Get access to over 1,300 articles and reader Q&As
Almost everyone reading this will be familiar with the depressing state of family leave in the United States. In contrast to many European countries, which typically have a minimum of four to five months of paid parental leave (and often much more), the United States has no paid maternity leave. The Family and Medical Leave Act guarantees some unpaid leave, but it doesn’t apply to all employers, and unpaid leave is often very difficult for families to afford.
If we all understand that paid leave is poor, what is less well understood is exactly how this translates to behavior. When do parents actually go back to work? This has generally been difficult to understand in detail due to data limitations.
Today, I want to talk about a new paper that provides our best descriptive evidence to date on parental leave-taking patterns in the U.S. I hope you’ll read this whole article to get the full picture, but in case you do not, here are two striking facts that I want you to know.
- One quarter of employed mothers are back at work by week three after giving birth.
- Fathers take, on average, three days of parental leave.
To me, these facts are a call to action that something must be done. More on what we might do at the end, but first, let’s look at the data.
How was this data collected?
The data here comes from a new paper by three researchers at the University of Nebraska; you can read the full (short!) paper here. The most important innovation in the paper is how they collected the data in the first place.
The authors start with birth records from 2005 through 2019 from the U.S. Social Security Administration. These data link children to their parents. They combine this with data from the American Community Survey, which is a survey run by the U.S. Census on an ongoing basis. This survey asks, among other things, about employment and work hours in the week prior to taking the survey. The researchers themselves do not see any identifying information about people, but the data is linked on the back end by the government.
By merging these data together, the authors are able to look at employment and work hours around the timing of childbirth. Put simply, they can see a child born based on the Social Security data, and then match it to their parents’ (if they are surveyed) employment history around the time of the child’s arrival. This allows the authors to look in detail at work behavior week by week around birth.
What does leave-taking look like on average?
In a word: short.
The graph below shows maternal and paternal employment and work behavior for the weeks around childbirth. For both men and women, employment is relatively flat. Although women’s employment slightly declines towards and shortly after birth, it is largely stable, similar to that of men.
Although there is little change in the share of women who say they have a job, most are not reporting work hours in the week right around birth. This does not go to zero, suggesting either noise in the data or that women have continued to work to some extent, even immediately after birth.
In these data, by week three, we are seeing a significant return to work, and by week 12, return to work is nearly universal. A skeptic may look at this data and say, look, maybe when people say they are working, they are really just sending a couple of emails, especially when we’re talking about three weeks after their baby was born. But the authors of this paper also looked at data from time use surveys, and in those, they see that moms of infants who say they are working are working on average five hours a day (this is in contrast to virtually zero hours worked for women who are on leave).
For fathers, we see a very small blip in work right around the timing of a child’s arrival, but they are quickly back to work.
It is worth pausing on this fact to paint a bit of a picture. Three weeks after birth, the vast majority of women are still facing significant physical recovery. They are very likely still having heavy bleeding and may have stitches. With a C-section, significant rest will still be recommended. Babies this age are very small, and breastfeeding is still being established. You’re not sleeping. Finding child care for a three-week-old baby is an enormous challenge. It feels fundamentally inhumane to ask people to go back to work at this stage.
How does leave-taking vary by state?
Attempts to pass paid parental leave at the federal level have been unsuccessful, but a number of states have passed their own legislation providing parental leave. The existence and extent of this leave vary by state. Unsurprisingly, leave-taking patterns also vary by state.
The map below shows the average number of weeks of leave taken by mothers across states.
This pattern across states partially reflects available paid leave. States like California, Rhode Island, and New Jersey have the highest leave-taking and also have (relatively) generous paid leave. There is variation beyond this, of course. New moms in South Dakota take an average of 4.2 weeks of leave, versus 6.9 in Pennsylvania, even though neither of these has paid leave at the state level.
Even with paid leave, new moms often do not take as much time as they could. Women in California are eligible for 18 to 20 weeks of paid leave by combining different programs, but on average, they only take 9.7 weeks. This could reflect a lack of information or difficulty accessing leave even when it is available. Many of these programs also don’t offer full income replacement, so it could also be a financial decision.
Can we do anything about this?
These data paint a somewhat dire picture. Our lack of paid leave for new parents is translating to people going back to work long before it is reasonable to expect them to be ready. We know from other work that maternity leave contributes to positive outcomes for both moms and babies. And we know that leave for dads can improve health for moms. There are many good data-based reasons (even putting aside the human reasons) to try to change this.
At the federal level, many bills have been proposed to create federal paid leave. They have not, so far, gone anywhere.
But all is not lost.
First, there is much more possibility for success at the state level; 13 states in the U.S. have at least some paid leave, and this number has been growing every year. This shows it is possible — now we just need to grow that number (Moms First is a great resource for figuring out what’s happening at the state level).
Second, these data suggest that even when people do have access to paid leave, they do not take all that they could. There are many reasons for this (workplace norms possibly being one of them), but a big one is probably knowledge about what is available. There is space for supporting information campaigns and other efforts to make sure everyone knows what is available to them.
One of the things this paper shows is that the extent of leave-taking has not changed very much over the past two decades. We cannot change that, but hopefully, we can make changes for the next two.
The bottom line
- Using linked Social Security and Census data, authors of a new paper were able to track parents’ week-by-week work patterns around the birth of their child.
- On average, U.S. mothers begin to return to work within three weeks of birth; by 12 weeks, nearly all are back to work.
- For fathers, we see a very small blip in work right around the timing of a child’s arrival, but they are quickly back to work.
- Leave-taking varies widely by state, but even where generous (by U.S. standards) leave is available, many new mothers take far less than they could.
- Expanding state programs and improving awareness could meaningfully increase leave-taking and its benefits.
Log in
This is incredibly sad for US families. In Australia, we are generally well past the point of providing adequate leave for birthing mothers and are shifting toward providing leave for other parents as well which is key for equality. The government provides 26 weeks paid leave at minimum wage split between two parents. In addition to this, my company provides 26 weeks full paid leave, and my husband’s company also provides 26 weeks full paid leave if he becomes the primary caregiver within 24 months.
This has allowed me to take almost a full year of leave with minimal disruption to my pay, and him to take another 6 months before having to put our child in (subsidised) childcare.
It would be interesting to see if there is any research done on the success of these parental leave policies and the economic outcomes.
I’m curious, do we have any data on how many people would like to take more leave if it were available? Also, I wonder how many people work in professions where leave is very difficult to accommodate (thinking of people like solo practitioner dentists, who may lose their business if they take significant leave.) I am 100% for generous leave, just would like to add to the picture.
In WA, if you own your own business or work for a small business, you don’t qualify for paid leave. A dentist owning their own practice would be responsible for finding a sub, working part-time, and/or scheduling patients around the leave. The nice thing about parental leave is that you know approximately when it will occur.
My hair dresser, who has her own business by herself has had 2 kids in the past 4 years. Both times she just didn’t have hair appointments for 6 months and gave recommendations for other hair dressers. It was about if a sad time for my hair, but it seems to have worked well for her and she’s done having kids now. Not sure how the finances worked, but she’s still in business!
I’m Canadian and about to return to work after 11 months. We have up to 18 months of federal job protected and subsidized leave here. I say subsidized because government pay tops out at $3000 per month (12 months of leave or less) or $1800 per month (12-18 months of leave). While these amounts look massive compared to what is offered in the US (essentially nothing) for many earners it is significantly less than what they would bring in while working. Many employers therefore top up these amounts, bridging the gap between the government subsidy pay, and the parents’ full salary. I’m a lawyer at a large corporate firm and was paid my full salary for 8 months.
Living through this past year, my first as a mom, I have not been able to stop thinking about parents in the US and how they can possibly make this situation work. Childcare costs are astronomically high, family help cannot always be relied on, and I can’t physically imagine having to go back to work so early after birthing a human.
All this to say — something MUST change. There are so many different models to choose from, since almost all other developed countries have some form of paid leave. I hope to see this become a serious matter brought forward in federal politics. If that feels like a pipe dream, remember that it was here once too. Now we have a system that, while it could be better, provides a meaningful support for families. We also got $10 a day daycare subsidies (success of the roll out still TBD) but that’s a fight for another day.
This is so fascinating to me. I am a lawyer in the U.S. and had my second child in November of this past year. While there is no formalized policy, I believe my employer would have paid my salary for up to 12 weeks’ time while on leave. The catch is that this would have directly impacted my year-end compensation and compensation for the next 3 years (as our compensation formula is based on 3 years’ of annualized data). In other words, I would still get paid day-to-day, but would effectively self-finance this period of leave through total compensation over the course of many years. The reality is, as the primary income earner in my household, that was not a gamble I could afford so I billed from the hospital room just prior and shortly following delivery. I worked remote, but on essentially a 3/4 basis for 12 weeks. It was awful. I tried desperately prior to my leave to convince my partners to implement a formalized leave policy. I proposed 12 weeks of guaranteed leave and that a partner taking leave during that time period would not be assessed their portion of firm overhead. In other words, the “cost” of leave would be shared equally among the remaining attorneys. I got absolutely nowhere. I would love to pick yours or anyone else’s brains about how to effectuate positive change in my own firm and law firms, generally, in the United States. I think we have to attack this issue from all fronts-local, state, and national levels AND demonstrating to employers why it makes financial sense for them, too.
Genuine question from a Canadian non-lawyer to the US lawyers: How is this multi-year pay penalty not considered gender-based discrimination?