But that polarization isn't limited to the politicians and what they do; instead, it's happening with all of us. As the political parties become less about regional differences or class differences, and more about ideological and values differences, the disparities between the parties and their members have gotten clearer. Moderate suburban northeastern Republicans and conservative rural southern Democrats are dwindling species, but it goes deeper than that; people are also less likely to pick and choose from an ideological buffet, and instead are becoming more and more down-the-line liberal or conservative on the entire spectrum of economic and cultural issues.
And it goes beyond that. With a more mobile society that's less rigidly based around family ties, people are increasingly free to move around and associate as they choose with like-minded people. That results in people increasingly sorting themselves along ideological lines in terms of who their friends are, what groups they join, even where they choose to live, all to enjoy the security of being around like-minded individuals (and avoid the cognitive dissonance of constant interactions with the oppositionally minded). It's not a totally new idea—it's been described well in many books, maybe most definitively in 2008's The Big Sort—but it's never been the subject of as wide-ranging and thorough (with a sample size of 10,000) a poll as the new study from Pew Research Center, "Political Polarization in the American Public."
Maybe the central finding of the Pew study is that Americans in recent years have tended to become much more "ideologically consistent." Pew asked respondents a number of policy and political values questions; in past years, people would have tended to take a more "a la carte" approach, maybe supporting a more populist economic policy but being opposed to abortion, for instance. Today, though, there's much less of that, as you can see in the graphic above, showing significant ideological overlap between the parties in 1994 and much less now, and showing the median Democrat and median Republican much further apart now.
In 1994, 10 percent of the population fell into either the "consistently liberal" or "consistently conservative" segments. That's more than doubled in 2014, to 21 percent. While 49 percent of the population took a roughly equal number of liberal and conservative positions, in 2014 that's down to 39 percent of the population.
Another way to look at that is in terms of where people fall in relation to the median members of each party. In 1994, only 64 percent of Republicans were to the right of the median Democrat, but in 2004, 70 percent were to the right of the median Democrat, and by 2014, 92 percent of Republicans are now to the right of the median Democrat. Meanwhile, in 1994, 70 percent of Democrats were to the left of the median Republican. That actually fell in 2004, to 68 percent of Democrats to the left of the median Republican, and by 2014, it had shot up to 94 percent of Democrats being more liberal than the median Republican.
There's more discussion, and more charts, over the fold ....
Watching the news in dismay from day to day, it may not seem like we're becoming more liberal, or that we're even going backwards. But that's not what Pew analysts see—they point out that the shift toward consistency in the last two decades has been greater among Democrats than among Republicans. In other words, the number of "consistently liberal" Democrats has grown from only 5 percent of Democrats in 1994, to 23 percent of Democrats in 2014, more than quadrupling. By contrast, 13 percent of Republicans were "consistently conservative" in 1994, and that's up to 20 percent of Republicans today. (Granted, that 20 percent still tends to be the loudest fifth of the party, which does a lot to affect our perceptions.)
You can see that at work when you compare the notoriously fractious Democratic Party of the 1980s and 1990s with Democrats in Congress today, who tend to put up a pretty united front with little Blue Dog resistance any more. On the other hand, that growth in consistency isn't so much about economic issues as it as about cultural issues. Pew attributes most of that to two issues that have evolved greatly over the last two decades: increased tolerance of gays and lesbians, and mostly positive sentiments about immigration.
One other thing worth noting is that being "consistently" liberal or conservative correlates with being more politically engaged. In other words, if you're a higher-information voter who pays a lot of attention to politics, you're more likely to commit fully to one camp or the other. The muddled middle who pick and choose aren't the best-informed, most-reasoned voters, despite what the Ron Fourniers of the world might tell us; instead, the pick-and-choose voters are the less engaged, less-informed ones. The middle-dwellers are also much less likely to vote or to give contributions than the ideologically consistent.
In fact, as you can see above, Pew has had to add a new category that goes even beyond "very unfavorable"—those persons who feel that the opposing party is a "threat to the nation's well-being." A total of 27 percent of Democrats and 36 percent of Republicans fall into that category. Among people who give political contributions, the existential threat numbers rise to 46 percent of Democrats and 54 percent of Republicans.
Interestingly, people seem to instinctively know what kind of community they can find their like-minded brethren in, as you can see in the graphic above. Seventy-seven percent of "consistently liberal" people (but only 21 percent of "consistently conservative" people) would like to live in a dense, walkable neighborhood. On the other hand, 75 percent of "consistently conservative" people, but only 22 percent of "consistently liberal" people, would like to live in a spread-out neighborhood where large houses are the main attraction. People seem to know in advance where they're likely to fit in, based on the types of people they already see living in those types of environments. That kind of knowledge only speeds up the "Big Sort," and the growing gulf between the cities, and the rural areas and exurbs.
Finally, Pew also asked questions about the nature of compromise, and how people feel about compromising in politics. For the most part, people are supportive of compromise, with 49 percent of all respondents saying that the ideal outcome is a 50/50 split in terms of each party getting what it wants. However, the more ideologically consistent one is, the less likely one is to accept 50/50; the average "consistently liberal" or "consistently conservative" person sees getting two-thirds of what they want as being ideal.
If you have a nagging sense that Democrats are hampered somewhat in negotiations with Republicans by being more willing to compromise, or at least instinctively feeling more comfortable with compromise ... well, you're right. When offered a choice between leaders who "are willing to compromise" and leaders who "stick to their positions," consistent liberals overwhelmingly prefer compromisers, 82-14. Consistent conservatives, however, prefer position-stickers, 63-32. (Operationally, rather than ideally, though, consistent liberals still prefer getting more of what they want. Sixty-two percent of consistent liberals still say that the ideal deal between Republicans and Democrats would give Democrats more of what they want.)
There's much more detail to the study, and I urge you to click through the whole thing to see more. However, there's one thing that Pew doesn't really cover, and that's why this polarization is happening. That may be unanswerable because it's a bit of a chicken-and-egg problem. Did the politicians sort themselves out first, and the people followed them? Or did voters start becoming more consistent first, and the parties followed the realignment from the grassroots up? And for that matter, how big a role did the mass media play in the shift? One of the defining characteristics of the period from 1994 to 2014 was the return of partisan media (something that was prevalent 100 years ago, but dwindled in the mid-20th century as the canon of "objectivity" took over).
There's one other important question about the study: What can we, as liberal activists, take away from it? The main thing that I would observe is that it really underscores the need, as has been said over and over at this site, to focus less on chasing elusive swing voters and more on activating base voters. There just aren't that many persuadables left, with more voters solidifying into the "consistent" camps. And moreover, the remaining swing voters are the ones who are the least informed and the least likely to vote, making them a riskier bet. Instead, especially as Democratic-friendly demographics (non-white and/or well-educated people) continue to be a larger and larger part of the population, the trick is to activate them and turn them into reliable voters.