A certain conventional wisdom has set in about the 2016 presidential race, particularly among those who don’t find the prospect of a Donald Trump presidency comforting: The electoral map highly favors Hillary Clinton, and she’s quite likely to win, unless a major terror attack hits between now and November. If that happens, all bets are off, and Trump could well bluster and fearmonger his way into the White House.
Some big names have made various versions of this argument. In Vanity Fair, Nick Bilton wrote last month that “national support for Donald Trump is hovering around 41 percent. But, you can bet that if any significant national-security threat is posed between now and November, Trump’s poll numbers will soar in the same way that Bush’s did in 2001.” New York Times columnist Thomas Friedman expressed similar certitude in a March column, declaring that “if, God forbid, there is a major terrorist attack on our soil between now and Election Day, Trump will reap enormous political benefits.”
Sunday morning’s massacre at Pulse nightclub in Orlando — in which 29-year-old Omar Mateen, who said he was inspired by ISIS, killed 49 people and wounded many others — seems to have ignited this discussion a bit (Financial Times: “Orlando massacre will boost support for Donald Trump”). It was the first terror attack to have occurred since both candidates secured their near-locks on being their respective parties’ presidential nominees, after all, and it has dominated campaign coverage over the last two days, with each candidate taking what feels like a characteristic approach to responding: Trump by publishing a bizarre tweet accepting congratulations “for being right on radical Islamic terrorism,” by re-upping on his proposal to ban Muslims from entering the United States, and suggesting, a bit obliquely, that Obama may have been complicit in the Pulse attack; Clinton by delivering a rather stateswoman-like speech calling for national unity and emphasizing the peacefulness of most Muslims.
Whether or not the Pulse attack ends up being seen in exactly the same light as other major terror attacks (as opposed to the different, if overlapping, category of “mass shooting”), it’s worth asking: Is it correct to believe that Trump benefits from such violence electorally, based on what we know about political science, and from the salience of terrorism as a campaign issue more broadly? The answer: It’s a lot more complicated than many pundits would have you think, and there are good reasons to think, given the specifics of these two candidates, that even a major terror attack wouldn’t lead to a surge in Trump’s general-election chances.
That’s not to say it’s a crazy idea on its face — in fact, there’s some sound underlying logic to it. Researchers have found that, all else being equal, when the threat of terrorism is salient, candidates who are male and/or Republican and/or incumbent seem to benefit over those who aren’t. Voters seem to trust them more to handle these situations. Trump has two of those three labels working for him. (While incumbency is very important to this question broadly speaking — political scientists speak of a “rally around the flag” effect that has benefited incumbent presidents, perhaps most notably George W. Bush — it isn’t relevant in this case since both candidates are trying to attain the office for the first time.)
But Elizabeth Zechmeister of Vanderbilt and Jennifer Merolla of the University of California, Riverside, two political scientists who literally wrote the book on this subject, Democracy at Risk: How Terrorist Threats Affect the Public, tell Daily Intelligencer that in the specific case of Clinton versus Trump, it isn’t exactly clear which candidate, if either, would accrue “benefits” from a terror attack.
“We do find that in our research on this there’s a general tendency for the public to orient themselves toward male Republican and incumbent leaders in times of terrorist threats,” says Zechmeister. But the combination of Clinton’s experience as secretary of State and Trump’s utter dissimilarity from any other presidential candidate in memory make the question a bit foggier in this specific instance, particularly in light of some of the research Merolla and Zechmeister have conducted on the subject of gender, party, and perceived trustworthiness during periods of mortal threat.
Two studies they’ve conducted, both of which touched upon voters’ evaluations of Hillary Clinton specifically, stand out here: one from 2011 and the other, drafted but as yet unpublished, based on data from 2012. The 2011 article (PDF), in which Zechmeister and Merolla, working with co-author Mirya R. Holman of Tulane, drew on data from 2005 in which college students were asked their views on Condoleezza Rice, Clinton, George W. Bush, and John Kerry. Invoking a sense of threat by having students read a news story about terrorism caused the students to evaluate Rice higher, Clinton lower, and Bush higher (by about as much as Rice), as compared to a control group, and had no statistically significant effects on how they evaluated Kerry. As Zechmeister pointed out, at the time Clinton had effectively no experience in foreign-policy issues. Rice, on the other hand, was secretary of State. “We interpreted this result as evidence that Rice’s partisanship and experience inoculated her to gender bias in this context,” she said.
Clinton herself ended up being a test case for the proposition that experience can ameliorate the effects of gender bias in this context. In 2009, she was named secretary of State, a position in which she served until 2013 — a period when the nation dealt with some bruising foreign-policy challenges connected to the disintegration of Libya and Syria and the rise of ISIS. So when Zechmeister and Merolla were examining a new set of nationally represented survey data collected in 2012, they kept in mind just how much foreign-policy experience Clinton had gained since the student sample from 2005.
“Our expectation had been that, since between 2005 and 2012 she gained experience as secretary of State, she would be somewhat inoculated by her experience against” the penalties she’d face for being a woman and a Democrat when terror threats were salient, says Zechmeister. That was, in fact, exactly what they found, in a study they haven’t submitted for publication yet which covers respondents’ evaluations of both Clinton and Representative Nancy Pelosi: “We find that ratings for Pelosi fall to a greater degree than ratings for Clinton and, further, that while Clinton’s ratings are slightly lower in the terror-threat condition compared to the control condition, the difference is not statistically significant,” Zechmeister explains. “In short, these results from the 2012 study are in line with the expectation that experience can counteract a decrease in evaluations for certain Democratic female leaders in times of terror threat.”
So that’s the Clinton half of the race: There’s less reason to think terror threats will cause voters to discount her now than there was before she was experienced on this front. (While it goes without saying that some voters are highly critical of Clinton’s handling of Libya and Benghazi, these tend to be solidly Republican voters, and this conversation is about how terror attacks could change minds.)
What about Trump? Here things get difficult, simply because he’s such an unusual candidate. “It does matter who she’s running against,” says Zechmeister of Clinton. “It is the case that in our research to date, we haven’t had a hypothetical matchup of someone like Clinton against someone like Trump. We typically look at these characteristics that individuals have while keeping everything else constant … but in the real world, not everything else is constant. He is quite a different candidate than we’ve researched before.” It’s not the norm, in other words, for a candidate to respond to a mass shooting by pointing to his past proposal to effectively ban members of a major world religion from entering the country.
Merolla sounds a similar theme in her emails to Daily Intelligencer. In theory, she says, Trump “does exhibit some of the qualities that people look to in times of terror threat: bold, assertive, aggressive leadership.” But on the other hand, general-election voters aren’t likely to respond in a uniform way to his inflammatory approach, especially given the level of political polarization at the moment. Things were obviously quite different after the San Bernardino shooting, which was when Trump first rolled out his proposed Muslim ban. At the time he was seeking to consolidate support among GOP primary voters, who are much whiter, more conservative, and more anti-Muslim than the general electorate. Trump hasn’t really moderated his approach appreciably since then.
As a result, says Merolla, there’s a chance he could squander some of the benefits on the subject of terrorism that he would be “expected” to enjoy given his gender and party affiliation. “It is hard to say for sure if Donald will get the same type of boost that we have found for other Republican male candidates, since he lacks incumbency status as well as foreign-policy experience, and it is not as clear if he will be as advantaged by party stereotypes, given that he is a party outsider,” explains Merolla. “It will likely end up being a more nuanced story where he will get a boost among some voters concerned about terrorism, but will not see gains among others (for example, Democrats, politically sophisticated independents, and Republicans).”
All of which points to the dangers of making sweeping assumptions about which candidate will benefit from as emotionally resonant a subject as terrorism. “We are a complicated people,” says Zechmeister. “We don’t tend to go to the polls or go into evaluating candidates thinking just about one trait that that person possesses.”
