Typically, at this point in a presidential election cycle, punditry is focused on campaign tactics and delegate counts — who is advertising where, who is endorsing whom. But the unexpected rise of Donald Trump has scrambled the usual rituals and led to a wave of commentary about a much more fundamental question: the future of the Republican Party.

To know whether Trump’s rise is a one-off or the opening act of a broader shift in American politics, though, we need to understand what’s motivating his supporters. Do they back him because they share at least some of his positions on issues or for other reasons, like his tone, his fame or his status as an outsider? Whether Trump has the potential to realign American politics hinges on the answer.

Already, we know a fair bit about Trump supporters. Demographically, they are often white, male and without college degrees. They are disproportionately drawn from the ranks of registered Democrats who vote like Republicans. (Or else they’re named Chris Christie.)

On Trump supporters’ attitudes and issue positions, however, there is less agreement. Trump captured headlines last June with an announcement speech that called immigrants coming from Mexico “rapists,” and for a time, immigration was the only issue on his campaign website. It’s not surprising that analysts have linked Trump support to anti-immigration attitudes as well as prejudice, authoritarianism and populism.

In a recent post, Sean Trende and David Byler push back against explanations of Trump that focus solely on prejudice: “There is also a strong strain of anti-elite sentiment in the country right now, and Trump taps into that.” What’s more, there appears to be an economic underpinning to Trump support. As John Sides and Michael Tesler show, people who are dissatisfied with their financial situation are more likely to back Trump as well.

Still, there is a well-known challenge in figuring out which comes first. Do people gravitate to candidates who share their political views, or do they adopt the political views of the candidates they already back for other reasons? We can get around that conundrum using the Institute for the Study of Citizens and Politics panel survey that Diana Mutz and I have been conducting along with colleagues. The panel has followed the same American adults since late 2007, with the most recent survey having wrapped up in early February of this year, shortly after the Iowa caucuses.

The panel gives us an unparalleled look at Trump supporters’ attitudes long before they even knew Trump would run, whether in 2007, 2008 or 2012. To make things simple, I recoded every measure to vary from 0 to 1, and compare the 250 Trump supporters to 109 Ted Cruz supporters and 78 Marco Rubio supporters. Trump’s overall support in the 2016 wave of the panel is 40 percent among Republicans, which tracks his national polling reasonably well.

Stance on a position ranges from 0-1, with 0 being totally against and 1 being totally in agreement with


What’s the image of Trump supporters that emerges from the panel? Back in late 2007, they rated themselves as much less conservative than Ted Cruz’s supporters, at 0.64 compared to Cruz supporters’ 0.76. But there are well-known limits to ideological self-placement, and if we focus on it alone, we risk missing much of what makes Trump supporters stand out. So let’s instead go issue by issue.

First, consider economic issues. Some have pointed to Trump’s support for Social Security and his on-again, off-again support for a health care mandate to suggest that he is something of an economic populist. But as these results make clear, back in 2012 Trump supporters were only slightly more supportive of government spending generally, at 0.20 compared to Cruz and Rubio supporters’ 0.15. (Keep in mind that 0 means “government should provide fewer services; reduce spending a lot” while 0.5 means the status quo.) Also, Trump supporters were only a touch less likely to support repealing the Affordable Care Act, at 0.76 compared to about 0.80 for Rubio and Cruz backers. Similarly, on raising taxes on the rich, they were opposed and basically indistinguishable from backers of other GOP candidates. In all those cases, Trump supporters look more like other Republicans than like New Dealers.

On social issues, the differences are more noteworthy. Trump backers were far more pro-choice than Cruz or Rubio supporters, at 0.63 (0.67 is equivalent to agreeing that “abortion should be available, but with stricter limits”). That’s not far from Clinton backers’ 0.73. By contrast, Cruz supporters’ 0.39 puts them closer to the view represented by 0.33: “Abortion should not be permitted except in cases of rape, incest, or when the life of the woman is at risk.” In that light, maybe Trump’s defenses of Planned Parenthood make more sense. Trump supporters were also a bit more supportive of gay marriage than Cruz supporters, although the difference isn’t nearly as pronounced.

Trump has also attracted attention for dissenting from the hawkish foreign policy views that are dominant in the GOP. For instance, he drew ire from GOP foreign policy elites after accusing former President George W. Bush of having lied about weapons of mass destruction in the run-up to the Iraq War. On the question of whether to remain in Iraq as of late 2007, Trump supporters scored a 0.62, which is relatively close to the 0.50 position (“The U.S. should set a deadline for withdrawing its troops”). That makes them less favorable toward the Iraq War than Cruz supporters (0.81) and closer to Rubio supporters (0.67). But when asked in late 2008, the future Trump supporters were slightly more likely to call themselves “hawks” than Cruz backers (0.68 versus 0.67), and markedly more likely than Rubio backers (0.52). Rubio presents himself as the most hawkish of the three candidates, but his base of support didn’t describe themselves that way. As political scientist Elizabeth Saunders has noted, 2016 is no exception to the rule: Voters do not seem to be picking their candidates based on their foreign policy views. Foreign policy has been a prominent point of division in the GOP debates, but those divisions don’t seem to be reflected in public support.

That brings us to a pair of issues that have defined Trump’s candidacy: trade and immigration. In his announcement speech, even before mentioning immigration, he talked about trade: “When was the last time anybody saw us beating, let’s say, China in a trade deal? They kill us. I beat China all the time. All the time.” And indeed, Trump supporters were far less sanguine about NAFTA in late 2007 than Cruz or Rubio backers. They score an 0.40 on NAFTA support, a far cry from the 0.50 among Cruz’s supporters or the 0.52 among Rubio’s. And on immigration, Trump backers are markedly less likely to have favored a pathway to citizenship in 2012 — 0.21 compared to 0.29 for Cruz backers or 0.37 for Rubio supporters. So on those signature issues, Trump supporters are at odds with other Republicans — and to some extent with the leaders of the GOP.

These results also reaffirm what others have pointed out about white Trump supporters’ levels of prejudice: They are higher than those of Cruz or Rubio supporters. For our 2012 measure of white-black prejudice, white respondents were asked to assess whites and blacks on different stereotypes such as intelligence and work ethic. We then subtract people’s views of whites from their views of blacks, so that 0 indicates someone who endorses negative stereotypes only about whites while 1 indicates someone endorsing negative stereotypes only about blacks. A 0.50 indicates a respondent who rates whites and blacks equally on average, while the 0.58 of Trump supporters indicates markedly more positive ratings of whites relative to blacks. Is prejudice among the distinguishing attitudes of Trump backers? In a word, yes.

But that said, one word isn’t enough to summarize a whole set of motivations. In the U.K., people who are concerned about immigration levels also tend to be politically disaffected: They see elites’ support for immigration as a prime example of politicians’ abandonment of working-class constituents. That mix of attitudes helped vault the anti-immigration, Euro-skeptic United Kingdom Independence Party to the political stage in recent years. There is some suggestion that the same dynamic is at work in the U.S. At 0.70, Trump supporters were notably more likely to say in 2012 that “at present I feel very critical of our political system” than were future Cruz or Rubio backers (0.64 and 0.62).

What’s more, if you take the 40 percent of the GOP electorate that backs Trump and multiply it by the 30 percent or so of the electorate that identifies as Republican, you get around 12 percent of the U.S. population whose first choice is Trump. That’s just one percentage point off the 13 percent that UKIP won in the May 2015 general election.

Do Trump supporters stand out for their levels of prejudice or their concerns about unauthorized immigration? Yes. But those are not their only defining features. They are also more pro-choice than other Republican primary voters. And their economic populism seems to be focused on trade and not on government spending. Down the road, we may well view the 2016 election as an aberration. But Trump’s rise is more than a celebrity trading on his publicity. One of the key ingredients of a political realignment — a split within one party on a durable, straightforward set of issues — is now in place.

Listen to the latest episode of the FiveThirtyEight politics podcast.