An interview with Lise Vesterlund, Professor of Economics at the University of Pittsburgh
Lise Vesterlund shares how The No Club began as a personal effort to manage overwhelming workloads and eventually became research that showed how women take on “non-promotable tasks” – the necessary but invisible work that doesn’t lead to promotions or advancement in a woman’s career. The No Club’s research shows that imbalance is often caused by an unconscious bias and expectation that affects pay and promotion. Vesterlund emphasizes that fixing the problem requires organizational change, and not a “fix the women” solution, and that awareness of promotable versus non-promotable work is the first step toward equity.
O’Donnell Center: What first inspired you to write The No Club and how did the idea first take shape?
Lise Vesterlund:
The project started 15 years ago as a personal journey when I and four other women formed a No Club to get our work lives under control. All five of us had accomplished careers, but we were working far too many hours and were not happy with the progress that we were making in our jobs. So, we started to meet every three weeks to understand why our careers had stalled and to talk about why our work lives were making us unhappy. It soon became clear that we were taking on too many tasks at work. We were all struggling to say no. And the tasks we were taking on were often those that, while helping our organizations, didn’t help our careers. Although we initially referred to these as crappy tasks, we ultimately gave them a more appropriate name and called them non-promotable tasks.
You might say: What are these non-promotable tasks? And it depends on where you’re working and where you are in your career. Common examples would be helping others with their work, being on committees, organizing events, mentoring, office housework, etc. Typically, they’re tasks that are not mission critical, they tend to be invisible, and they tend to not rely on your specialized skills. We were doing a lot of these.
While we were spending time trying to figure out how we could manage our own loads of non-promotable tasks, we started discussing the challenges with our colleagues, and were surprised to learn that they also felt that they had an excessive load of non-promotable tasks. That caused us to approach the question from a research angle. What started as a personal journey became a quest to understand who is taking on the non-promotable work and whether and why there might be gender differences. Was an excessive load a problem faced equally by everyone? Or was it more pervasive for women?
To assess differences in workloads we collected data from organizations and conducted a broad literature review. Every source of data revealed that women are doing more of the non-promotable tasks than men. We found in one professional services firm that women were spending 200 more hours per year on the non-promotable tasks than their male colleagues. That’s a solid month of additional non-promotable work! Similar patterns were found in other organizations and in previous studies examining work allocations. The evidence was clear. Women everywhere were spending more time on non-promotable tasks.
The second question was why this happens. There are many potential reasons why men and women may spend their workdays differently. It could be that women have different preferences; it could be that they’re better at certain things. To understand what gives rise to different work allocations, we wanted to study the phenomena in the lab, and in doing so we wanted to capture settings that mirrored those where we ourselves often took on non-promotable tasks. We were all far too familiar with the pressure we felt when attending a meeting where a request was made to volunteer for a task that no one wants. Writing a report, planning an event, or joining a committee to handle a challenging issue. We wanted to capture the tension and dynamics of deciding to volunteer, and ultimately, reluctantly, raise one’s hand.
We came up with a very simple setting and began by examining who volunteers. Over ten rounds we randomly formed groups of three and presented members with incentives to volunteer to click a button on a computer. If nobody clicked a button within two minutes, everybody would get $1. If you clicked the button, you would get $1.25 and the two other members would get $2. So everybody would be better off if a volunteer was found and would prefer someone else to volunteer.
Consistent with our findings in the field, we found that women in the lab volunteered more than men. Women clicked 48% more than men. And they kept doing this over the ten rounds.
In the field we couldn’t determine what drove the gap. It could be because women had different skills, but this wasn’t a factor in the lab. Men and women both know how to click buttons. So differences in skill were not driving the gap. It could be because men and women have different preferences. Women could be more altruistic and maybe this is why they volunteer more. Subsequent studies ruled out that the gender gap in volunteering resulted from differences in preferences. We found in all-male and all-female groups that females were no more altruistic than males; the two types of groups volunteered at the same rate. Men knew how to volunteer when we removed women from their group.
This and subsequent studies revealed instead that the gender gap in volunteering resulted from a collectively held belief that women more than men will volunteer. For example, in a study where participants are asked to volunteer, we find that women are asked 44% more often than men, and they confirm the commonly held belief by accepting the request 50% more often than men. Intriguingly, both women and men ask women more. Everybody expects women to volunteer more.
These two results were key to addressing our research questions. First documenting that independent of organization, women more than men, spend time on non-promotable tasks, and second, documenting that this difference isn’t merely a result of differences in preferences or skill, but is affected by the expectation that women more than men will take on this work.
O’Donnell Center: I agree. There’s this unconscious decision that we as women make or that others perhaps put on us as women. How much of that is it because we feel responsibility to volunteer or an expectation that we feel we need to fulfill? And why might that be?
Lise Vesterlund:
We have a tendency to say that women are somehow special. Maybe they’ve been socialized to be more caring or have developed a preference for helping others. What’s so interesting about our laboratory studies is that they reveal that the gender gap in volunteering isn’t driven by differences in preferences. If you put women together with women, they’re not volunteering more than groups of all men. Women paired with women know they can rely on the other women, so they volunteer less, but when they are paired with men, they know that it is in their interest to volunteer more. Their behavior is driven by our gendered beliefs, rather than a desire to fulfill a responsibility to behave in a certain way. When we get into a setting that has both males or females or minority and majority individuals, we tend to think that the less- empowered or underrepresented group will take the less- regarded work, and they too have internalized that belief. So, it is in their interest to take it up. Documenting that these beliefs are affecting work assignments is key to justifying intervention. It points to the distribution of work being inefficient.
O’Donnell Center: You alluded to this earlier, but what are two to three key takeaways from your research, especially as you worked through the book, and perhaps as it relates to your own research that you found interesting or surprising?
Lise Vesterlund:
In terms of what surprised me the most – it was that our stark and simple experiment could uncover gender differences in volunteering, and that we so consistently find that the gap is driven by beliefs. It was also surprising to see how pervasive the problem is. We have yet to find a profession where work allocations are not gendered.
We’ve worked with many organizations that will say, “Yes, we understand that this is probably a problem in some organizations, but it is not a problem in ours.” When we’ve challenged those organizations by saying OK, let’s look at the data and see where we are, we have found gendered work allocations in every single one.
A takeaway is in understanding the impact of differential work assignments. I’ve spent most of my career looking at gender differences in the labor market, and we spend a lot of time looking at wage differentials and promotion differences. These differences in labor market outcomes are sustained by differences in work assignments. I now see gendered task assignments as the structural problem that prevents women from reaching their potential in organizations.
If we don’t give peers of young employees similar work assignments, then we will fail to identify their talent, and we will never get to equal pay and advancement.
O’Donnell Center: I’m curious, in your own research and after publishing this book, have you already seen changes in the pay gap, gender differences, and equitable distribution of work assignments?
Lise Vesterlund:
Several organizations have been eager to change. My own institution, University of Pittsburgh, has made several changes. However, in the field we have yet to ensure policy adoption in a way that allows for causal inference of impact.
Our finding that gendered beliefs contribute to the difference in work allocations demonstrate that organizations are making mistakes. And organizations recognize that they’re making mistakes. We shouldn’t be allocating work to the employee who is least reluctant to take it on; we should be allocating it to the employee who holds the comparative advantage. When we wrote the book, we hoped that all organizations would recognize this.
And indeed, some of the changes are super simple. For example, don’t ask for volunteers, instead take turns or randomly assign the task. Get managers to take responsibility for the distribution of work. Get them to reflect on the problem, so they don’t subconsciously assign less desirable work to those least empowered. Accountability for work allocations is critical.
Regrettably, organizational change has been slower than what we hoped for. Admittedly we are at a historic low point when it comes to concerns for diversity, equity, and inclusion. Many organizations have stepped away from values that they previously saw as core to their mission. Many have lost sight of the fact that these measures were in place to ensure meritocracy. They were in place to ensure that talent be identified and promoted.
While we still believe that this is not a “fix-the-women” problem, slow adoption has made it clear that in many organizations it is a “women-fix-it” problem. Women supporting one another and becoming one another’s allies can help equalize the distribution of work. They can speak up to support one another and propose obvious improvements to the way that work is allocated.
While we initially saw this as an obvious lift for organizations, we now hope that individuals will carry the message forward. Young employees, especially women and minorities, need to know that there’s promotable and non-promotable work, they need to know that the right balance is needed if they’re going to succeed in the organizations they’re joining.
O’Donnell Center: It is very important to advocate for diversity, and for women especially. The book mentions the concept of equal distribution and representation in workplace settings compared to men. Can you tell us more?
Lise Vesterlund:
Conflict often arises around equal representation and distribution. For example, many academic institutions strive for equal representation on committees, although only a third of their faculty are female. Equal representation in this case causes women to carry twice the service load of men.
This is not to say that women shouldn’t be on committees, but they need to be seen as a scarce resource. Focus on getting representation where it matters most. Place women on hiring and executive committees rather than holiday committees. For faculty of color it is a substantial problem. We expect faculty of color to over represent in every dimension and magically maintain a comparable research portfolio.
Either equalize service loads or find other ways of protecting time on research. Compensate faculty with an excess service load with a reduced teaching load or greater access to research funds.
O’Donnell Center: After you’re exposed to and know about non-promotable work, you can’t not see it. My hope is that this is more widely seen and understood, not just by women, but also by men and those in leadership. So, let me ask: Where do you see the research going? You expose some of what’s already there, but what’s next in your own research and behavioral science? What questions remain after the book was published?
Lise Vesterlund:
There are a number of open questions left after the book. We published the book in several other countries, including in my home country, Denmark. Even though the book was very well received, some saw the book as an anthropological study. The sense was that this is a problem in the U.S., but not a problem in Denmark.
So a natural question is whether gendered task allocations solely is a U.S. phenomenon? It turns out that it isn’t. Over the past five years, I and my co-author Maria Recalde have been working with a team of researchers in Bergen at the Centre for Experimental Research on Fairness, Inequality, and Rationality (FAIR) to understand the distribution of work in Norway. Norway is the second-most gender equal society in the world, just after Iceland, and despite this we have uncovered that gendered work allocations are a very pervasive problem there. We’ve been working with organizations and been following students as they are getting their MBA and entering the labor market, and we can see that, indeed, the problem extends to other countries.
A second question is when gendered work arises and what the consequences of the work are? We’ve found that the onset is very quick. Comparable men and women are within their first year on the job market given different work assignments and these affect their career trajectory.
In laboratory studies we have studied what happens when workers randomly are given work portfolios that vary in the share of work that is promotable. Forget about gender, forget about the employee’s credentials. Imagine simply that we give one new employee a portfolio with more non-promotable tasks and another new employee a portfolio with more promotable tasks. How does this random assignment affect evaluation, compensation, and promotion?
Our research shows that your randomly assigned portfolio of tasks has a substantial impact. If you are given more non-promotable work, then you are seen as a lesser employee, even when you hold the comparative advantage on the promotable task.
If you think about two new employees, one with more billable hours than the other, then the person with more billable hours, even if they’re not better at it, will be seen as the better-performing employee. Giving detailed information about how good employees are on the tasks doesn’t help, the manager cannot unsee that they’re getting a larger revenue from the worker who was randomly given more promotable work. Once you have a larger load of non-promotable work, you’re then seen as the lesser employee, and this depresses both your compensation and your advancement.
In fact you can’t even negotiate for improved compensation, because there’s somebody with a better task portfolio. So, the consequences, even from randomly assigning work, are severe.
A number of questions remained after the book, and we are in the process of addressing these. However, the insights from our work go beyond understanding the onset and consequences of gendered work. What I hope, for research on gender, is that we begin to understand the dynamics and the structural problems that give rise to gender differences in compensation and advancement.
We need to take a few steps back to understand what happens every day at the office. What is the process that generates differential treatment?
O’Donnell Center: Really fascinating. My last question, Lise, what advice would you give to young researchers or a person early in their career?
Lise Vesterlund:
While we make clear that this is an organizational problem, there are things that an individual can do to better navigate excess requests for non-promotable work. Step one is to become aware of what work is the promotable and what work is non-promotable. A manager may say, that all work is promotable, and that’s fine, but some work is more promotable than other work, and the less promotable work is effectively non-promotable. Awareness of the non-promotable work is critical.
Second, get better at saying no, and yes. Joining a No Club is a good starting point. Find allies that can help you navigate requests.
Become aware of the bias you have for saying yes. What are the situations where you end up saying yes when you really don’t want to? Is it because you get caught off guard, and you just don’t have enough time to think? Is it because you underestimate how long the task takes? Is it because somebody gives you a compliment and you suddenly feel honored? “Oh, you’re so good at putting the holiday party together, you should do it again.” Does this type of diva moment push you to say yes?
In my case, my problem was that I didn’t think about my implicit no. Every time you say yes, you’re saying no to something else. So what is that implicit no? Is it that you aren’t doing more of the promotable work? Is it that you are working late? Is it that you don’t get to have time with your spouse, with your kids, with yourself?
Be aware of the bias you have for saying yes. If you can pause a little bit, become cognizant of where the danger zones are for taking on too much. Think about when you can and can’t say no, and improve the way you do it. The response is often, “Well, women just don’t know how to say no, they need to learn how to say no.”
The consequences of saying no are different for men and women. So caution is needed when declining requests. Our book provides a lot of guidance on this front. Saying no strategically reduces the potential for back lash. In cases where you have room to say no, give a short explanation for why you can’t do the work, point to the promotable work you will be cut off from. Then help the requester solve the problem by presenting them with an alternative solution. You know the organization from a different level and may be able to point to someone for whom the task is promotable.
When you say yes to a non-promotable task, try to negotiate your yes. Can you offload another non-promotable task? Can you divvy up the work and say, I’ll do part A if John will do B, and Carl will do C. Can you put a time limit on it? Can you negotiate for a promotable task instead?
When I first was hired at Pitt, I was asked to be in charge of the website because they thought that my sense of color was so good. I thought with a PhD in economics that maybe there were other things where I could better spend my time. I thanked them and instead asked to be on the recruiting committee.
Sometimes you can negotiate for something that is more promotable, so looking at these requests, as a way of saying, I want to help the organization, but the best way to help the organization is if we fully utilize my skills.
O’Donnell Center: Great advice. Lise. Thank you. Anything else that you want to share, any final words or anything we might have missed during our chat?
Lise Vesterlund:
Getting the word about non-promotable tasks out there is empowering. We want everyone to become familiar with the concept, so they don’t get blindsided. Thank you so much for putting this together!