2 political scientists have found the secret to partisanship, and it’s deeply depressing

VOX

2 political scientists have found the secret to partisanship, and it’s deeply depressing

Updated by Ezra Klein on April 24, 2015, 10:40 a.m. ET @ezraklein

Politics isn't about who you love. It's about who you fear.

That's the upshot of a draft paper by political scientists Alan Abramowitz and Steven Webster that attempts to untangle a mystery about modern American politics: how can there be record levels of party loyalty and straight-ticket voting at the same time that fewer Americans than ever before are identifying as Republicans and Democrats?

To answer the question, Abramowitz and Webster test a host of political characteristics to see what best predicts party loyalty. The real key, they found, was fear of the other party: "Regardless of the strength of their attachment to their own party, the more voters dislike the opposing party, the greater the probability that they will vote consistently for their own party’s candidates."

It's worth saying that a bit more clearly: you're more likely to vote Democratic if you hate Republicans than if you love Democrats, and vice versa. What parties need to do to keep you loyal isn't make you inspired. Rather, they need to make you scared.

Partisans like their own party less, but they like the other party a lot less

You don't actually need a regression analysis to see that hate, not love, is driving the changes in American politics. Abramowitz and Webster include this graph of people's attitudes toward their party, and toward the opposing party, and it gets more disturbing the longer you look at it:

What's changed isn't how much we like our own party. It's how much we hate the other party. (Chart: Abramowitz/Webster)

The thick line atop the chart shows people's attitudes toward their own party since 1980. The change there isn't dramatic, but it's noticeable: on net, people like their own party a bit less than they did in '80s and '90s.

Consider that for a minute: even as people have become more loyal to their political party, they actually like their party less. That's a very strange finding. But it's more than explained by the line at the bottom of the chart. That line shows people's attitudes toward the other party — and there, you see a sharp, sustained drop in favorability.

This, then, is the last 30 years of American party politics in a sentence: we like the party we belong to a bit less, but we hate the other party much more.

Parties thrive on their members' fears

This chart from Pew's massive survey around partisan polarization is a powerful companion to Abramowitz and Webster's work:

Pew

People often compare politics to sports. Both have teams, and winners and losers, and the different sides often hate each other. But here's where the analogy falls apart: in politics, many believe — often correctly — that if the wrong side wins, the consequences will be grievous. The other side isn't just disliked. It's feared.

What's more, politicians from both parties have an incentive to keep it that way. This chart also comes from Pew, and it's powerful: it shows that the more you hate the other party, the likelier you are to vote, and the likelier you are to donate money to candidates from your party:

Pew

How demographics drive partisan animosity

The question all this raises is why voters are coming to fear the other party more with each passing decade. The answer Abramowitz and Webster offer is that it comes down to changing composition: the two parties are becoming more demographically dissimilar from each other, and that's making it easier for them to hate each other.

Abramowitz and Webster note, for instance, "In 2012, according to the national exit poll, nonwhites made up 45 percent of Democratic congressional and presidential voters but only 11 percent of Republican congressional and presidential voters." Meanwhile, the Republican Party has become much more religious: in 1980, 48 percent of observant white voters were Republicans; by 2012, that had risen to 78 percent.

This is leading to sharper polarization on major cultural issues like abortion. "Between 1980 and 2012 the proportion of Democratic voters taking a strongly pro-choice position on abortion rose from 38 percent to 61 percent," write Abramowitz and Webster, while the proportion of Republicans "favoring a ban or stricter limits on abortion rose from 42 percent to 51 percent."

In other words, as the parties become more distant demographically and culturally, they become more distant on policy, too. And so voters of each party are being at least somewhat rational in their increased fear of the other party: a liberal Democrat actually has more to fear from the Republican Party of 2012 than she did from the Republican Party of 1980, and vice versa.

Why Republicans will dominate Congress and Democrats will turn out for Hillary

This theory leads to at least a few major conclusions about American politics. The one Abramowitz and Webster emphasize is that it gives Republicans an enduring advantage in congressional elections. Republicans have long had a geographical advantage in both House and Senate elections because their voters are more spread out than Democratic voters. In the past, Democratic candidate could compete by distancing themselves from the national party and making the election about local issues. But as the country grows more partisan, that's becoming harder, and the result is that Republicans are likely to be dominant at the congressional level for some time.

Jonathan Chait, meanwhile, pulls some good news for Democrats out of the data. Just as Republicans have a natural advantage at the congressional level, Chait believes Democrats have a natural advantage at the presidential level, where the voters skew younger and less white. And if fear of a Republican president means the Obama coalition persists after Obama, then it's going to be a long time before Republicans win back the White House.

Both of these lessons speak to the same underlying reality: as American politics becomes more partisan, and more based on fear of the opposing party, individual candidates matter less. So for all the talk of how much less enthusiastic Democrats are for Hillary Clinton than they were for Barack Obama, it's not likely to matter all that much because Democrats are going to be extremely enthusiastic about beating the Republicans, and vice versa.