It is hard to read an article about the child-welfare system without learning about the “foster-care-to-prison pipeline.” From Teen Vogue to NBC News, the idea that time spent in foster care contributes to higher levels of involvement in the criminal-justice system is practically an article of faith. Even right-leaning groups such as the Parental Rights Foundation support the idea that foster care itself is a problem that results in (and is not simply correlated with) worse outcomes for kids. The notion is often cited as a reason to reduce — if not eliminate — its use.

But what if time in foster care actually helps kids? A recent paper from the National Bureau of Economic Research (NBER) suggests that this is indeed the case. Research by Jason Baron, of Duke University, and Max Gross, of Mathematica, finds that time in foster care reduces the likelihood of adult criminal involvement by 25 percentage points and reduces the likelihood of conviction by 21 percentage points. These declines were driven by a significant reduction in violent crime in particular. And the effects are larger than those of many other evidence-based interventions, including the cognitive-behavioral therapy employed in the much-celebrated Becoming a Man counseling program.

The NBER research was conducted by studying 120,000 child-welfare investigations involving kids in Michigan between the ages of six and 16, from 2008 to 2016. The researchers were able to approximate a random sample by using the random rotation of child-welfare investigators. Some had a history of being more lenient and leaving kids on the margins in their homes, while others were more likely to remove such children and place them in foster care: By looking at the cases that could go either way and then comparing the kids who were left in their homes (with support services) and those who were removed, the authors could understand the effects of time in foster care.

What is going on here? Why is it that kids in foster care end up doing better than those who have been abused or neglected but left in their homes? For one thing, the authors found a lower likelihood that those in foster care would suffer maltreatment again by the families that originally abused or neglected them, even years after they left foster care and after they were reunited with their families, as most of them are. If maltreatment leads to more trauma and more behavioral problems, then less maltreatment might lead to less criminal activity.

Similarly, the authors found that there were better academic outcomes among kids who had spent time in foster care. “Foster care substantially reduces absences from school, improves math test scores, and appears to increase the likelihood of high school graduation and college enrollment.” Improved academic (and therefore job) prospects can also have the effect of decreasing criminal activity. And this study was not the only one to find such a result.

An earlier study from the University of Michigan’s Youth Policy Lab, authored by Max Gross, also found that removing children from homes to foster care had positive long-term benefits. Looking at “borderline” cases similar to the ones in the NBER study, in which some Michigan caseworkers might have chosen to leave a child in the home, with support services, and others would have removed the child, the authors concluded that foster placement reduced the likelihood of a child’s being reported for maltreatment again.

It is true that a 2008 study of Illinois foster care published by MIT professor Joseph Doyle and constructed in much the same way found that time in foster care had negative impacts on children. But the foster-care system has changed significantly since then. Not only has the number of kids in foster care shrunk significantly, but reforms have also improved the quality of placement settings and reduced the length of placements. More important, perhaps, Illinois’s child-welfare system is an outlier, with foster care there lasting nearly two years longer than the national average. Michigan’s system, by contrast, is representative of the rest of the country’s.

Other research since then has confirmed clear benefits of foster care. A study in the journal Pediatrics of infants born in California between 2010 and 2016, for instance, found that the risk of death for those placed in foster care was only half of what it was for other infants reported for maltreatment. Foster care can clearly protect children from harm. This doesn’t mean, of course, that every report of maltreatment should result in a child’s being placed in foster care. Rather, “during periods of foster care placement, infants may receive a level of case management and care coordination that protects against death and other health and developmental adversities,” the authors note. “Differences in death rates signal potential unmet service needs among infants who remain at home after reports.”

Better-coordinated services might help, but it’s also true that incentives matter. “We find evidence that birth parents make improvements while their children are temporarily in foster care,” the NBER paper notes. What kind of improvements? “After removal, birth parents work closely with social workers and receive fully funded services” — such as substance-abuse treatment or counseling — “to address challenges in their lives.” The paper’s authors argue as well that “parents might also benefit from time and space away from their children to focus on themselves.” (In-patient rehabilitation may be more effective, for instance.) Those parents whose children are not in foster care also receive services, but “the incentives to make changes are stronger for parents working to regain custody of their children than for parents who face the threat of child removal.”

This is the key. Most of the child maltreatment in this country is driven by our drug crisis. Some estimates suggest that substance abuse is involved in over 80 percent of child-maltreatment cases. Drug abuse (often in combination with mental illness) prevents parents, particularly those of small children, from properly caring for kids. This includes the constant monitoring needed by infants and toddlers — feeding them, changing them, and getting them medical care; keeping them from touching hot stoves or running into traffic or ingesting substances they should not. It means not leaving them in the care of other dangerous individuals.

The removal of children in order to incentivize a change in parental behavior can seem heartless. Writing a few years ago for the Institute for Family Studies, Amber Lapp interviewed a woman whose children were removed because of her addiction, after which she went to a rehab program. “Well, I think a lot of people like me, they stay on [drugs] because they lose their kids,” she told Lapp, who concluded: “We need a system that is by default integrative, acknowledging and nurturing the mother-child relationship and recognizing that, for many mothers, keeping the connection strong with their children may be the most powerful motivation to get better.”

Unfortunately, that doesn’t seem to be the case. Getting off drugs is hard. And, as we have seen in the massive move toward drug legalization and the widespread cultural acceptance of drug use, there are fewer and fewer reasons to do so. In a recent essay in the Wall Street Journal, my AEI colleague Sally Satel and Kevin Sabet, CEO of the Foundation for Drug Policy Solutions, argue that involuntary drug treatment in which participants are threatened with jail time may be more effective than voluntary treatment:

Data from the federal Substance Abuse and Mental Health Services Administration show that in 2022, a staggering 94.8% of people with a drug or alcohol use disorder within the past year “did not seek treatment and did not think they should get treatment.” Those who do voluntarily enter treatment usually don’t complete it. About one-third of voluntary patients drop out of treatment before completion, according to government data. Other studies show that up to 80% leave by the end of the first year.

Satel and Sabet note that addiction is “not a classic, involuntary illness; it is a behavior that entails choice and responds to consequences.” Consequences such as losing custody of your children.

No one, least of all those of us who distrust big government, wants to believe that foster care is the best solution for kids. But the truth is that for at least some adults, it will give the necessary push to get help and become the parents their kids actually need.