It’s even worse than previous research suggested (and
it’s not great for white kids either).
|Graphic from Frank Edwards, et. al., "Contact with Child Protective Services is pervasive but unequally distributed by race|
and ethnicity in large U.S. Counties" (Proceedings of the National Academy of Sciences, July 27, 2021).
CORRECTION: Due to a data coding error, the authors of the study cited in this post have corrected the figures they originally presented. The new data show that the invasion of Black homes by family policing agencies investigating alleged child abuse or neglect is still pervasive, ubiquitous and extremely high - but not quite as high as originally estimated. They found no errors in their figures for "substantiations" foster care placement or termination of parental rights. Figures in the post below are corrected accordingly. The link to the study now goes to the corrected version. The graphic above also is from that version.
By now, people working in “child welfare” know, or at least
should know, of the national study published in 2017 which
revealed that about one-third of all children and more than half of Black
children will be forced to endure a child abuse investigation at some point
during their childhoods. Earlier this
year, a California study produced similar results.
But, at least when it comes to most of America’s largest
counties, these data underestimate the problem.
A new study
of data from America’s 20 largest counties (with the five counties of New York
City counted as one) reveals a system that is arbitrary, capricious, cruel and
very, very racist.
The study reveals an infrastructure of surveillance of
families – especially nonwhite families -- that is, in the words of the study
authors “pervasive” “ubiquitous” and "extremely high."
The data tell us that, step by step, brick by brick over
more than half a century America has built a monstrous machine inflicting
state-sanctioned emotional child abuse on a huge proportion of nonwhite
For example: The study estimates that in Los Angeles County, about 58%
72% of all Black children – fifty-eight percent seventy-two percent – will be forced to
endure a child abuse investigation during the course of their childhoods. Neither this, nor the previous studies,
provides data specific to income, but of course, we know that the widespread confusion of poverty with neglect means we’re
talking almost exclusively about impoverished families. So if 72% 58% of all Black children in L.A. must
endure this, imagine how rare it is to avoid this trauma if you’re poor and Black.
Riverside County and San Bernardino Counties, next door to
L.A., are nearly as bad. So are Wayne
County, Michigan (metropolitan Detroit), San Diego County, California and Middlesex County, Massachusetts
(suburban Boston). Clark County, Nevada (metropolitan Las Vegas) Cook County, Illinois (metropolitan Chicago) and, especially Maricopa County, Arizona (metropolitan Phoenix) are worse. In none of America’s 20 largest counties does
the percentage of Black children enduring a child abuse investigation fall
Phoenix and Los Angeles are also #1 and #2 respectively
among America’s largest cities in NCCPR’s
calculation on overall rates of tearing apart families and consigning
children to the chaos of foster care over the course of a year.
Even for white
children, in all but two
three of these counties, more than 20% will have to endure
a child abuse investigation.
Yes, an investigation is a big deal
Before going further into the data, I want to pause here to
respond to anyone thinking: Well, an investigation is no big deal, they’re just
social workers knocking on the door to offer help Indeed, some weird comments
made by one of the researchers, discussed below, suggests he may believe that. But just to set the record straight, this
story from The New Yorker gives some idea of what it’s really like:
You will hear a knock on the door, often late at night.
You don’t have to open it, but if you don’t the caseworker outside may come
back with the police. The caseworker will tell you you’re being investigated
for abusing or neglecting your children. She will tell you to wake them up and
tell them to take clothes off so she can check their bodies for bruises and
You must be as calm and deferential as possible. However
disrespectful and invasive she is, whatever awful things she accuses you of,
you must remember that child protection has the power to remove your kids at
any time if it believes them to be in danger. … If you get angry, your anger
may be taken as a sign of mental instability, especially if the caseworker
herself feels threatened.
You may never find out who reported you. If your child
has been hurt, his teacher or doctor may have called the state child-abuse
hotline, not wanting to assume, as she might in a richer neighborhood,
that it was an accident. But it could also have been a neighbor who heard
yelling, or an ex-boyfriend who wants to get back at you, or someone who thinks
you drink too much or simply doesn’t like you. People know that a call to the
hotline is an easy way to blow up your life. [Emphasis added.]
On one of the incredibly rare occasions when this actually
happened to a white, middle-class foster parent, it sparked days
of outraged news stories and demands for action from a state
legislator. The fact that this will
happen to almost every impoverished Black child in the City of Los
Angeles, almost every impoverished Black child in the City of Phoenix, and
more than half of Black children nationwide somehow doesn’t provoke the same
Almost every allegation is false
All that is required for a caseworker to “substantiate” an
allegation is for her to check a box on a form stating her personal conclusion
that it is slightly more likely than not that the “abuse” or “neglect” occurred. There is no trial beforehand, no neutral
arbiter hears all sides. Unsurprisingly,
the only study we know of to second guess these decisions, and it’s a very old
study, found workers two to fix times more likely to wrongly “substantiate” an
allegation than to wrongly declare one “unfounded.”
And yet, even with this incredibly low standard of proof and
propensity to over-substantiate, this study found that, on average, in more than 90% of
cases, the workers found that the allegation was false. Even among Black children, where racial bias
and the confusion of poverty with neglect make it more likely workers will check
the “substantiated” box, on average workers decided at least 80 percent of
reports were false.* (And, of course,
among those that are “substantiated” the vast majority are “neglect” – which often
means the family is poor.)
So even by the workers’ own assessment, nine times out of
ten, children are forced to endure this trauma for nothing. That also means that roughly 90% of the time,
workers are spinning their wheels, chasing down false allegations, making it
less likely they’ll find the very few children in real danger.
1 in 5 Black kids in L.A. endures foster care
Graphic from Frank Edwards, et. al., "Contact with Child Protective Services is pervasive but unequally distributed by race and ethnicity
in large U.S. Counties" (Proceedings of the National Academy of Sciences, July 27, 2021).
But, of course, the harm doesn’t necessarily stop with an
investigation. The new study has
stunning data on the cumulative effect of all that investigation: More than 20%
of Black children in Los Angeles will be forced into foster care at some point
during their childhoods – the figure is nearly as high in Phoenix. Phoenix also is where it’s proportionately
most likely that a Native American child will be forced into foster care, and
second most likely for Hispanic children.
Though family police almost always mean well, that means all
these children endure the same trauma endured by children taken from their
parents at the Mexican border. Again,
from the New Yorker story:
If the caseworker believes your kids are in imminent danger,
she may take them. You may not be allowed to say goodbye. It is terrifying for
them to be taken from their home by a stranger, but this experience has
repercussions far beyond the terror of that night. Your children may hear
accusations against you—you’re using drugs, your apartment is filthy, you fail
to get them to school, you hit them—and even if they don’t believe these things
they will remember. And, after your children see that you are powerless to
protect them, this will permanently change things between you. Whatever happens
later—whether the kids come back the next week, or in six months, or don’t come
back at all—that moment can never be undone.
Just ask someone who went through all this, such
as, say, this 14-year-old what “just an investigation” really means:
I’m scared when I hear a hard knock at the door. I think
they are coming. I was scared to go to school because they will come to the
school and remove me and put me in a foster home. All because if my Mom and Dad
don’t do what they want, never mind they are not abusing us.
I will be so glad when I am 18 and my brother is 18. Then
I know [no one] will ever be able to put us in a foster home again.
That’s best case – it doesn’t even begin to account for the
high rate of abuse in foster care itself.
When it comes to child welfare’s death penalty – terminating
children’s rights to their parents (a more accurate term than termination of
parental rights) Phoenix is again #1 – by a wide margin. Nearly three percent of all children and
nearly six percent of Black children in metropolitan Phoenix will have their
parents taken from them forever.
The family policing bias Olympics
If one thinks of this study as scoring a family policing
bias Olympics, different counties may “win” individual events. But who does “best” all around? Probably Middlesex County, Mass.
In a state known for discrimination
families in child welfare, Middlesex had the worst rate among the counties
studies for inflicting child abuse investigations on such families. It also had the worst rate for Native
Percentage of Middlesex County, Mass. Children likely to be
subjected to family police investigation by race:*
Native American 50%
Middlesex also is the county in the study in which Hispanic
children are proportionately most likely to have to endure foster care.
Of course, this is the state where the head of the family
policing agency and the state’s “child advocate” are doing everything possible
that racial bias is a problem.
The other disparity – geographic disparity
Communities are consistent about overinvestigating and
overremoving Black children, but there is no consistency in how often they do
it. That, in itself, shows how arbitrary
and subjective the whole process is.
The rate at which caseworkers check a box on a form saying
they think it’s slightly more likely than not that “abuse” or “neglect”
occurred, (what the study wrongly describes as “confirmed maltreatment”) is nearly ten times higher in metropolitan Detroit than in metropolitan Seattle. A child is three times more likely to be
placed in foster care -- and 17 times more likely to lose rights to her or his
parents forever – in Phoenix than in New York City.
Of course, it’s theoretically possible that Phoenix is such
a cesspool of depravity compared to New York City that there really is 17 times
the amount of child abuse severe enough to merit termination of parental
rights. But it’s more likely the result
of the culture in a state family policing agency in which caseworkers thought
it would be a great idea to wear T-shirts emblazoned with the words “Professional
These data track with NCCPR’s comparisons of rates of child
removal for states and, as noted above,
for big cities, which find enormous differences, even when factoring in rates
of child poverty.
It’s probably worse in rural America
Bad as these data are, odds are the omnipresence of the
child welfare surveillance state is even worse in much of rural America. That’s suggested by the fact that, even when
adjusting for rates of child poverty, the states that tear apart the most
families tend to be states that don’t have any big cities – states like
Montana, Wyoming and Vermont. Of the 20
states most prone to tear apart families, only two, Massachusetts and Arizona,
have counties big enough to be included in the new study.
Substituting surveillance for foster care is not
In a weird press release announcing the study – but not, it
should be emphasized, in the study itself - one of the researchers, Christopher
Wildeman, suggests that New York City is somehow a success story because, even
though the proportion of children investigated is a little above average
(though not, it should be noted, above average for the places in this study),
and even though New York City caseworkers “substantiate” allegations at a
relatively high rate, they take away proportionately fewer children and
terminate parental rights at one of the lowest rates among the counties studies.
Somehow Wildeman leaps from this to the conclusion that
family policing is “working” because ultimately the families aren’t destroyed
forever. UPDATE: Asked for comment about this view, the lead researcher for the study, Frank Edwards, told NCCPR: "I think that the massive surveillance and separation of Black, Brown and Native kids is at crisis levels and needs to be dramatically reduced rapidly.”
In fact, the New York data tell a very different story from Wildeman's spin. They show only that, thanks largely to
pressure from grassroots family advocates and a network of high-quality family defense
providers, the city family policing agency has been stopped from doing the very
worst things it can do to families.
But, as the Movement for Family Power documents in this report,
the decline in foster care numbers has been replaced by an almost precisely
equal increase in oppressive, needless surveillance of families. (One outcome the study does not measure is the
proportion of families forced into such surveillance in each county).
So no, success is not substituting needless family
surveillance for needless foster care; success is substituting community-based
community-designed concrete help for families for both forms of family
*-These are estimates based on the graphics in the study.
The authors did not supply accompanying tables so each may be off by a
percentage point or two.