A raft of new polls out this week carried almost unanimously good news for Hillary Clinton, staking the Democratic presidential nominee to significant leads over Donald Trump. But there’s one potential warning sign in these polls should the race narrow: Clinton’s lead over Trump shrinks when voters are allowed to choose one of the major third-party candidates in the race.
Yet the Libertarian Party’s presidential ticket — composed offormer GOP Gov. Gary Johnson of New Mexico and former GOP Gov. Bill Weld of Massachusetts — appears to draw more from voters who might otherwise be aligned with Clinton, especially younger voters.
The same is true of Green Party nominee Jill Stein — though to a lesser degree, since Stein doesn’t earn nearly the same level of support as Johnson.
Here’s why Democrats should be concerned: As Trump’s support has dwindled in recent days — leaving the GOP nominee with just his fervent supporters — some soft voters might be moving into Clinton’s camp when asked on a two-way ballot, but defecting to a third candidate when given other options.
“Trump voters are mainly Trump voters, but Clinton voters are still not quite happy that they’re going to end up voting for her,” said Monmouth University pollster Patrick Murray, who has studied the role of third-party candidates in pre-election polls.
The extent to which Johnson and Stein are cutting into Clinton’s lead is small but fairly consistent. In the 15 major national polls conducted since the first night of the Republican convention last month that included both the two-way and four-wayballot tests, Clinton has had a smaller lead over Trump in nine, and her lead has been unchanged in five of them. In only in one of the surveys has Clinton’s lead actually increased when Johnson and Stein are included.
Most of those differences are slight: The average difference in Clinton’s margin is only about 1.4 percentage points. If Clinton can maintain a significant lead nationally over Trump, Johnson and Stein will have no impact on her general-election chances. But if the race tightens again, they could be a factor, especially in states with long histories of voting for third-party candidates, like Minnesota, Colorado, New Mexico, Nevada and New Hampshire — third-party presidential candidates have won more than 2 percent of the aggregated vote in those states since 2000.
Many of the national polls offer a good test of how the race changes with Johnson, and sometimes Stein, added. Take this week’s Fox News poll: Clinton leads Trump on a head-to-head ballot test, 49 percent to 39 percent.
But immediately after, respondents were asked the same question — only with Johnson included. In that case, Johnson took 12 percent, dropping Clinton’s share to 44 percent, and Trump’s to 35 percent.
It was similar in a CNN/ORC poll this week: Clinton posted a 9-point lead in the head-to-head matchup with Trump, 52 percent to 43 percent. But when Johnson and Stein are thrown in, Clinton’s lead slips to 8 points, 45 percent to 37 percent. (Johnson takes 9 percent, and Stein 5 percent.)
Why is Johnson (and Stein, for that matter) drawing more from Clinton when he is a former Republican in a year whenRepublicans are far more likely than Democrats to defect to the third-party candidates?
Young voters. In the Fox News poll, Johnson gets 19 percent of voters younger than — but only 13 percent among voters 35-54 and 6 percent of voters 55 and older said they would vote for the former New Mexico governor.
In the CNN/ORC poll, Johnson and Stein take a combined 21 percent from respondents younger than 45, but just 8 percent among voters 45 and older.
It all sets up a push-pull between Republican voters who aren’t on board with Trump, and younger voters who don’t like either candidate.
“The Libertarian is the stronger [of the third-party candidates], which is more likely to pull from the right,” said Democratic pollster Nick Gourevitch, whose firm, Global Strategy Group, is working for the pro-Clinton super PAC Priorities USA. “You have the young voter thing, which is more likely to pull from the left.”
But the way pollsters are structuring their questions could also influence the results. Murray, the Monmouth University pollster, said the way most pollsters are approaching the question — asking the Clinton-Trump matchup first, then offering either Johnson or both Johnson and Stein — might be inflating the third-party candidates’ vote shares.
“The evidence suggests that when you ask the four-person as follow-up question, it actually overstates the third-party vote share,” Murray said.
The irony for Johnson, who is trying to crack 15 percent in the polls to guarantee a spot in the first debate next month, is that he actually wants to flip the order of those questions.
“Right now what’s happening is the question is getting asked by all the pollsters: Trump or Clinton? That’s the question. Not Trump, Johnson, Clinton, but Trump and Clinton,” Johnson told the Los Angeles Times recently. “And then as the secondary, which is now happening in more cases than not, they add Johnson as an afterthought. And in that afterthought, we’re coming out at 12 percent. If they came in right at the top as Johnson, Trump and Clinton, I think it would be at 20 percent, and a lot of that has to do with the dissatisfaction with the two major candidates.”
Murray says the evidence indicates Johnson is mistaken. “That actually would hurt him,” Murray said. “When you ask it as a follow-up question, it gives people a chance who reluctantly said Clinton or Trump on the first question a chance to say they aren’t that happy with their choices.”
Johnson’s other complaint about the question order concerned the news coverage the polls receive. “The opening question is Trump and Clinton, and 99 percent of what gets reported is just that line,” he told the Times.
That may be the case. In all the national polls this week, the news organizations led with the result of their head-to-head matchup, with the third-party candidates included almost as afterthoughts.
Weld punted when asked Wednesday during a CNN town hall whether he and Johnson were comfortable with the possibility of pulling more voters from Clinton and handing the presidency to Trump.
“You know, at this point, that’s pure speculation,” said Weld. “We believe in ourselves and our ticket, and we’re voting Libertarian.”