Kathryn Anne Edwards on Women Leaving the Labor Force in the COVID-19 Pandemic

Chancellor: Hello and thanks for joining us for the Poverty Research and Policy Podcast from the Institute for Research on Poverty at the University of Wisconsin–Madison. I’m Dave Chancellor. For this episode, I talked with Kathryn Anne Edwards. Dr. Edwards is an economist and Professor at the Pardee RAND Graduate School and we are also proud to note that she is a former IRP Graduate Research Fellow. We talked about the striking number of women, and especially moms who have left the U.S. labor force since the start of the coronavirus pandemic. We talked about some of the patterns she’s seen around why women are leaving the labor force and how the lack of support for working parents could roll back the gains we’ve seen in women’s work and the economic benefits that have come with them. Let’s turn to the interview.

Chancellor: Kathryn Anne Edwards, thanks so much for joining us for the Poverty Research and Policy podcast. Today, we’re talking about women leaving the labor force in the COVID-19 pandemic. But before we get to that, I’m hoping you can introduce yourself and tell us a little bit about the kind of research that you do.

Edwards: Absolutely. Well, howdy, and thank you for having me, I’m Kathryn Anne Edwards. I’m a labor economist at the RAND Corporation and proud Wisconsin alum. I got my PhD at the University of Wisconsin and was a graduate research fellow with IRP. I look at inequality in whatever way it manifests in the labor market between genders, between races, between educational levels. When someone loses their job, when they lose the ability to work because they become disabled or because they retire. And really the work just looks at people who work.

Chancellor: At the end of November this past year, you wrote a piece titled Women Are Leaving the Labor Force in Record Numbers that looks the really striking disparity in the number of women who have dropped out of the labor force or we’re no longer looking for work since the start of the COVID-19 pandemic. Tell me what’s going on here.

Edwards: In any recession, we expect for workers to leave the labor force, so you lose your job or you’re looking for a job and you’re having a hard time, and it’s natural to kind of give up and get discouraged and stop looking in in order to be included in the labor force, you have to be actively seeking work. This would be the difference between a retiree and a student versus someone who’s unemployed. Right? The student doesn’t need it, doesn’t want a job, and neither does the retiree. So, they’re not in the labor force. In recessions, we always see that there’s this fringe group of workers who are notionally wanting a job. But because they can’t find one, they kind of technically drop out of the labor force because they don’t look as hard or as or as long. And then there’s people who leave entirely and decide to wait it out. That happens every recession and this recession. In April, the U.S. economy shed twenty-two point five million jobs in a month. And so naturally, we would expect as a consequence of that that people would sit out the labor force and sit out the labor market during the recession. But the other thing that’s going on and that’s unique to this recession is that we have large scale closure of in-person schooling for K through 12 and large-scale closures of daycare centers, which means that in addition to this kind of push from the a very bad labor market, there’s this pull home because your kids need schooling. What we were looking at in November is how much of what is what how much of women dropping out of the labor force is because of the jobs they had or because of the labor market right now or because of their kids. We didn’t have a great way to say like a traditional causal ID that you would require in economics. But we wanted to build some intuition. So, we just compared women by the age of their kid and by the number of kids they had, because for anyone out there who’s a parent taking care of a 15-month-old versus a 15 year old requires a different amount of attention. And so, what we found was that women with more kids and women who had kids who were of the difficult age of two to five dropped out in larger numbers. So what percentage is the labor market and what percentage is caregiving? We couldn’t speak to precisely. But what we could say is that both are happening. And for sure, there is a caregiving aspect that is driving women out of the labor force.

Chancellor: You started the article talking about what economists Claudia Goldin called the quiet revolution in women’s work that started back in the 1960s or 1970s. And so you’re kind of clearly looking at this drop off that we’re seeing here today in a historical context. What is that context and how does it help us understand what’s going on now?

Edwards: One of the most remarkable features of the U.S. labor force in the 20th century was that women’s rate of working doubled between 1950 in 2000, and it was a slow and steady increase that was maintained through all kinds of economic conditions and all kinds of changing demographics and all kinds of labor market opportunities. But the type of work that was typical for the American worker, what Claudia Goldin had found in her research was that this, you know, relatively stable increase masked a rather tectonic shift in thinking among women workers, which is they no longer view themselves as working until they got married or working until they had a kid. They viewed themselves as having a long-term career and in having a career, became part of their identity. And she this revolution was entirely about the mindset of how women approached work. And she shows not necessarily just through the number of women working, but their mindset towards how there was this revolution in women as workers. That quiet revolution, sustained labor force participation increases up until 1999, and that is when women’s labor force participation peaks. And since then, we’ve been in a 20-year stall. And so, the way that I think about what’s happening now is that this this quiet revolution could have also been characterized really as an unsupported one, because at no point between 1950 and 2000 did we have large scale interventions to support childcare. Have we ever had universal access to sick leave, universal access to paid family leave, the type of things that have been shown in other countries in context to support maternal labor supply? We’ve never had them here. And so this kind of incredible accomplishment, women really, you know, did on their own. But we’ve arguably we’ve hit a wall and arguably we hit it 20 years ago. And I think what the pandemic is really bringing to light is just how much of a constraint childcare makes for parents and how much of that constraint is often absorbed by the mom.

Chancellor: So I want to talk more about that idea of unsupported work in a bit. This is kind of a foundational thing. Can you just walk us through in high level terms what actually happened with the labor force in 2020? I mean, we’ve all been living through this pandemic, but what actually happened starting in March of 2020 and what did that look like throughout the year?

Edwards: So at the end of 2019, according to the Bureau of Labor Statistics, 50 percent of all employees in the U.S. are women. It’s only the second time that it’s happened in U.S. history. The time before was during the height of the Great Recession. When we start 2020 year, women should not be thought of as the minority part of the workforce. They are half of workers and we’re coming off of a very long economic expansion. And so unemployment is low, wages are rising. When the pandemic hits and we go into closures and a pullback in consumption spending. The economy sheds twenty-two million jobs in a single month, which is April. So, in May and June, there’s a really quick snap back that, you know, about half of those jobs are recovered very quickly. And then, you know, through the rest of the summer into the fall and now even through the winter, job growth has just it has been anemic relative to the size of the hole. And we are still 10 million jobs down than we were from this time last year. Now, the jobs that were lost were not randomly distributed through the economy. They tended they tended to be concentrated in certain industries and sectors that are associated with in-person consumption. Leisure and hospitality, for example, it lost 45 percent of its jobs in a month and it hasn’t gained nearly half of them back. One of the reasons why this has been the women’s recession is because the industries and occupations that were hit hardest were those that have a higher share of women workers to begin with. But I should be really clear that even within those industries, more women lost jobs than men. It’s not that half of, you know, leisure and hospitality had an even job loss among the men and women who are employed there, even within an industry or within an occupation that had a high number of women workers. Women were also much more likely to lose their job than men. And this is this brings us into kind of the big mass of people who are out of the labor force. So, you know, the mechanisms for women leaving the labor force could be twofold, right? It could be that they lost their job and then their kids didn’t have school. They just weren’t going to try and go back. Or it could be that they left their job and that they didn’t they weren’t affected by the initial job loss. But eventually they decided to leave their job. And it’s important to stress as far as the labor market goes, these are just the two things that we can really easily see in real time through monthly statistics, who is working and who is not and who is searching and who is not. But there are countless ways in which parents have been affected by the recession and not having in-person schooling. And we don’t have the way to measure of people who have turned down promotions, gone part time, you know, been passed over, you know, might be on probation, cut back their wages are going to miss out on five years of growth because of the actions that they’re taking to accommodate.

Chancellor: Now, why is this something that we’re seeing with moms and not with dads?

Edwards: So, first thing to say is that this doesn’t mean that it’s not happening to dads. It doesn’t mean that there haven’t been dads that have left the workforce or dads that are working from home and really struggling, whether while their wives have to go in. But the reason why we think it’s moms is one they’re the ones that have seen larger labor force declines, especially when you limit between workers who have children. And to you know, it’s not just the pandemic. It’s likely reflecting years of decision making that the pandemic is kind of revealing the patterns of. So, we tend to think that people specialize with household production. Right. Like, if I if I split the chores with my husband, I don’t do the dishes stop at exactly halfway through and say, OK, now it’s your turn to finish the dishes. Right. And I don’t unload half the laundry and then leave half of the wet clothes in there and say we split chores. So, you need to do this. You know, you tend to divide task across, you know, partners and not divide the actual task themselves so that that applies to child care and that applies to jobs. And we think that more than likely women will take you know, one person has to take the flexible job. At least one person has to take the job that has the health insurance and the retirement and things like that. Someone has to have the stable job. Someone has to have the flexible job. Someone has to be there at 5:00 or pay the penalty, the permanent lateness of penalty from the day care for being late. And so that those choices are really you’re moving back inches and inches from the labor force and full participation. Right. I took the job that was that paid less. They gave me more time. I, I skipped the promotion that required me to travel more and it’s just an inch and an inch and an inch. And then the pandemic just split that miles open. And so, when so when one person had to give it would make sense. It was the person that was already earning less that was already in the more flexible job and that was already taking care of that kind of taking. Primary parent for a caregiving that is not necessarily the woman in every marriage, but on average, we think it’s more likely that it’s the woman who’s taken the secondary earning job, the woman that’s probably that’s take the more flexible job and the woman that’s less likely to be in person. I think one thing I say over and over again in the recession is that an experience doesn’t negate the average, and the average doesn’t negate a different experience.

Chancellor: I think when we’re thinking about this issue, especially child care, you mentioned earlier that one of the ways that you get at this is by looking at the differences for women who have different age children in the research that you do on this. Can you tell me about that?

Edwards: We went into the Current Population Survey and looked at households that had children and compared the labor force participation, the change in labor force participation for women and men in those households based on kind of the composition of the kids ages. And what we found was that, you know, if you think toddlers are the hardest to take care of, it would require the most time. You know, they were the those it’s parents of toddlers who we saw had the largest labor force decrease relative to the start of the recession, whereas with teenagers it was smaller and with a zero-year-old, it was smaller.

So we tried to use the variation and the amount of caregiving that kids require in order to figure out if there was a real caregiving kind of pull out of the labor force. It’s not perfect, you know, I mean, someone’s going to come by in years and come up with some amazing study that looked like a child care center that closed randomly, but like one next door didn’t. And it all meets all the qualifications you would need for perfect causal identification to be, you know, a good economics study. You know, in a recession, you have to build intuition where you can guess.

Chancellor: One thing you said there is interesting. You said that women who have a zero-year-old—basically an infant or very young child—their labor force participation dropped a little bit less so than you might otherwise expect. Why’s that? What do you think is the story there?

Edwards: So we found that for very young. Zero to one, we didn’t see a meaningful decline in labor force participation relative to the start of the pandemic, you know, versus someone who had a two- to six-year-old where we saw on a five-point decline in women’s labor force, labor force participation compared to, you know, roughly zero decline for a woman from zero to one. So, couple of things could be happening there. It’s a change. It could be that in general, when someone has a zero-year-old baby, they do leave the labor force. So, there wasn’t an additional inducement out from the pandemic, or it could have been that women who were on maternity leave were able to basically extend their leave and stay at home with the kid and come back maybe part time or work remotely, which would have been harder if they had to go in person every day. That would be my sense of what could be going on. And also that those kids I mean, speaking from experience of having started the recession with a three month old, you know, you can kind of just like strap them into a bouncer and leave them for a couple hours, which is very different than when a, you know, a two year old walks up to you and just like demands to be held so that I am I am not unbiased and how I divided these age groups and in thinking about this.

But, yeah, there’s you know, the real question is, you know, at zero years old, even not in a pandemic, a woman could be deciding her labor force participation going forward and that the relative change, the pandemic might not have been large enough because it’s a time associated with lots of labor force decisions. Whereas, you know, possibly if your kids for, you know, you were going to work, if you were going to drop out because of kids, you probably did it already. Whereas the pandemic was kind of an additional like a shock relative to what you had established, whereas there’s lots of movement around zero years old when someone has a kid, when a woman has a kid.

Chancellor: We’ve talked a lot about women who have left the labor force or just some of the adjustments that have been taking place in the home. But what about women who have stayed in the labor force? You know, I don’t want to ask how they’re doing. I don’t think anybody is really doing well right now. But what do we know about that and kind of what’s the story there that you see?

Edwards: Well, for both conversations of women who left and women who stayed, there’s very large in both respects, women who stay, women who go. It’s important to remember that we expected this to be a very different story based on your income and therefore very different story based on your race and ethnicity. So, for some women who have stayed on the job in our remote working, you know, we have, you know, scattered surveys through private companies that, you know, they’ll poll a thousand women or they’ll ask a workplace to put in a poll and find out what’s going on. And they’ve all shown, you know, women are miserable, increased stress, increased drinking due to stress and pulling back on hours, trying to pull back on their workload, moving part time, all these types of like small withdrawals and additional stress. So, I think there was a McKinsey survey that’s associated with Lean In that found that something like one in four workers was either going to quit thinking of quitting, had reduced their hours or had reduced their voluntarily reduce their pay. But that that really does have a sample problem because it’s pulling a lot from the corporate American workplace and it’s pulling a lot from older, whiter women, you know, the other side of this would be women of color who are, you know, around 30 percent of Hispanic women and 30 percent of black women who work are in service occupations, which means that they’re in personal care and service, they’re in food service or, you know, grounds cleaning and maintenance like these are in person face to face occupations. And they have very high unemployment rates. And if they are on the job, their job has become much more dangerous. And what the effect has been on them in terms of covid transmission, in terms of how they’re able to do childcare, I think that question gets harder to answer because, you know, we have less of an inclination to survey them in private studies to see how they’re doing.

Chancellor: So I want to go back to Claudia Golden. You wrote in your article that what we saw in 2020 and what we’re likely to continue seeing this year in 2021 are in some ways revealing sort of the limits to the quiet revolution. Right. And can you explain what you meant by that and kind of how we might think about that?

Edwards: Yeah, I mean, I think maybe what I learned is Claudia Goldin is nicer than I am because she called it the Quiet Revolution. But I really it struck me as an unsupported one. Right. Women doubled their labor force participation and they did it without sick days and they did it without family leave and they did it without affordable, accessible child care. And they did it without, you know, longer school days. We the U.S. has never made the investments in working women the way that other countries of comparable income have. But the evidence base we have from those other countries and the investments they made is that if you know basically anything you do between eight, 30 and five, 30, that puts a kid outside of parents’ hands in a safe, affordable place, increases maternal labor supply. And there’s an article by Lisa Kahn, which I believe is in the American Economic Review, where she says that, you know, the U.S. used to be the leader of women’s labor force participation in over the past 30 years. We’ve fallen in this ranking because we’ve had we you know, strangely enough, we are more likely to have women working full time and we’re more likely to have women who earn more. But we have a lower labor force participation conditional on that. And so what that would suggest is that there are a lot of women who marginally could be brought into the labor force that we that don’t have the policies that support them. That’s one interpretation. There’s definitely someone out there who would say, no, of course not. We don’t need childcare for women to work. That’s crazy talk. But that’s what struck me about what Goldin found is that first, we have not made progress on our labor force participation since 1999. And second, we have not made progress on any type of formidable child care or family leave investment ever. And I’m reminded of, you know, my mother should have had these things. And I remember my mom telling me that when we were younger, there was only one daycare in our town and she worked in the city. And so that in order to make sure that we could get daycare, she slept outside of the daycare and in a tent on the front lawn so that when the line opened up, she would be there and she asked me if it was better now. And, you know, I would say, well, we don’t have to sleep outside. We just get some Internet queue. But aside from the physical conditions, no, it’s not. You know, it is that desperate for a lot of women. And I think here’s a good place to bring up that center based care in a lot of ways is luxury that you can afford it. A lot of children are in informal care arrangements with family members. And that’s another thing that the pandemic really jeopardized, because if my if my child care is my older mother who was in the covid, you know, high risk group, if I go to work, I’m either putting my mother at risk or I’m losing child care. Right. And that there’s this kind of cascades down from who has access to what based on what they can afford. And so that’s why I called it the unsupported revolution in my essay, was because there’s a shocking lack of support for working parents. And it’s often framed as working moms, but it’s truly working parents. You know, marriages are not the same as they were in 1950, and neither is women’s labor force participation. But on the margin this pandemic has shown, it’s still largely a story about mothers. But truly, it’s a story about parents.

Chancellor: So we’re not calling this the quiet revolution anymore—we’re calling this the unsupported revolution, but what do we do to make this a second revolution? You mean you hinted at this pretty strongly, in your last response there, what would you advocate for? What are the things that are going to make a big difference here?

Edwards: Well, I mean, I should caveat by saying that Claudia Goldin is a much better economist than me, and maybe when I become head of the American Economic Association and renowned in everything that I do, maybe then we can change the name that she gave. Up with it with that caveat that she’s really good at what she does. And so we can keep calling it the quiet one. You know, the big actors at the end of the pandemic are going to be the federal governments and firms. So, when the federal government side, the question is, are we going to make the investment that we should have made 40 years ago, are we going to have mandated paid sick leave for private sector workers? Are we going to have universal access to paid family leave? Are we going to have affordable and accessible child care? And are we going to work to make school match work days so that the school day matches the workday? Those are really questions for the federal government, states have made movements kind of on all fronts in a limited way, but if it’s about getting access to every worker, it would be a federal floor that’s established by the other actors. In all of this are firms and your employer determines a lot about your work fate, and we still have a glaring gap in wages between men and women. We still have a glaring lack of female leadership at private companies and then, you know, a glaring lack of female leadership in the public sector as well. And I would argue in academics, the question for firms is, you know, they typically utilize information about someone’s productivity and attachment to a job and filter that into higher wages or any filter that into wage determinations. So there was a there’s a paper that came out last year that looked at the amount of leave that women took for maternity. And they looked at the variation between women who took some of the leave that they were entitled to versus women who took all of the lead that they were entitled to. And they found that years later, the women who took all of the leave were paid less still. What’s interesting about that is that what the author interprets that as is signaling that firms interpret this as a signal of how committed you are to your job and that if the culture is to not take as much leave, even taking the lead that you’re entitled to is a signal that you’re not as good and that that makes perfect sense in the context of things like. Corporate law or finance these occupations in which people are expected to work and in the office late into the evening, right, it’s their productivity, their means of signaling their productivity is their presence after hours. Right. Which is hard for a parent to do, especially if the other parent can’t take the kid. So what makes the pandemic so special is that if you have kept your job and you’re like me and you’re in a closet, you know, I’ve spent a lot of signals to my employer over the past year. Right? They know what my husband looks like. They know how many kids I have and how many dogs I have. They know, like when my kid naps, like they know everything and they know how whether or not I’ve been able to maintain my productivity in the pandemic with all these extra demands on my time. And so the question is, how are firms going to interpret these signals? You know, we’re in a really incredible moment. A fourth of the U.S. workforce is at home right now. And when they get called back into the office, you know, this goes back to what kind of signal are they going to take to mean productivity? And it and it goes back to the fact that half of women work, half of the workforce in December 29th, 2013 were women. So firms have to decide, are women the exception or are they the rule? Are there measures of productivity built on signals that are just not going to be met by women, are not going to get my mothers or maybe not going to be met by the part of their firm that’s working from home permanently? And they have to. This requires very thoughtful design and this requires very thoughtful kind of adjustments and investments as to what is culture and what is what is productivity and how do we measure it. So, in some ways, the federal government has it a little bit easier because there’s a very large evidence base to draw from, you know, paid family leave and childcare will get more women to work. How firms do this, I think a lot of them are having to remake their culture and might not know where to begin. But a lot, you know, rest in their hands.

Chancellor: So what do you want to leave us with today?

Edwards: I would say two things, I think for a lot of parents and for a lot of women, it can feel like they may be. We’re talking about this choice that they made as if it’s really expendable and easy and, oh, yeah, they decided to leave. And, you know, I think it’s hard not to end up selling Kallus as you talk about these really momentous decisions a lot of women had to make. And so I think if I would leave you with any thought, it’s that I think a lot of women might tell you it didn’t feel like a choice at all. You know, we paint it in our discussions as you decided to leave the labor force. But if you’re backed into a corner, it doesn’t really seem like you have a lot of directions to go in. And I would argue that, you know, as the posture of the federal government, it has been one to really kind of put the blame on the personal failures of women rather than the policy failures that leave them with little options. You know, in welfare reform and the Personal Responsibility and Work Opportunity Reconciliation Act of 1996, the policy towards childcare is “get a man,” right? That that is that is what we told women, was that they need to get married. It is in the legislation. And it happened a month before the Defense of Marriage Act, where we made it clear that we wanted women to find husbands. And I think it’s hard now you twenty-five years later, to kind of realize what that posture has left us with. And it’s left us with a position where we tend to blame women for their choices as opposed to their lack of choices, you know, you’re a single mother, get a husband, you’re struggling, you’re married, but you’re struggling. Make your husband do more. Right. And I think that that type of judgment should be pointed upward to the people who can make a larger difference. Which is, if you can’t find childcare, call your congressman.