The economics of late grandmotherhood

We show how fertility delays compound across generations to raise age at first grandmotherhood. Under the 1960 first-birth schedule median grandmotherhood came at 44. Under the 2024 schedule it is 57, cutting the healthy years with grandchildren from 28 to 15. Longer life does not make up the loss of years when the grandmother is healthy enough to supply money and care. Delay is most pronounced among the college-educated, with reproductive-control access accounting for over half the fall in grandmotherhood by age 55 among White college graduates. When a daughter times her first birth to maximize her own payoff, the loss her choice imposes on her mother is external to her calculation. We call this the Rotten Adult Kid Problem. Delay raises both the price of a grandmother’s care and wealth per grandchild.

That is from a new NBER working paper by Dora Costa and Matt Kahn.

Comments

And great grandparents as well. My wife was cared for by her great grandmother while both her mother and grandmother worked. There was far more family based care in the past that us not available today.

Respond

It holds for all relations - aunts, uncles, cousins - as you disperse the family temporily it is ever less likely that your kids will have meaningful time with somebody in each role. The simple coordination hack is folks getting married in their mid 20s and having kids shortly thereafter. Coordinating beyond that is very hard - and gets harder as jobs and other life concerns tend to pull families apart as time goes on.

There is a huge amount hedonic utility that the average family forgoes by adopting norms of delayed coupling, marriage, and childbearing. It gets worse if you live far away (age robs us of long distance travel before it robs us of transversing the neighborhood) and worse still if you have a demanding profession.

Respond

The extended family is a holdover from our last quarter million years of evolution. Speaking with Charlie Marx: The bourgeoisie has saved many from the idiocy of village life.

And if you disagree, you don't know my extended family.

Respond

Quite possibly. Yet people spend good money, and more valuably, vacation time to attend family reunions and the rest. And those who are better attached to their extended families, on average, live longer and all the other nice correlates.

Honestly though, seems like the optimal setup would be for careers and other major life choices to easily allow folks to choose to stay near family or not. To delay childbearing or not. But aside from some Dues ex machina intelligens, the current setup were we eat the seed corn of future workers is a recipe for bad times to come.

Respond

Whether extended family is vestigial baggage or a survival mechanism depends on the setting. If you live in a low-trust country with limited State capacity, then extended family is how you navigate the territory.

Respond

Absolutely! And it depends on which family.

Respond

Oh, that was your extended family?

Respond

Cf. Tyler's recent Conversation with a historian who argues that Europe outpaced China after 1400 because its society was based on nuclear families instead of extended-family clans.

Respond

My wife was partially raised by her grandmother and, because of this, convinced me to start our family earlier than I did. It was tough but, now that I now appreciate the time with grandkids while I can still play with them.

Respond

Barack Obama's American grandmother was 39 when he was born. He benefited a lot from having a vigorous grandmother in her prime.

Respond

That is a good example that illustrates the nuances of this issue. You are more likely to have a "vigorous grandmother" raising you if you are the product of a teenage romance. However, leaving aside the (controversial here) question of whether college-age people should be getting married and having their first children, there is also the issue of the grandmother's own life and preferences. I don't think every 40-something woman who has just finished the difficult task of raising children wants an immediate repeat of the experience. Most women will step up over a sense of duty but 19-20-year-olds are not fully formed adults so the grandma is effectively raising two generations of kids instead of just one.

Respond

We call this the Rotten Adult Kid Problem

The average age of mothers giving birth in the US is now 29.6. This paper is labeling the typical US mother as a rotten adult? Really? Is this how economic papers write?

To have an average first-birth grandmother age of 44, would require an average first-birth mother age of 22. Having young, healthy grandmothers sounds great, but having a successful marriage and start having kids by age 22 is rather difficult in today's society.

Respond

Making kids is easy. I’m fairly old but I’m pretty sure I remember how. There’s a widespread belief that successful marriages and successful parenting require age and financial security that were not commonly present in any age before very recent times. I would counter that most successful marriages were started before the partners were fully formed and set in their ways. The foundation of the marriage were the hurdles and challenges overcome together. Further, the least successful marriage is the one which never happened, and the least successful child is the one who was never born. Both of those hidden failures multiply when young people delay marriage and children “until they’re ready.” Nobody is ever truly ready for either adventure until they have been undertaken. Nobody gains the maturity desired through study or preparation. Marriage and parenthood are both learn by doing skill sets. Delays in marriage and parenthood are both indicative of a general societal unwillingness to take the necessary risks for success.

Respond

That name is probably a riff on Gary Becker's *rotten kid theorem* which would be familiar to many people who got econ PhDs in a certain era (but perhaps few other people!). That theorem showed that, under the right conditions even a selfish, no-good rotten kid would behave kindly towards his parents.

This problems seems to cut the other way, but I think the reference to theorem is what was driving the name.

Respond

I think this might be a joke paper? Dora Costa and Matt Kahn are married. Maybe, this is written for a (presumably childless) daughter? I think this sentence gives it away: "When a daughter times her first birth to maximize her own payoff, the loss her choice imposes on her mother is external to her calculation."

I hope this paper is revealed as parody, given the commenters on here that appear to be (supportively) taking it seriously! Imagine taking it seriously nowadays that one should want to become a grandparent in one's forties!

Respond

These two appear to be serious: [https://matthewekahn.substack.com/p/delayed-grandparenthood-an-economic].

Respond

And, in addition, moreover, the external cost is merely an emotional one. Reminds me of Amartya Sen's paper about the impossibility of a Paretian [classical] liberal.

Best to ignore those costs.

Respond

This is an interesting/compelling thought that I’d not considered before.

Thank you for sending!

Respond

My mother scooped these two by about ten years, purely via women's intuition.

Respond

Yeah, but grandmothers had more grandchildren in 1960, so presumably time and money would be divided.

A grandma in 1960:
Had 2.5 children
Those children had 3 children
7.5 grandchildren for 28 years = 3.74 child-years

A grandma in 2024:
had 2 children
those children have 1.7 children
3.4 grandchildren for 15 years = 4.4 child-years

Years per grandchild have gone up.

Respond

57 vs 44 is a significant difference in capability for the grandmother. How do you subtract that out of the greater child years?

While life expectancy continues to lengthen, the age of peak ability doesn't lengthen nearly as fast. And of course with fewer children and a smaller age range of children, while in the past the oldest kids may have assisted with child care, today the age difference is rarely more than 4-5 years, not really enough for hte oldest to have substantial knowledge.

Respond

The 28 versus 15 years cuts off at 72 - mostly still pretty capable, but wide variance. And older Americans are healthier now than they were in the 60s.

Respond

I probably know of more cases where the child has to take care of the parent than the grandparent helps significantly with the grandchildren and of the cases where the grandparent is significantly involved-all of them are a 1st generation immigrant in-law. My anecdotal experience is the right side of the bell curve but nowhere near elite; the healthier the grandparents are the more they are still focused on their own consumption and self-actualization and avoid childcare, generational compounding of delaying fertility leads to functional alienation even if everyone is alive. It makes it even harder and less appealing for the women to become parents or have more than one child. I expect fertility numbers to really tank as my cohort and below rolls into the completed fertility column in all of these data sets, like my cohort from high school looking like a TFR of Shanghai. Priors need to be updated.

Respond

This raises misogyny to a whole new level: selfish grandmothers. My grandmother didn't have her first and only child until she was age 42. Then I wasn't born until she was 84. Yet, she had enormous influence on my life. Why? She was a physician, practiced until she was 80 years old. I think of her every day of my life, and whether she would be proud of me. My grandmother was way ahead of her time, and I tell people who ask that she was a product of the 60s. As they are doing the math in their heads, I tell them, yes, she was a product of the 60s: she was born in 1868. No, she never changed my diapers but she is my inspiration.

Respond

I had old grandparents too. What people don't get is, if you have 8 freaking kids that survive, the last one gonna be born when you're pretty old. That's been going on for a long friggin' time.

Respond

What people don't get is, if you have 8 freaking kids that survive, the last one gonna be born when you're pretty old.

What, in your thirties? You set a low standard for "pretty old".

Respond

Your grandmother sounds remarkable, and her influence shows that grandparents contribute in ways beyond childcare. But her experience doesn’t erase the tradeoff the paper examines.

As coauthor Matthew Kahn writes: “We acknowledge in our paper that we are averaging across different adult women with different aspirations and life goals.”
https://matthewekahn.substack.com/p/delayed-grandparenthood-an-economic

Having children later can reduce the healthy years grandparents and grandchildren share. Why is it misogynistic to study that effect?

Respond

Paraphrasing Brando in "The Wild One":

"Hey, Johnny, against what theory are you rebelling against?"
"Whaddya got?"

Respond

Does free family childcare provided by grandparents show up in GDP? If not, isn’t it worthless to a GDP-focused economist?

Respond

It probably subtracts from GDP, as the kid isn’t in daycare and the grandma isn’t in assisted living.

Respond

So family is a net negative or GDP sometimes misleads. I am pretty sure it’s the latter.

Respond

Interesting, but they're looking at just one grandmother's lifespan. Children have two grandmothers, who thanks to lower mortality rates are more likely to be alive compared to decades ago.

So children born in 1960 might've had younger grandmothers, but might've been less likely to have two or even one grandmother still alive.

But factors that I suspect are bigger are these: how many of the grandmothers live within say an hour drive of the child? Until recently, increased geographic mobility meant people were more likely to live far from their parents and hometowns. Tyler's posted articles however that say that geographic mobility has diminished in recent years.

And the role of divorce. If there's one divorce and one remarriage by one divorced parent, maybe the child lost access to half of their grandparents.

Or maybe the child now has eight grandparents.

I don't know what the net effect is, but it could be significant.

Respond

P.S. We see both patterns when looking at students' applications for financial aid. They might list financial resources that include grandparents from more than two households chipping in. Or they might include a note from a single parent (usually a mother) saying that the national financial aid formulas that call for $X of support from parents do not work because the ex-husband is refusing to help pay for the kid's college tuition. Again I don't know what the overall average is.

Respond

What kind of idiot explicitly lists grandparent support on financial aid forms? All this means is that the school measures less need and gives you less aid.

Respond

An honest, law-abiding person?

Respond

How common was it for grandchildren to never know all 4 grandparents because one or more died before the children were born? Happened to me and I think that was quite common. I’m in my 40s and my parents were 23 when they first became parents

Respond

The economics of late grandmotherhood

Nothing that can't be solved by Carousel.

Respond

Very interesting paper—and contrary to my intuition, which is: the older the grandmother as a nanny, the cheaper. Children can intertemporally optimize when to have children themselves, waiting to buy grandmother-care services more cheaply as her opportunity cost of time falls.

But this does not perpetuate delay. If you were born to an older mother, your mother is already older when you reach childbearing age—so you have less reason to wait. Delay today therefore creates a force toward less delay tomorrow.

Respond

Grandma can't complain too much. It's half her fault.

Respond

I kind of assumed that grandparents were more likely to help with childcare etc. if they were a bit older themselves.

A grandmother in her fifties is pretty likely to still work (quite often full time), whereas in her mid- to late sixties she is more likely to have retired and therefore have the time (and in many cases health and inclination) to be involved in the care of her grandchildren. Same for grandpa.

It wouldn't surprise me if having grandchildren would turn out to be a contributing factor to older women leaving the workforce (thereby helping out their children who struggle to afford childcare). If that's the case, later children = more older women in work.

Respond

Even when the grandmother is healthy, it doesn't help if she is thousands of miles away, and many people move to large metro areas, where the jobs are, while the grandparents are elsewhere, and its not that they can afford to move to a HCOL location. In those cases, it's still all heavy parental effort, or expensive prices.

Respond

It's a crapshoot, like so much in life. My grandparents on my mother's side were both 57 when I was born. Fairly old, on average. My grandmother was in poor health as long as I remembered and died at 72. My grandfather lived to 90 and was quite active until at least 85.

Respond

Grandparents? There's 70 year olds raising K-12 age kids (adopted).

Respond

The article (or the above summary) overstates the case. The summary states that the healthy years for grandmothers were reduced from 28 to 15 between 1960 and 2024. That statement doesn't take into account the longer life span of females in 2024 than in 1960.

Women in 1960 lived an average of 74 years. Women in 2024 lived an average of 81 years. Thus the number of years a grandmother can be helpful should be increased by 7 years for 2024, e.g. 28 years in 1960 vs 22 years in 2024.

Respond

Not dying and being able to healthily care for a child are different things. Sure health care has probably improved how many times a 70 year old can bend down and pick up a toddler, but that doesn't mean that an 80 year old is going to be doing it well.

Respond

The number of women who have the balance, mobility, eyesight, and strength to handle infant care into their 80s is very, very small.

While we are also extending the healthspan when lifespan goes up, it does so much more slowly. A lot of the earlier mortality was simply something killing people before they were old enough to have difficulties with complex tasks. But even if we keep you alive by treating your heart attack better, you still going to have to deal with decaying eyesight, muscular atrophy, and the rest.

And with geographic dispersion, I would expect this to fall even further as significant travel is much more trying on the elderly than popping by another house in the neighborhood.

Have your kids young - in you mid 20s. Your job will hate you, but everything else will be easier.

Respond

Have your kids young, because it's a lot more fun to be a young grandparent.

Respond

80 is the new 70.

Respond

Derp, they mention this in the extract Tyler posted. Longer life =/= healthier life. And some point dem bones give up.

Respond
Respond

Add Comment