:format(webp)/cdn.vox-cdn.com/uploads/chorus_image/image/50860467/502742138.0.jpg)
There are fewer newspapers more solidly Republican than the New Hampshire Union Leader, the daily newspaper in Manchester. But for the first time in a century, the paper’s editorial board isn’t endorsing a Republican for president.
Instead, the paper is urging readers to vote for Libertarian Gary Johnson, with an additional front-page message from its publisher blasting Donald Trump:
The man is a liar, a bully, a buffoon. He denigrates any individual or group that displeases him. He has dishonored military veterans and their families, made fun of the physically frail, and changed political views almost as often as he has changed wives.
Americans are being told that we have to choose the lesser of two evils. No, we don’t.
The Union Leader was anti-Trump in the primary and urged readers to vote for Chris Christie, and it didn’t matter. But Clinton’s lead in the polls is getting narrower in New Hampshire. The RealClearPolitics polling average for the state has her leading Trump by 5 points, down from an 8-point lead in mid-August.
And political science research suggests that a surprising endorsement — like, say, one of the most Republican editorial boards in the country refusing to endorse a Republican — could make a difference.
Newspaper endorsements matter when they’re a surprise
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/7057493/shutterstock_130253288.jpg)
According to a study published in 2011, unusual endorsements matter because it’s not what readers expected to hear. Newspaper endorsements change the most minds when they break with the usual pattern to endorse a candidate of the other party.
Two political scientists, Chun-Fang Chiang of National Taiwan University and Brian Knight of Brown University, studied the effect of newspaper endorsements in 2000 and 2004, using a survey that asked voters in the days leading up to the election about which newspapers they read and which candidates they preferred.
The researchers sorted newspapers on a spectrum based on how likely they were to endorse Democrats for president. They found that when Democratic-leaning newspapers endorsed Republicans for president, or vice versa, readers were slightly more likely to support the candidate the newspaper endorsed.
If newspapers endorsed the candidates that typically lined up with the editorial page’s ideology, though, they didn’t really convince anyone. The effects were greatest among people who had seen the endorsement, as you might expect, and among older readers, who were more likely to read the editorial page.
The Union Leader is the second newspaper in two weeks to break with tradition and decline to endorse Trump, after the Dallas Morning News endorsed Clinton on September 7. And there are still many large newspapers that endorsed Republicans for president in the past two election cycles that haven’t weighed in. While the newspaper industry’s influence has waned since 2004, it’s conceivable all this might make a slight difference.
