in partisanship between White women voters who have a college degree and those who do not, in particular, is quite large.
By contrast, there is little variation in the partisanship of Black and Hispanic voters by these characteristics, and the relationship has varied less over time.
Partisan affiliation by age cohorts can get confusing as generations and sub-generations move through life, but Pew confirms that people born in the 1940s, 1950s, and 1960s overall lean Republican, while those born in more recent decades lean Democratic.
As is pretty well understood, the growing percentage of Americans with no religious affiliation lean strongly Democratic, along with certain minority religious groups (e.g., Jews, Muslims, and Black Protestants), while white Evangelicals lean heavily Republican and Catholics tend to mirror the general population with a slight Republican lean. In another Pew finding that won’t surprise anyone, voters in urban and rural areas have become steadily more polarized (the former in favor of Democrats, the latter in favor of Republicans) while the suburbs remain a closely divided cockpit. Also unsurprisingly, liberals have all but vanished from the Republican coalition, and conservatives are down to the mid-single digits among Democrats.
The persistence of ideological polarization has clearly fed and even intensified partisan polarization; this isn’t some artificial construct of the news media or party tacticians, notes Pew: “Republicans and Democrats do not just hold different beliefs and opinions about major issues, they are much more different racially, ethnically, geographically and in educational attainment than they used to be.”
So the bottom line is that while big changes in both polarization and in the partisan balance are always possible, you should look for a lot of verification before assuming one is underway right now or that any one election will create a new template. (After all, such allegedly “realigning” elections as 1994, 2008, and 2010 were shown to be no such thing by the immediately following elections.) It’s another matter entirely whether a potential 2024 presidential victor and his supporters could get tired of the burden of winning elections (particularly since their Republican Party has lost the national popular vote in seven of the last eight presidential contests) and decide to take our country in an anti-democratic as well as an anti-Democratic direction.