Government & Politics

Do You Discriminate Along Party Lines?

A Stanford study hints at the broad consequences of political polarization.

May 07, 2018

| by Dylan Walsh


A worker at a factory in Kfar Saba near Tel Aviv cuts fabric printed with U.S. flags. | Reuters/Nir Elias

Partisan opinion is spilling over into our everyday lives. | Reuters/Nir Elias

It’s rancorous these days, politics. Consider Kerry Maguire, profiled in the New York Times, who gave her husband an ultimatum when he announced his intention to vote for Donald Trump. It’s him or me, she explained.

From dinner tables to the Capitol floor, conversation between Democrats and Republicans is shot through with antipathy. “A lot of discussion in both the national political discourse and academic communities has suggested that polarization today is less about issues and more about a visceral dislike of the other side — tribalism as opposed to reasoned disagreement,” says Neil Malhotra, a professor of political economy at Stanford Graduate School of Business. “A lot of the data we have on this, though, is from surveys, where people can simply engage in cheap talk.”

But are these partisan opinions more than cheap talk? Do our feelings about people of the opposing political party actually change how we behave toward one another? Malhotra, with three colleagues — Stanford GSB PhD candidate Christopher McConnell, Yotam Margalit of Tel Aviv University, and Matthew Levendusky of the University of Pennsylvania — ran a series of experiments to probe this question. They found that partisanship not only manifests as neighborly smearing or cheerleading, but also spills into our economic behavior.

Paying for Partisanship

Malhotra and his colleagues ran two general tests. In the first one, after a short demographic survey that included a question on political affiliation, people carried out a short job editing a website for grammatical and typographical errors. The text that they edited included mention of where the company’s founders met: working for a nonprofit (the control group), working on fundraising efforts for the Republican Party, or working on fundraising efforts for the Democratic Party. After completing this task, the participants were asked how much they would charge for a similar editing job in the future.

The researchers found that when Democrats edited for company founders who had met at the Democratic Party fundraisers, they proposed a discounted fee for future jobs. The same was true when Republicans did work affiliated with their own party. There was no difference in proposed fees, however, between people who edited for founders who had met working for the opposing political party and those who edited for founders who had met at a nonprofit.


What happens when we get into territory when someone donates to Trump and they lose their job because of it? Or when someone isn’t hired in the first place because of a particular social media post?
Neil Malhotra

The second test offered nearly 1,800 people a $50 Amazon gift card at half price. After receiving the deal by email, those who were interested had to register, and a portion of those who registered would ultimately receive the card. Everybody was told that the cards were leftovers from a fundraiser, but with three different treatments: one-third of participants were told the cards were leftover from a nonprofit fundraiser, one-third were told they came from Democratic campaign fundraisers, and the final third were told that they came from Republican campaign fundraisers. As with the first test, when the cards came from fundraisers for an aligned political group, people were more likely to register to purchase them; registration rates were lower, and comparable, when people were told they came from a fundraiser for the opposite party or a nonprofit.

“People were willing to charge their own party less when working for them, but they were not really giving a penalty to the other party,” Malhotra says. “And we found similar results when people were selling products: We are more willing to buy from fellow partisans, but we don’t seem to penalize counter-partisans.”

When Politics Becomes Prejudice

These were academically surprising results, as most studies of polarization demonstrate aversion to other groups more than they do fealty to one’s own. But the implications extend beyond disharmony among scholars. As Malhotra and his colleagues write in their article, published in American Journal of Political Science, “partisanship can systematically condition economic behavior.” This has broad — and disconcerting — implications.

“What happens when we get into territory when someone donates to Trump and they lose their job because of it? Or when someone isn’t hired in the first place because of a particular social media post?” Malhotra wonders. “You could imagine the same kind of thing happening in the rental market, say, or in any number of scenarios.”

Across most social divides — race, gender, religion — laws have been put in place to protect equality. This is the result of historical evolution, still underway, that appears to be directed toward more inclusivity. Politics, though, stands outside of this progress. Shaming of one political party by another is widely accepted; it is perhaps integral to the political process itself. This, says Malhotra, is an insidious problem.

“There is potential for our findings on partisan-based discrimination to be more troubling than racial- or gender-based discrimination because there are no social norms against it,” he says. “There’s no real social shaming if you don’t hire someone because she’s a Democrat or Republican, but do we really want a society segregated by partisan identity?”

For media inquiries, visit the Newsroom.

Explore More