How Third Party Voters And Non-Voters Could Shape The Election

Hillary Clinton and Donald Trump  may have enjoyed favorability bumps in the aftermath of their respective conventions, but the pair still has the highest disapproval ratings for presidential candidates since 1980 . Many registered voters are consequently voting against the candidate from the opposing party rather than for their own, according to Pew Research .

Unsurprisingly, there’s been an increase in two groups that could influence the general election outcome: third-party backers and non-voters.

Disaffected voters who refuse to support either Trump or Clinton are increasingly turning to third-party alternatives. Libertarian candidate Gary Johnson is currently polling at a level of support higher than third parties in the previous four election cycles: 8.5 percent, according to the HuffPost Pollster general election average . At this point in the 2012 election , Johnson polled at only 3 percent. The most popular third party candidate in 2008 and 2004 was Green Party nominee Ralph Nader , who averaged between 3 and 4 percentage points.

When Nader ran against former candidates George W. Bush and Al Gore in 2000, he had an average of 5.14 percent of support at approximately this point in the election. Some analysts attribute Bush’s victory to the third party presence, as Nader could purportedly have cost Gore decisive votes.

Johnson currently polls about 3 percentage points higher than Nader did. This creates the potential for an even greater influence on the election outcome.

Post-convention polling is relatively volatile, but data suggests that Johnson’s presence in the race pulls slightly more votes away from Clinton than Trump. A two-way race between the Democratic and GOP presidential hopefuls puts Clinton ahead by 5.9 percentage points, according to the HuffPost Pollster general election average. Insert Johnson into the race and Clinton’s lead shrinks to 5.2 percentage points less than a percentage point difference.

Johnson’s impact in the general election becomes more pronounced at the state level: He’s polling at high single digits in Florida, Ohio and Pennsylvania, according to July research from Quinnipiac University . Factoring in Green Party candidate Jill Stein, the third-party candidates collectively occupy between 11 and 13 percent of the vote share in those states. And according to the University of New Hampshire Survey Center , Johnson alone polls at 10 percent in the state. Stein is at 5 percent.

Nationwide, a Washington Post-ABC News poll isolated the impact of third-party candidates on two disaffected voting blocs: Democrats who supported Sen. Bernie Sanders (I-Vt.) and Republicans who supported a non-Trump candidate for the GOP nomination. About one in five (21 percent) of pro-Sanders Democrats indicated their vote would go to a third party. Fewer non-Trump Republicans (17 percent) indicated that they would support a third party.

Other Sanders Democrats and non-Trump Republicans might not show up to vote, as Sanders’ campaign uniquely appealed to a demographic that not only traditionally exhibits poor levels of voter turnout , but also one that Clinton typically struggles with : millennials.

Millennials, in addition to other disaffected slices of the population, may abstain from voting altogether. The average number of respondents who have checked “not voting” in all publicly available polls since March is 5.8 percent. During this timeframe over the past three election cycles , the average was much lower ranging between 1.52 to 3.57 percent. (The “not voting” question was not asked in 2000).  

An internal quirk within a Reuters/Ipsos survey  highlighted the significance of this increase late last month. It introduced a “Neither/Other” option that skewed poll results so significantly that Reuters/Ipsos removed the question and re-conducted the poll. “More recently, the ‘Neither/Other’ option appeared to lead to an underreporting of Clinton’s support,” said Cliff Young, pollster and president of Ipsos Public Affairs, which partnered with Reuters on the poll . The pollsters estimated the Clinton shortfall at 2 to 4 percentage points.

Although Reuters and Ipsos removed the question, the option to vote for “neither” candidate, in reality, exists. And the Reuters/Ipsos poll quirk indicated that voters’ natural propensity to choose this option might also unfavorably impact Clinton on the whole.