When former US secretary of state Madeleine Albright gives speeches, there’s one line she returns to again and again: “There’s a special place in hell for women who don’t help each other.” Her signature sentence has appeared, with occasional variations, on Starbucks cups, T-shirts, and mugs, and supplied Taylor Swift with a ready retort after comedians Amy Poehler and Tina Fey cracked wise about her love life.
Most feminists would readily agree with Albright’s sentiment—namely, that women who help and advocate for one another play a crucial role in advancing gender equality. But the quote also hints at the gendered expectations that powerful women face.
It’s one thing to believe that women should support one another. Definitions of what constitutes supportive behavior, however, may vary. In some cases, it’s not such a far leap to the decidedly sexist belief that women should act as bottomless wells of altruism and kindness, supporting their male and female colleagues alike via praise, assistance, and advice, while never uttering a word of criticism.
That puts women who rise to positions of power in a difficult spot, as illuminated by a recent study (pdf) from Martin Abel, an assistant professor of economics at Middlebury College, published in the Institute of Labor Economics. All managers need to be able to give tough feedback at times. But Abel’s research finds that both men and women discriminate against female bosses who dish out criticism, even when the feedback is worded identically to the feedback given by male managers.
No praise? No thanks
For the experiment, Abel recruited 2,700 people on Mechanical Turk, Amazon’s online gig platform, to do transcriptions for a fake firm, randomly assigning each worker a (fake) manager with a male or female name. The names, chosen to balance out possible associations with age, race, and level of education, were Brittany, Chloe, Christine, Ebony, Emily, Jennifer, Lynn, and Shanice for women, and Darius, Doug, Ethan, Josh, Justin, Michael, and Tyrone for men.
Partway through the task, 60% of workers received either positive or negative feedback from a male or female manager, based on whether their work was above or below average as determined by an automatic scoring calculation. The standard text read:
Hello, This is NAME. As mentioned in the task introduction, I’m overseeing your performance in transcribing the receipts. I just went over some of the receipts. Your performance has been above / below average. I was pleased with / disappointed by your effort and attention to detail. Going forward, remember that your continued commitment will improve / lack of commitment will harm the quality of our services. NAME
After completing the transcriptions, workers answered a survey with questions about their job satisfaction and their interest in working for the firm in the future.
The results? Both men and women were much more negatively affected by criticism when it came from a female manager. “In fact,” Abel notes, “criticism from female managers doubles the share of workers not interested in working for the firm in the future and leads to a 70% larger reduction in job satisfaction than criticism from male managers.”
“Criticism from female managers leads to a 70% larger reduction in job satisfaction than criticism from male managers.” Notably, criticism from a female manager didn’t affect the amount of effort workers put into the task, which Abel suspects had to do with the nature of the MTurk platform. “Similar to Uber drivers, people are penalized heavily for negative reviews (i.e. getting their task rejected),” Abel explains in an email. “If you get criticized, you know you face that risk, so you are unlikely to reduce your effort.”
But in general, he notes, research shows a strong relationship between job satisfaction and productivity—so in another context, the fact that workers’ attitudes soured after receiving criticism from a woman would be likely to impact their performance. Moreover, he says, the fact that workers were less likely to work for the firm again after receiving criticism from a female boss “can be seen as a form of productivity reduction, since retention is a key objective in many industries.”
“What does she know, anyway?”
The study found a correlation between workers’ differing reactions to male and female managers and their gendered expectations of bosses. Abel explains in the paper that the workers surveyed were “about three times more likely to associate giving praise and appropriate use of tone with female managers. By contrast, they are about twice more likely to associate giving criticism and strict expectations with male managers.”
On a psychological level, criticism can sting even when you expect it. But it’s far more upsetting when the criticism seems to come out of left field—and because women face social pressure to come across as sensitive and undemanding, a stern word or two from a female boss can provoke a disproportionately big reaction.
While both men and women bristled at the criticism from female bosses, there was at least one key difference between workers of different genders: Men also demonstrated a tendency to dismiss the validity of criticism from women, judging their feedback as less accurate. “By contrast, female workers’ perception of feedback does not vary by manager gender,” Abel finds.
The discrepancy suggests there may be other factors at play, in addition to gender expectations, influencing women’s reactions to female bosses. The study was inconclusive about what those factors might be. But it did rule out implicit bias, since there was no correlation between workers who held stronger biases about the roles of men and women and their reactions to criticism from a female manager. Abel also rules out the hypothesis known as “attention discrimination,” which holds that women’s negative feedback simply gets overlooked; workers actually spent more time processing criticism from women.
Among workers in their 20s, though, the variation in reaction depending on the boss’s gender vanished entirely. Importantly, Abel acknowledges that since the study was conducted on Mechanical Turk, his findings may be more accurately reflecting the experiences of workers and bosses in the gig economy, who are more likely to have short-term, remote relationships, as opposed to the dynamics in a traditional office environment. But as he also notes, the gig economy is rapidly expanding. A recent McKinsey study suggests that 20% to 30% of working-age Americans and Europeans participate in it in some way, whether by choice or necessity. How gig workers respond to the women charged with overseeing them is a matter of great concern to the future of work more broadly.
For those feeling discouraged about the challenges faced by women leaders, the study does identify a bright spot: Younger workers are less likely to react differently to criticism from female bosses, and among workers in their 20s, the variation in reaction depending on the boss’s gender vanished entirely. Abel warns that “young workers may adopt attitudes that lead to gender discrimination as they get older.” But he says it’s also possible that millennials and members of Gen Z may indeed remain receptive to criticism from women.
That would ultimately benefit both women in power (who would presumably have greater ability to make changes in the workplace) and workers themselves, who would be able to benefit and grow from constructive criticism. After all, there may indeed be a special place in hell for women who don’t help other women. But a workplace where women aren’t allowed to challenge other people is a far cry from heaven.
Sign up for the Quartz Daily Brief, our free daily newsletter with the world’s most important and interesting news.
More stories from Quartz: