There is no study of what grandparents notice and keep to themselves. Nobody has sat a representative sample of them down and collected the observations they swallow at the kitchen table, so every list of this kind, including the ones written with real affection, is a list of guesses.
But one observation turns up in every version of it, and that one has been measured. It is the sense that this generation does more. More hours, more attention, more of the parent pointed at the child. Time-use diaries have been recording exactly that since 1965, which means the most common unspoken comparison in three-generation family life can be checked against numbers rather than argued about.
The observation is correct
Giulia Dotti Sani and Judith Treas pooled time-use diaries from 11 Western countries covering 1965 to 2012, and modelled daily child-care minutes while holding constant parents’ age, partnership status, number of children, the age of the youngest, employment status and whether the diary day was a weekday.
The Artful Age
A weekly letter on aging well, family across generations, and the creative life after the kids leave home.
For the average mother across those countries, predicted child-care time went from about 54 minutes a day in 1965 to about 104 minutes by 2012. It roughly doubled.
For fathers the change is larger in proportion and smaller in absolute terms, which is its own kind of fact. Sixteen minutes a day in 1965. Fifty-nine by 2012.
Two things about that are worth holding onto. The first is that these are diaries, not recollections: people recorded what they were doing as the day went, which is why this evidence survives the ordinary problem that everybody remembers their own era generously. The second is that all of it happened across precisely the decades when mothers moved into paid work in enormous numbers. The extra hours did not come from anywhere obvious.
The shape is stranger than either generation assumes
The American data has a twist in it that the pooled international model smooths over, and it is worth separating out rather than blending.
Garey and Valerie Ramey linked 13 American time-use surveys from 1965 to 2008 for a Brookings paper called The Rug Rat Race. In the United States, child-care time fell for roughly three decades first. Then, from the mid-1990s, it rose sharply, in an S-shaped curve that flattens in the early 2000s.
So the era a grandparent is comparing against was not a stable baseline. In the American series it was the bottom of a long decline, and the thing that happened afterwards was a reversal rather than a drift.
The size of that reversal was not evenly spread. Less educated mothers added more than four hours a week. College-educated mothers added more than nine. Fathers followed the same pattern at smaller magnitudes. And a substantial part of the increase was not time with small children at all. It was time with older ones, and specifically the business of coordinating and transporting them to their activities.
The Rameys put the total in a way that is hard to unsee: by 2008 the adult population’s child-care time amounted to nearly a fifth of all time spent in paid work, and the increase alone, valued at market wages, came to over $300 billion a year.
Why, and what it is not
The Rameys tested the explanations people reach for first. Selection effects. Income effects. Safety fears. Parents simply enjoying child care more. More flexible work schedules. They report that none of them fits the data.
What they propose instead is competition for university places. College-bound cohorts began swelling in the mid-1990s, at the same moment the curve turns, and their argument is that scarcity of good slots raised the amount of preparation a parent had to supply. Their most persuasive piece of support is a comparison with Canada, where there is much less of a national market for elite admissions. Since the mid-1980s, parents with less education increased their care time by roughly the same amount in both countries, about four to five hours a week for mothers. But college-educated mothers added almost nine hours a week in the United States and only one in Canada, and among college-educated fathers the figures were six and two. The educational gap widened in America and shrank in Canada. One caveat the authors flag: the Canadian measure covers care of adults as well as children, so the two series are not identical.
Dotti Sani and Treas, working from a different discipline and a different dataset, reach a reading that is no more comfortable. They see the rise as driven substantially by intensive parenting among the highly educated, a practice which, in their words, confirms privileged social status by differentiating those parents from the classes below them. They conclude that in some countries the trend appears to worsen the passing-down of inequality rather than soften it.
Neither of these is the story anybody tells about themselves. Nobody drives across town on a Tuesday evening thinking about positional competition for university slots.
What the evidence will not carry
Dotti Sani and Treas say plainly that their data fell short of what was needed to detect significant trends country by country, and that the confidence bounds for the two education groups often overlapped. The finding lives in the pooled model. French mothers were a documented exception to the pattern.
The Rameys are making an argument by elimination and comparison, not producing an identified causal estimate. They rule out five explanations and offer a sixth that fits the timing and the cross-country contrast, which is a good deal weaker than proof. They also note, in their own conclusion, that the race may not be wasteful at all, since parents may not fully capture the social benefits of preparing their children.
And neither paper measures a single child outcome. Nothing here says the extra hours are good for children, or bad for them, or neutral. That is a different question with a different and contested literature.
What this does to the sentence that never gets said
The noticing is accurate. That is the first thing, and it is not nothing, because most unspoken family observations are a mixture of something real and something projected, and this one turns out to be measurable and confirmed.
What does not survive is the interpretation usually bundled with it. The extra hours are not softness, and in the American data they are not mostly about hovering over a toddler either. A large share of them are a parent in a car at six in the evening, performing something closer to logistics than indulgence, inside a competition nobody in the family chose and few would defend out loud.
Which suggests the silence is guarding the wrong thing. The sentence that gets swallowed is a criticism, and a criticism is the one form of this observation that is both unwelcome and uninformative. The version that is neither is a question: not why are you doing so much, but what are all these hours actually for, and who decided that.
There is a last symmetry worth naming, and it is the reason this is not a scolding in either direction. The same diaries measured the other era too. Sixteen minutes a day, for fathers, in 1965. That number is also a thing somebody might notice and decide not to mention, and the fact that it almost never gets said from the younger side is the same courtesy, running the other way.