One city inadvertently allowed daycares on just 4 percent of its land. Until we fix rules like these, subsidies will mostly benefit the better-off families who can grab scarce slots.
You’re right. But also worth noting *why* they’re so prevalent - because earmarking $$$ to A Program is the electoral version of just slapping that purchase on your credit card. You get the benefits - your purchase/the political boost of being able to say that you Did Something™️ about homelessness or whatever - right now, and you get to defer the actual work (paying for that purchase/actually solving poverty) until Someday™️.
Another excellent piece. Some Australian perspective
* Proper parental leave would significantly reduce the need for under-2 daycare. Most/many new parents would prefer to stay home with infants but can't afford it. Australia not as good as many European countries, but much better than US. https://www.pm.gov.au/media/paid-parental-leave-expand-6-months
Oh fascinating about the over supply. The article says because of a specific policy and a specific framework - but I don’t really get the mechanism. Is it because government builds gov-run daycares?
Yes parental leave would help a lot! Infant care is really hard to finance, so mostly in in-home daycare in the US apparently!
I don’t understand the chronic oversupply story. Maybe just industry special pleading, but wouldn’t be credible if there were an obvious shortage of places
This gets to the heart of systems design on systems that aren't designed, but rather accrete over time. Most of the rules make sense on their own, but in aggregate are completely unworkable.
You write that "The lowest caps don’t track with safety (though large group sizes lead to more illnesses)." There's no source for this statement--the report that you link doesn't say this. Can you provide a source?
The paper says that not enough research has been done to know, in large part because there are ethical issues with setting up controlled studies on this. That's very different from saying that there's no relationship between caps and safety.
I focused on this because I know that staffing ratios are the single biggest driver of costs, and I was very interested to learn that they don't have a relationship with student safety, but I don't think that's the case based on the evidence here. As a parent of a two year old in daycare myself, I would need a lot more evidence to be comfortable if I heard that my jurisdiction was considering reducing staffing requirements.
I believe they clearly do have a relationship with safety when you compare to *no caps* (which was the norm before 90s). I spent most of my caps-research time looking for comparing between states' caps (since today, they all have caps, though with wide-ish ranges).
My reading of the data (I could be wrong, there's just not much data!) is that, within the existing caps that states have, the safety incidents are too low to find a difference. Most, if not all, of the studies of between-existing-caps are about "quality" like child attention, not safety. Which I also took as an indicator that safety was seen as moot or null.
Other meta-studies that look into the question of regulation impacts didn't find impacts from ratios, but rather other things.
Example: a review of accidental injury (data from the 80s and 90s) reviewed the impact of ratios, inspection frequency, and educational requirements - and found that the only thing to have a substantive impact was training by daycare center directors. I believe training is defined as some training beyond high school. Caps were much wider back then, so I'd take this also as strong evidence. My Claude summary says of the paper ratios had a small effect on accidents, but I don't see that in the paper abstract or results (Im paywalled), I'm guessing authors didn't mention this much if it is in there because it might not have been robust or because authors didn't consider this worth the effect of keeping kids out of formal care which had other protective effects. https://www.sciencedirect.com/science/article/abs/pii/S0167629603001012
Another more recent NY child fatality study of ~20 years of data found that being in a center was the most important factor, likely because there was more than one caregiver and more screening of caregivers (as opposed to in-home daycares). Having multiple adults seems important to prevent loss of emotional control, and abiding by rules (e.g. no blankets in cribs). https://www.fcd-us.org/fatalities-and-the-organization-of-child-care-in-the-united-states-1985-2003/
I realize that because incidents are so rare so ratios are hard to study - but the broader context is also that lots of studies did look into these and found that *other things* did show relationships. E.g. having multiple adults. If current ratios had a clear impact, they'd have shown up in any of the many studies. It's not the case that ratios haven't been extensively studied.
I don’t agree with your conclusion in your last paragraph—absence of evidence is not evidence of absence. However, I really appreciate all the research you’ve clearly done on this and other topics, and that you’re sharing it with your readers.
I was literally going to be like "Abi should talk to Allister" about this, and...lo and behold! This is such a high-leverage issue because so much of the child care conversation surrounds e.g. childcare subsidies — which are great, for sure, but which presumes the capacity to construct supply, which by no means can simply be postulated!
Reformas would help, but a better approach woud be a child allownce that parents could use for formal. informal or parent provided day care. Let's make a familily allowance par of our socia insurance sustem just like Social Security, Unemployment Insurance, of Medicare, paid for with an earmarked tax soure, a VAT.
I think you missed the main point: that money is only a part of the problem, and probably not the main part. Abi is describing all the self-inflicted damage imposed through badly designed efforts to address "safety."
"Until we fix rules like these, subsidies will mostly benefit the better-off families who can grab scarce slots."
TY! These types of "sounds good!" policy proposals are so prevalent (and tend to be most noticed / pushed by those who will take advantage of them).
Thanks Matt!
You’re right. But also worth noting *why* they’re so prevalent - because earmarking $$$ to A Program is the electoral version of just slapping that purchase on your credit card. You get the benefits - your purchase/the political boost of being able to say that you Did Something™️ about homelessness or whatever - right now, and you get to defer the actual work (paying for that purchase/actually solving poverty) until Someday™️.
Or even checking whether it had the intended impact! That should be part of the process.
“Why the hell would I audit results?? The results could be bad!”
-idk Gavin Newsom probably
Great specific suggestions!
Thanks Ben!
Another fking banger! Ty for this
Thanks so much Richard!
Another excellent piece. Some Australian perspective
* Proper parental leave would significantly reduce the need for under-2 daycare. Most/many new parents would prefer to stay home with infants but can't afford it. Australia not as good as many European countries, but much better than US. https://www.pm.gov.au/media/paid-parental-leave-expand-6-months
* At least according to the industry, we have a chronic oversupply of places https://nsw.childcarealliance.org.au/news-current/why-oversupply-is-occurring-when-will-it-stop
Oh fascinating about the over supply. The article says because of a specific policy and a specific framework - but I don’t really get the mechanism. Is it because government builds gov-run daycares?
Yes parental leave would help a lot! Infant care is really hard to finance, so mostly in in-home daycare in the US apparently!
I don’t understand the chronic oversupply story. Maybe just industry special pleading, but wouldn’t be credible if there were an obvious shortage of places
Another terrific piece. Now my favorite substack and I read as lot!
Aww Robin! Thank you!! This piece in particular took me a looot of time so this means a lot!! :)
This gets to the heart of systems design on systems that aren't designed, but rather accrete over time. Most of the rules make sense on their own, but in aggregate are completely unworkable.
You write that "The lowest caps don’t track with safety (though large group sizes lead to more illnesses)." There's no source for this statement--the report that you link doesn't say this. Can you provide a source?
Sure - the link under illnesses had looked at safety and didn't find conclusive directions apart from illnesses. :)
The paper says that not enough research has been done to know, in large part because there are ethical issues with setting up controlled studies on this. That's very different from saying that there's no relationship between caps and safety.
I focused on this because I know that staffing ratios are the single biggest driver of costs, and I was very interested to learn that they don't have a relationship with student safety, but I don't think that's the case based on the evidence here. As a parent of a two year old in daycare myself, I would need a lot more evidence to be comfortable if I heard that my jurisdiction was considering reducing staffing requirements.
I believe they clearly do have a relationship with safety when you compare to *no caps* (which was the norm before 90s). I spent most of my caps-research time looking for comparing between states' caps (since today, they all have caps, though with wide-ish ranges).
My reading of the data (I could be wrong, there's just not much data!) is that, within the existing caps that states have, the safety incidents are too low to find a difference. Most, if not all, of the studies of between-existing-caps are about "quality" like child attention, not safety. Which I also took as an indicator that safety was seen as moot or null.
I see what you’re saying, but it’s certainly not enough to convince me that kids’ safety wouldn’t be compromised by increasing ratios.
Makes sense. For context, I was also basing this on Niskanen's report on the range of ratios (https://www.niskanencenter.org/states-have-room-to-experiment-with-childcare-staffing-ratios/) which cited the same study. That study seems to be the largest meta-analysis, and it found weak evidence even about social-emotional health and cognitive development from ratios.
Other meta-studies that look into the question of regulation impacts didn't find impacts from ratios, but rather other things.
Example: a review of accidental injury (data from the 80s and 90s) reviewed the impact of ratios, inspection frequency, and educational requirements - and found that the only thing to have a substantive impact was training by daycare center directors. I believe training is defined as some training beyond high school. Caps were much wider back then, so I'd take this also as strong evidence. My Claude summary says of the paper ratios had a small effect on accidents, but I don't see that in the paper abstract or results (Im paywalled), I'm guessing authors didn't mention this much if it is in there because it might not have been robust or because authors didn't consider this worth the effect of keeping kids out of formal care which had other protective effects. https://www.sciencedirect.com/science/article/abs/pii/S0167629603001012
Another more recent NY child fatality study of ~20 years of data found that being in a center was the most important factor, likely because there was more than one caregiver and more screening of caregivers (as opposed to in-home daycares). Having multiple adults seems important to prevent loss of emotional control, and abiding by rules (e.g. no blankets in cribs). https://www.fcd-us.org/fatalities-and-the-organization-of-child-care-in-the-united-states-1985-2003/
I realize that because incidents are so rare so ratios are hard to study - but the broader context is also that lots of studies did look into these and found that *other things* did show relationships. E.g. having multiple adults. If current ratios had a clear impact, they'd have shown up in any of the many studies. It's not the case that ratios haven't been extensively studied.
I don’t agree with your conclusion in your last paragraph—absence of evidence is not evidence of absence. However, I really appreciate all the research you’ve clearly done on this and other topics, and that you’re sharing it with your readers.
I was literally going to be like "Abi should talk to Allister" about this, and...lo and behold! This is such a high-leverage issue because so much of the child care conversation surrounds e.g. childcare subsidies — which are great, for sure, but which presumes the capacity to construct supply, which by no means can simply be postulated!
Reformas would help, but a better approach woud be a child allownce that parents could use for formal. informal or parent provided day care. Let's make a familily allowance par of our socia insurance sustem just like Social Security, Unemployment Insurance, of Medicare, paid for with an earmarked tax soure, a VAT.
I think you missed the main point: that money is only a part of the problem, and probably not the main part. Abi is describing all the self-inflicted damage imposed through badly designed efforts to address "safety."
I agree and let's try to refome that, but a child allowance lets parents work around some dysfunctional systems.