When Donald Trump delivers his first State of the Union, he will do so as one of the most unpopular first-term presidents in modern history.
According to polling averages compiled by FiveThirtyEight, a staggering 56.3 percent of Americans disapprove of Trump’s job performance. If you look at this moment in past presidencies, disapproval ratings range from 45.7 percent (Barack Obama) to 14.1 percent (George W. Bush in the wake of 9/11) to 33.1 percent (Bill Clinton) to 41.7 percent (Ronald Reagan) — none even coming close to Trump’s figure.
And yet Trump has nonetheless reshaped American public opinion across a number of topics. Political scientists have known for decades that political elites like presidents help form public opinion, especially among supporters, and evidence has recently begun to accumulate suggesting that people’s opinions follow those of leaders they respect more often than vice versa. That is, it’s likelier for voters to follow elite opinion than for politicians to follow public opinion.
Nowhere has this been more evident than in views on Russia. Trump’s presidency has seen Republicans following their leader and warming to Russia and Vladimir Putin, as Democrats report more strongly negative views.
I asked Joe Williams, a senior political analyst at YouGov, to pull all the organization’s polling of Republicans, Democrats, and the public at large on Russia and Putin from 2014 to the present, to show how thinking on the issue evolved as Trump and his ties to Russia gained prominence.
The trends we found are stark. Here’s what has happened to Putin’s favorability rating among Democrats and Republicans during the campaign and Trump’s early presidency:
(You can see the complete data in this spreadsheet, including some surveys left off the above chart.)
In the last pre-campaign survey in December 2014, the American public as a whole, still shocked by Russia’s aggression in Ukraine and the downing of Malaysia Airlines Flight 17 that spring and summer, was very anti-Putin. Only 11 percent of the public had favorable feelings about the Russian leader, including 8 percent of Republicans.
But by September 2016, as Russia’s favoritism for Trump became more and more obvious, Republicans started to flock to Putin. By December 2016, it was a major trend, and as of this past July, the last time YouGov asked, Republicans were still substantially likelier to have favorable opinions of Putin.
You get a similar result if you ask Democrats and Republicans if they consider Russia an ally, friendly to the US, unfriendly to the US, or an enemy (full spreadsheet here). The share of Republicans saying “ally” or “friendly” has shot up since 2014:
YouGov has been asking this as a weekly tracking question since February 4, 2017, enabling us to get a granular sense of how specific events in the Trump-Russia investigation have changed how the public sees Russia.
Take, for instance, Trump’s firing of FBI Director James Comey on May 9, 2017. This was a particularly dramatic move, one that occurred the same week that Trump revealed classified information to senior Russian officials. A week after Comey’s firing, the New York Times revealed that the FBI director had detailed memos on his meetings with President Trump chronicling Trump’s attempts to pressure Comey to shut down the FBI investigation of former National Security Adviser Michael Flynn. The day after that, Deputy Attorney General Rod Rosenstein appointed Robert Mueller as special counsel to investigate Trump’s ties to Russia.
If you follow the tracking polls, what you see is that over that period, favorable feelings toward Russia remained static among Democrats but shot up among Republicans as the issue took on more salience, and Trump was increasingly associated with the country:
It’s not just YouGov. You can see the Republican détente with Russia in just about every organization’s polling. Consider this Gallup table comparing views of Putin in February 2015 to those in February 2017. In 2015, Republicans liked Putin less than Democrats did. By 2017, a huge gap had opened up, with Republicans on the more pro-Putin side:
As the year dragged on, Putin’s popularity deteriorated among both Democrats and Republicans:
But the gap remained, and Democrats still registered a favorability rating one-sixth that of Republicans.
This is especially striking since, for the past decade, Democrats have generally been more friendly to Russia than Republicans have been. Pew Research Center found that Republicans were likelier to name Russia as an “adversary” of the United States than Democrats were — until late 2016 when Trump’s closeness to Russia flipped the partisan valence of the question:
“It’s pretty historic,” Gabriel Lenz, a professor at UC Berkeley who studies how political elites influence public opinion, says. “The percentage who think of Russia as an adversary … the parties have been aligned on that for 60 years or something.”
It’s not just Russia
Trump’s closeness to Russia is one of the most striking reversals of Republican Party policy under his tenure, and thus the one that has effected the most dramatic pivot in public opinion among his followers and detractors.
But it’s hardly the only example. While elite opinion on trade had been polarized well before the 2016 race — with Democratic presidential candidates promising deals with more protections for US workers (recall Obama’s 2008 promise to “renegotiate” NAFTA) and Republicans toeing a more orthodox pro-trade line — public opinion, to a surprising degree, was not. Trade has always been a controversial issue, but roughly equal shares of Democrats and Republicans supported it. Pew Research Center polling found that alignment began to break down when Trump became a political force:
NBC News/Wall Street Journal found the same thing, with data going up to February 2017:
Within the past year, the trend has abated somewhat, as Republicans have become more pro-trade and started to close the gap with Democrats. That might have something to do with the low salience that trade policy has had in Trump’s first year. In the aftermath of major tariffs the Trump administration recently announced, it’s possible the trend will reverse:
Attitudes toward the news media have also become polarized. To some extent, this is unsurprising; the out-of-power party typically has a more positive view of the press than the in-power party. This was true under Reagan (when Democrats liked the press more), Clinton (when Republicans did), and Bush II (when Democrats did again), though notably not as much under Obama.
But Pew has found the biggest gap between the parties even under Trump, a result no doubt caused in part by Trump’s frequent denunciations of the “fake news” media.
Gallup in August 2017 revived a question they last asked in 2003 about whether the media gets the “facts straight.” Rather than seeing continuity with the Bush years, when Democrats were out of power and could be expected to trust the media more, what they found was a massive increase in Democratic trust and decline in Republican trust:
Because of when the polling was conducted, it’s impossible to isolate the effect of Trump on each party’s answers here as opposed to the later Bush years, Obama, other factors, etc. But it’s hard to imagine, given how profoundly Trump has reconfigured opinion on Russia and trade, that his media bashing has had no impact.
Trump even shifted national polling around the NFL when he condemned anti-racist protests by players. A Morning Consult poll found that on September 21, before Trump jumped into the controversy, “25% of Trump supporters said they had a very favorable view of the NFL and 11% had a very unfavorable view.” By September 28, after he waded in, those numbers had dramatically “changed with 33% of Trump supporters [reporting] they have a very unfavorable view of the NFL and 16% report having a very favorable view.”
On Israel, too, opinions have shifted under Trump. As recently as May 2016, 43 percent of Democrats told Pew Research Center that they primarily sympathized with the Israelis in the Israeli-Palestinian conflict. By January 2017, the share was down to 33 percent. As of January 2018 it’s down to 27 percent, as Republican support for Israel continues to reach higher:
It’s not clear if this is because of Trump personally or simply a continuation of a trend that existed before him. Either way, Trump is overseeing a massive polarization of the parties on the conflict.
The political science behind why voters adopt politicians’ opinions
Trump is hardly alone in wielding this kind of influence. Political scientists have found that elite opinion can have profound effects among a politician’s supporters.
Lenz, in his book Follow the Leader?, documented a number of cases where a candidate’s supporters adopted the candidate’s view over the course of a campaign. For instance, in the 2000 election, he found no evidence that voters changed their minds based on candidates’ views on Social Security privatization, even though that was the dominant domestic policy issue during that presidential race.
What he did find evidence for is voters changing their views on privatization to align with their preferred candidate. “When citizens became more favorable toward Bush early in the 2000 campaign, for instance,” he explains, “they later switched to support his policy of investing Social Security funds.”
This same phenomenon was powerfully illustrated in recent experimental work by Stanford’s David Broockman and Washington University in St. Louis’s Daniel Butler. Broockman and Butler got a number of state legislators to randomly send out letters to constituents; some explained a policy position of the legislator in detail, some only briefly and with minimal justification, and the others were neutral controls. They also surveyed the constituents receiving the letters, both before and after. They found that receiving letters often led constituents to adopt their representative’s opinion, even when they had disagreed before.
As my colleague Brian Resnick notes, this finding has held up even in studies involving Trump. Brigham Young University political scientists Michael Barber and Jeremy Pope polled 1,300 people shortly after Trump’s inauguration, seeking to find if Republicans would be likelier to support liberal policies — like a higher minimum wage, background checks for gun owners, less restrictive abortion laws, or the nuclear deal with Iran — if they were told Trump supported those liberal policies. About 200 respondents were given this treatment, where they were told Trump backed a liberal idea.
“On average, across all of the questions that we asked, when presented with a liberal policy, Republicans became about 15 percentage points more likely to support that liberal policy,” Pope told Resnick.
Political scientists disagree about how disturbing this kind of finding is. While many view it as evidence that politics is, in essence, tribalistic and disconnected from actual policy ideas, others, like UChicago’s Anthony Fowler, have argued that voters are more rational and more likely to determine which politicians they support based on policy (rather than the other way around) than the most cynical interpretations of the data suggest.
“Suppose the voter selects her preferred party based on issues of importance to her like tax policy, health care, and women’s rights,” Fowler writes. “Then, a new issue arises like Social Security privatization. … Then, after she watches the Bush-Gore debate, she learns where the candidates stand and thinks about the issue for the first time. … If the voter knows she agrees with Gore on the issues important to her, she can probably trust his position on this new issue.” So even if she adopts her candidate’s preferred position, where she previously had none, Fowler argues that the voter could be acting rationally.
But whether you interpret followership cynically or optimistically, it’s a fact of political life in the US.
And it may be a problem for a politician like Trump, whose follower base is dwindling.
Trump’s unpopularity really matters, and it works against him
Despite the shifts documented above, Trump doesn’t win over his followers on every issue. And in general, he might be making the public less sympathetic to his views rather than more so.
Even when Trump does convert followers to his point of view, that doesn’t mean that he’s converted the public at large. There are more independents and Democrats than there are Republicans, so if independents and Democrats are both getting more anti-Russia and Republicans are growing more pro-Russia, then that would mean Russia is getting less popular overall.
Furthermore, when a president is unpopular, people are less likely to tell pollsters that they’re a member of that president’s party. That can bias the president’s apparent popularity among his party upward, since pollsters are then only asking people who are sufficiently committed that they haven’t stopped identifying with that party.
Trump is, of course, unusually unpopular, and sure enough, you can see that unpopularity having a pronounced effect on overall public opinion on specific issues. My colleague Matt Yglesias refers to it as a “reverse Midas touch.”
Take Obamacare, for example. For much of President Obama’s second term, the bill’s popularity was underwater. In April 2016, for instance, the Kaiser Family Foundation found that 49 percent of adults viewed the bill unfavorably compared to only 38 percent viewing it favorably.
Since Trump took office and the Republican effort to repeal the law began, this has reversed. The bill has consistently had higher approval than disapproval, with polling this month finding a 50 percent favorable to 42 percent unfavorable spread.
In December, Pew Research similarly found that for the first time in its polling, more Americans say the law has helped the country than say that it’s hurt:
“Much of the increase in positive views of the impact of the Affordable Care Act — and support for the law — has come among Democrats,” Pew’s Hannah Fingerhut writes. “Opinions among Republicans and Republican leaners have shown less change.” The Trump presidency then has mostly served to entrench pro-Obamacare sentiment among Democrats while not reducing it enough among Republicans to compensate.
Perhaps the most surprising place where this is true is immigration. On that issue, Trump hasn’t really polarized the public so much as he’s united it in opposition to restrictive immigration policies.
UPenn political scientist Daniel Hopkins was noticing this as early as during the campaign, in a piece for FiveThirtyEight. He found that even comparing January/February 2016 to October 2016, you could see Democrats, Republicans, and the mass public alike moving in a pro-immigrant direction:
UC Irvine’s Michael Tesler similarly documented a decline in public support for Trump’s border wall in August 2016, arguing that the campaign itself was enough for Trump’s views to turn off the American public. The trend has continued since then. Quinnipiac, for instance, found that the share of Americans opposing a border wall has grown from 55 percent in November 2016 to 63 percent in January 2018:
Quinnipiac also found that while 48 percent of voters in 2014 and 55 percent in 2015 supported a pathway to citizenship, by September 2017 the share had shot up to 68 percent.
Fascinatingly, it appears that Republican opinion on immigration is either holding steady or becoming more pro-immigrant, contrary to Trump’s views. An NBC poll conducted in August showed Republicans warming to immigration:
Pew, while finding that the partisan gap on immigration has expanded, also finds Republicans warming to immigrants since 2014 or so, just not as rapidly as Democrats:
Again, there remain big gaps between the parties on how to handle DREAMers and the border wall. And while the public will often say in the abstract that they think immigration strengthens the country (and the trends in that question do matter), it’s clear that anti-immigrant sentiment is still quite common nonetheless.
But it’s hard to escape the conclusion that the American people have become more pro-immigration under Trump, not less.