Politics around the world is increasingly marked by polarization, but many people would assume that workplaces allow for a diversity of views.
Perhaps not. Political polarization spills over into the workplace and affects how people do their jobs, research suggests. Chicago Booth’s Emanuele Colonnelli, EPGE Brazilian School of Economics and Finance’s Valdemar Pinho Neto, and Northwestern’s Edoardo Teso find that bias rooted in political partisanship is more influential in hiring decisions than gender or racial bias—and that political discrimination might even affect not just individuals but corporate performance.
Decades of studies have documented the workplace effects of all kinds of bias, including racial, religious, gender based, and age based, but little has been written on political discrimination. To address this gap, the researchers turned to Brazil, a country with polarization so extreme that in 2021 the Carnegie Endowment for International Peace wrote that the situation “has become a major risk not only to the country’s democracy but also to its capacity to address its most urgent policy challenges such as the coronavirus pandemic.”
Colonnelli, Neto, and Teso used several methods to tease apart the relationship between partisanship and employment. They began by compiling data on individuals and corporations between 2002 and 2019 that came to cover 87 million workers across 7.5 million employers. The information included occupations, wages, hours, and employee gender, educational level, and political affiliation, among other things.
They then analyzed how the political preference of a company’s owner related to those of its workers, finding compelling evidence of partisan discrimination at play. In aggregate, job candidates who shared the politics of a company’s owner were more likely to be hired at that company. Their tenure as employees averaged 5 percent longer than those who didn’t share the same politics.
In Brazil, business owners were on average 1.5 times more likely to hire workers who belonged to the same political party than would be expected if owners and workers were randomly matched. They were far more likely to hire someone of the same political party than to hire someone who shared their race or gender.
According to the data, job candidates of one political party were 40–75 percent more likely to be hired by someone of the same political party. For gender, the figure was 15–31 percent, and for race, 3 percent.
The discriminatory effect of political affiliation was stronger at smaller companies, where “personal interactions between owner and worker are likely more frequent,” write the researchers. Sorting along party lines was greatest among managers and weakest among low-level workers, they find, surmising that this is because owners and managers work more closely. When the owner of a company switched parties, the change showed up in a sudden shift in the politics of new hires.
Colonnelli, Neto, and Teso then conducted a field experiment, sending a set of fake résumés to 150 business owners and asking how interested they would be in hiring each candidate. To encourage truthful and well-considered responses, the researchers told the owners that after the experiment they would receive real résumés similar to those in which they expressed interest. The résumés were essentially identical to each other except for signals about an applicant’s political affiliation—introduced through cues such as volunteer experience with a particular campaign.
In the end, résumés of candidates whose partisanship aligned with an owner’s attracted a level of interest roughly 9 percent higher than those of candidates who were affiliated with a different party.
Finally, the researchers surveyed 1,000 workers and 1,000 owners at different companies. They explained their findings of political bias and then simply asked the participants what they thought drove the results. Owners and workers overwhelmingly cited discrimination, and 29 percent of business owners acknowledged outright that they took political views into account when hiring.
The upshot of this bias may ultimately hurt companies. Looking at the performance data for employees and companies, the researchers find that workers who identified with the same party as an owner tended to make more money and earn promotions more rapidly, even if they were less qualified than coworkers of a different party. And more generally, companies with a workforce more closely aligned with an owner’s politics appeared to grow more slowly than similar companies with a more politically diverse workforce, the analysis demonstrates.
Colonnelli suggests interpreting these outcomes “with a larger grain of salt” than usual because companies with a more politically aligned workforce could be different in many other ways than the researchers are able to observe in the data. That said, he adds, “it looks like there are real costs of partisanship in the workplace.”
Emanuele Colonnelli, Valdemar Pinho Neto, and Edoardo Teso, “Politics at Work,” Working paper, June 2022.
The decisions of companies, governments, and educators will help to shape the ultimate outcomes of the A.I. revolution.
A.I. Is Going to Disrupt the Labor Market. It Doesn’t Have to Destroy It.
Research examines people’s willingness to cut ties with companies whose actions they deem immoral or unethical.
How Much Will Consumers Sacrifice for a Moral Stand?
Increased polarization in the United States has meant more political homogeneity across digital, social, and civic spaces. But what about workplaces, where so many Americans spend so much of their time?
Capitalisn’t: The Political Polarization of Corporate AmericaYour Privacy
We want to demonstrate our commitment to your privacy. Please review Chicago Booth's privacy notice, which provides information explaining how and why we collect particular information when you visit our website.