Sheryl Sandberg famously suggested women 'lean in' to salary negotiations in her book.

Sheryl Sandberg famously suggested women 'lean in' to salary negotiations in her book. Alison Yin/Invision for Facebook/AP Images

Harvard Researchers: Not All Women Should 'Lean In' For Salary Negotiations

Women are less likely to ask for more, and therefore get less than men.

“Leaning in” to salary negotiations is a big part of Facebook COO Sheryl Sandberg’s manual for how women can empower themselves in the workplace. Women are less likely to ask for more, and therefore get less than men. Through salary negotiations, women can help themselves. But does that mean women should always negotiate?

New research says “no.” In an experiment where women must negotiate, they do worse financially than when they can choose whether or not to negotiate, according to a study published in a Harvard Business School working paper.

When women were forced to negotiate, they received wages lower than the suggested amount 26% of the time. When they had a choice and negotiated, they lost only 3% of the time. Men had more or less the same success rate as women whether they were required to negotiate. Unlike women, however, men who chose to negotiate didn’t do much better than when they were forced.

Women have sound reasons for being cautious when entering salary talks. Previous research shows that women can be penalized for trying to negotiate, even by other women. Much of that penalty comes from the double bind women can find themselves in: when women bargain like men, they can be perceived as pushy.

The study, by a team of professors from Harvard, Stanford, and the University of Pittsburgh, was conducted through laboratory experiments with 392 Stanford undergraduates who played the roles of employee and firm. As such, it can’t account for all the variables that women would face in a real-world negotiation. There’s also the fact that some participants had to negotiate (whether they wanted to or not) in order to create a control group, which isn’t a realistic scenario. And because genders were obscured from the participants, the experiment didn’t address biases women must confront in actual salary discussions.

Quartz asked Christine Exley, a Harvard researcher who was an author of the study, about the project and its conclusions over email. Her answers have been edited for concision and clarity.

Quartz: Your study examines whether forcing individuals to negotiate their salaries would benefit them. What does the existing research say about the subject?

Exley: A robust finding in the literature is that women enter negotiations less often. We expected to confirm this pattern and did.

A more mixed finding involves how women do in negotiations. We set up our environment to purposefully limit several factors that are known to disadvantage women in particular in negotiations. For instance, our negotiations are anonymous and thus limit backlash or image concerns. So, in this sense, we were not surprised that in our “Forced treatment,” whenever everyone negotiates, that women and men achieve similar returns.

Prior literature has not focused on whether women financially benefit from always “leaning in” or negotiating, relative to having the option about when to enter negotiations or not. So, learning more about the selection of women into negotiations was the focus of our study.

As I understand the results, you find that women should not “lean in” every time; that women do better when they pick and choose the opportunities to negotiate, in part because some know that they’re not as skilled at bargaining. But isn’t that the status quo, where women choose to negotiate based on how confident they are in their ability to haggle?

You are correct that we find that women should not “lean in” every time, that women do better when they pick and choose the opportunities to negotiate. This contrasts with the commonly held belief that women should “lean in” or negotiate more. In a recent poll that we conducted, 70% of adults in the US thought women should negotiate their salaries more often. The often-cited statistics about the “wage gap” are frequently coupled with recommendations that women should negotiate more often. Indeed, the US Department of Labor encourages women to “aim higher and negotiate better.” So, I do not think our results confirm the status quo—at least not the status quo belief that women would benefit financially if they would just negotiate more often.

Our results highlight a potentially different strategy—in particular, one that focuses on constructing an environment such that women can learn what they bring to the table. Where women do know what they bring to the table, women not only negotiate as well as men but they also make wise financial choices about when to negotiate or not.

So, when women are forced to negotiate, they do more poorly than when they have a choice (and can opt out of situations that don’t favor them), but men do about the same when they are forced or not? Why is that?

That’s right!

We construct an environment where men and women achieve similar and positive returns to negotiations, as shown in our Forced treatment. Yet, women avoid more negotiations than men when given the opportunity to do so in our Choice treatment. This is a scenario where there could be a temptation to jump to a premature conclusion—women are doing well in negotiations, just as well as men, but not negotiating as often. They should negotiate more!

Our empirical results serve a reminder as to why such a premature conclusion can be very wrong. Indeed, since women are positively selecting into negotiations, forcing them to always “lean-in” would be financially harmful.

As to why we do not observe significant evidence for men selecting into negotiations, there are many potential reasons. A leading reason may be that men were selecting into negotiations more often as is, so there is less room for them to benefit by negotiating even more. There is also some evidence that it may be more difficult for men in our setting to know if and when they are likely to be good negotiators—while there seem to be some “high ability” women who always achieve higher returns across all scenarios, men’s relative ability seems to depend on the scenario at hand.

Looking at your study in the context of the broader conversation about pay disparities between men and women—the starting point for the “lean in” argument—do you think it’s a fair conclusion that women’s ability to bargain isn’t the reason? That is, does your study support the idea that when women trail men in pay, it’s not because they can’t negotiate a fair starting wage?

Our study certainly pushes the needle away from thinking that women earn less just because they are worse negotiators. Ruling something out—that negotiation ability never contributes to a gender wage gap—is much harder and cannot be shown in any single study. That being said, we tend to agree with the growing call for more research around what policies work, as opposed to thinking women just need to get better at some particular thing.

Based on what you learned, how should the “rules of the game”—the framework for negotiating—be constructed? Is there any incentive for employers to make more information available to potential employees (men and women)?

A first step, that increasingly seems promising, is to have the “rules of the game”—regardless of what they are—known and well-defined. A next step, and building off of prior literature as well, may involve ensuring individuals know what they bring to the table. Pay transparency could help with this.

I think a lot of employers are motivated to find and implement procedures that do not disadvantage particular groups, such as women. Employers also benefit from workers who feel they are appropriately valued and did not “miss out” just because of how the negotiation process is set-up, or indeed, not set-up.