Should there be a 'supply-side abundance agenda' for childcare?

Fertility rates are declining and, from what I can tell, we don’t really have a definitive handle yet on why. Many potential culprits have been offered—higher opportunity costs for women, delaying pregnancy making it harder to conceive, and a decline in coupling accelerated by smartphones, to name a few. The underlying causes for a topic as complicated as this are almost certainly multifactorial.

One potential contributor I’m especially interested in is the rising cost of childcare. Intuitively, if real costs for childcare have gone up dramatically, and the opportunity cost of having kids has gone up, it stands to reason more potential parents (especially women) would be more hesitant.

The majority of Americans say that providing free childcare would encourage more people to have children. Personally, while I don’t yet have kids, I plan to at some point. As I’ve done my own research on the question of “how much should I budget for childcare over the next ~18 years if I still want to work and also be able to keep investing in other important parts of my life?” one quickly arrives at a version of: “wait, how is everyone affording this?!?” Now we’re in the territory of making meaningful tradeoffs. “OK I could move closer to my parents and out of San Francisco, but I love San Francisco. OK, I could relax the goal of wanting to spend time with my friends/pursue other interests, but I’d really rather not. OK fine, if I spend a significant portion of my income on childcare, what are the things I can no longer do?” Most current or potential parents have worked through some version of this reasoning.

To be clear, I’m not opposed to making tradeoffs (life is famously full of them!). But I do wonder if we can think more expansively here…

If childcare cost does turn out to be a significant driver of declining fertility rates, what can we do about it?

Why have childcare costs gone up? Economists seem to think that the vast majority of increasing costs can be attributed to supply-side problems (e.g., the Baumol effect), with demand playing a role but a smaller one (e.g., expected hours your kid is supervised and how attentively) [1].

Intuitively, at least part of the story here is that historically significant sources of childcare ‘supply’ have gone down—namely many women choosing to work instead of choosing to be full-time parents, as well as grandparents/other family members living farther away (or are older because people are having kids later). This of course pushes the source of care to paid sources of labor—nannies, daycare, etc.

OK, so let’s focus on the supply-side problem for now. Before getting to potential solutions, let’s first describe some basic characteristics of a good solution:

1. Kids have the amount and type of care they need

2. Parents have real, attractive choices about who provides that care and in what quantities (i.e. if one or both parents want to work, they have good, affordable alternatives)

3. Caretakers are fairly compensated for their work

If those are the goals, what are some options? A non-exhaustive list to get us started:

Category 1: Increase the ‘supply’ of labor devoted to childcare

1. Encourage more parents to go back to being full-time parents. If a parent wants to be a full-time parent I am all for it. But I will admit to being skeptical of (and increasingly concerned by) some of the culture/messaging we’re seeing in this category—specifically the recent-ish uptick in the whole tradwife meme. Surely we can be more creative about solutions here than just trying to culturally meme women into going back home…

2. Make it easier/cheaper for grandparents to live nearby. I’d be curious if California’s ADU initiative had any noticeable effect on birth rates (from what I’ve read, too small/noisy to say). Intuitively it feels like the potential impact here might be modest at best, in part because grandparents are on average older and less capable of the physical work involved in childcare.

3. Supplement caretaker wages to attract more people to this line of work without raising prices for parents. The Nordic countries do a lot of this. Seems promising.

4. Make it easier/more attractive for caregivers from other countries to work in the US. What would a well-designed nanny visa program look like? This seems like a very promising area to dig into (though I suppose at some point this could be detrimental to the countries people are coming from). There’s probably a lot we can learn from (and edit) from what Singapore and Hong Kong have tried.

5. Create net new caregivers with the help of AI/robots. I’ll be the first to admit that robot nannies seem very far off. I am also not suggesting the role of parent be fully outsourced to a robot (or anyone for that matter). But I think we have to at least acknowledge that this is the only idea so far that scalably brings net new caregiver ‘supply’. Maybe AI nannies can watch a sleeping kid while a parent gets dinner with friends (basically a more capable monitor), or help with cooking/cleaning that often falls to the mother, etc.

6. …what else?

Category 2: Increase the ratio of child to caretaker

6. Reduce/edit regulation that limits child-to-staff ratios to further amortize the cost of caretakers over more kids. Some ideas suggested here. Of course, eventually there are diminishing returns to the child here, but there’s likely some room here.

7. Improve housing structures and density impact how easily families can share childcare costs—harder to scalably implement in existing cities, but perhaps a sneaky important design principle to consider in parts of the world that are likely to house a lot of people in the future.

8. …what else?

Should there be a ‘supply-side abundance agenda’ for childcare?

If rising childcare costs do turn out to be a significant contributor to declining fertility rates, then it makes me wonder what a supply-side abundance agenda for childcare would look like that considers both what’s possible near-term and long-term. I’m especially curious about (4) and (5) from the list above.

[1] ChatGPT