In late 2015, Congress agreed as part of a bipartisan funding deal to produce a landmark, $750,000 report on how to cut child poverty in America.
The result of pressure from California Democratic Representatives Barbara Lee and Lucille Roybal-Allard, the provision called for the National Academy of Sciences to convene a group of experts to produce “a nonpartisan, evidence-based report that would provide its assessment of the most effective means for reducing child poverty by half in the next 10 years.”
That group has now, more than three years later, produced its exhaustive, 600-page report.
The report estimates that child poverty costs us $800 billion to $1.1 trillion every year due to increased crime, worsened health, and lower earnings when poor kids become adults. There is no one approach to reducing it, the committee concludes, but it did outline four separate options policymakers could pursue — two of which would cut child poverty in half in the next decade.
- A “work-based package,” which increases the Earned Income Tax Credit (EITC), makes the Child Care Tax Credit fully refundable, boosts the minimum wage, and scales up WorkAdvance, a training program “in which program staff work closely with employers to place disadvantaged individuals with moderate job skills into training programs for specific sectors that have a strong demand for local workers.” This plan does the least to reduce child poverty out of the four options, cutting it from 12.6 percent to 10.2 percent, and lifting 1.8 million kids out of poverty. It would cost about $9.3 billion per year.
- A “work-based and universal support package,” which includes the EITC and Child Care Tax Credit policies from the first package but pairs them with a $2,000 per year child allowance paid to all parents of children under 17. Adding a child benefit makes this plan much better at reducing poverty, cutting the rate from 12.6 percent to 8.3 percent and lifting 3.2 million kids out of poverty. That’s a reduction in child poverty of about a third: much better than the work-focused plan, but not enough to meet the goal of halving child poverty. It would cost about $44.3 billion per year.
- A “means-tested supports and work package,” which includes the EITC and Child Care Tax Credit boosts from the first two options, but adds a 35 percent increase in Supplemental Nutrition Assistance Program (SNAP, or food stamps) benefits, and dramatically increases access to Section 8 housing vouchers. This plan cuts child poverty by half: The rate goes from 12.6 percent to 6.2 percent, lifting 4.7 million kids out of poverty. The housing and food benefits do substantially more to reduce policy than the “work-based” benefits. It would cost about $90.7 billion per year.
- A “universal supports and work poverty reduction package,” which includes a bigger increase in the EITC than the first three packages, includes a minimum wage increase to $10.25 per hour and makes the Child Tax Credit refundable, offers various anti-poverty programs to legal immigrants who are currently barred, and, most importantly, includes a child allowance of $2,700 per year, as well as a $1,200 per year publicly funded minimum child support payment for single parents entitled to child support from their former partner (oftentimes, these partners aren’t able to regularly pay child support). This plan also cuts child poverty by half: the rate falls from 12.6 percent to 6.1 percent, lifting 4.8 million kids out of poverty. It would cost about $111.6 billion per year.
The authors expressly rule out and do not include several policies popular in DC think tanks as insufficiently evidence-based to include in the report: work requirements like those in the 1996 welfare reforms, marriage promotion, and subsidizing long-acting reversible contraceptives (LARCs) in the hope of preventing women from having children they don’t have the resources to support.
The costs of each of these plans, even the most expensive, are dwarfed by the annual social costs of child poverty as estimated by the committee.
The committee that drafted the study comprises a who’s who of child poverty researchers, including economists Greg Duncan (who chaired the group), Janet Currie, Irv Garfinkel, Hilary Hoynes, Robert Moffitt, and Timothy Smeeding; psychologists Vonnie McLoyd and Eldar Shafir (whose popular book Scarcity documented the mental consequences of poverty), and, perhaps most surprisingly, Ron Haskins, an architect of the 1996 welfare reforms during his time as a House Republican aide.
The authors speak for no one but themselves, and these conclusions read as what they are: the outputs of a committee with both welfare reform skeptics like Smeeding and supporters like Haskins, whose members had differing beliefs on the importance of work, the justifiability of giving money or benefits to people who cannot or do not work, and the proper size of the federal budget, among several priorities.
But I think the report still leaves us with some clear lessons for where child poverty policy should go in the future.
Promoting work doesn’t get you much of anywhere
One thing the report suggests is that we’re hitting the limits of how much we can reduce poverty through work alone.
In the 1990s, US social welfare policy took a dramatic shift away from guaranteed income support for poor households in the form of the Aid to Families with Dependent Children (AFDC) program and toward an approach to poverty reduction with a heavy focus on work. Temporary Assistance to Needy Families (TANF), which replaced AFDC after the 1996 welfare reforms, imposed work requirements and prioritized job placement over offering cash; the Earned Income Tax Credit, which the Clinton administration dramatically expanded in the 1993 budget, funneled money toward the working poor, not the disabled, elderly, caretaker, or otherwise non-working poor.
The authors are clearest about the limits of this shift in respect to TANF-style work requirements. “There is insufficient evidence to identify mandatory work policies that would reliably reduce child poverty, and it appears that work requirements are at least as likely to increase as to decrease poverty,” they conclude.
But not even EITC expansion and other work-based subsidies were sufficient, the experts concluded. “The committee was unable to formulate an evidence-based employment-oriented package that would come close to meeting its mandate of reducing child poverty by 50 percent,” they write.
Moralizing doesn’t get you much of anywhere, either
Conservative approaches to poverty reduction have traditionally shied away from increased government spending in favor of increased government hassling of welfare recipients, urging them to work more but also engage in better family planning and marry before having children.
The experts reject basically all of these as non-evidence-based. “In the early 2000s, an ambitious attempt to develop programs that would improve couple relationship skills, promote marriage, and improve child well-being failed to boost marriage rates and achieve most of their other longer-run goals,” they conclude.
They do evince some sympathy for long-acting reversible contraceptives (LARCs), like IUDs and implants, noting evidence from Colorado and Delaware that increased access to this kind of birth control can reduce unwanted pregnancies substantially. But this is a bit conceptually odd, if not unsettling, as an answer to “child poverty”; it doesn’t make any actual children better off, it just prevents them from existing.
Moreover, the experts note that there’s evidence that government family-planning efforts exhibit racial bias. It’s easy to imagine a LARC-promotion program with reduced child poverty as its goal disproportionately trying to prevent black women from having children in a way that undermines rather than promotes reproductive autonomy.
The authors ultimately decline to recommend LARC promotion as a solution for child poverty.
You need to give people money
A few years ago, another group of eminent poverty researchers — including Duncan, the NAS committee chair, and fellow NAS committee members Smeeding and Garfinkel, as well as fellow poverty experts Jane Waldfogel, Kathryn Edin, Luke Shaefer, David Harris, Sophie Collyer, Christopher Wimer, and Hirokazu Yoshikawa — offered a stirring call for a large child allowance, of $3,000 to $3,600 per year, paid out monthly and perhaps in greater amounts for young children, in a paper for the Russell Sage Foundation.
“To be sure, the costs involved would be substantial,” they wrote. “But this investment would lead to large and direct reductions in child poverty, and might also have a significant effect on the poverty and well-being of future generations.”
The NAS report, as a consensus document, does not come to that clear of a conclusion. But the same takeaway — that we need increased unconditional cash or cash-like support for families with kids — is implicit in every aspect of the study.
Take, for instance, the below table, which breaks down the effect on poverty and deep poverty of each policy included in the report’s four policy packages:
:format(webp):no_upscale()/cdn.vox-cdn.com/uploads/chorus_asset/file/14773063/child_poverty_effects.png)
The single policy that does more than any other to reduce poverty is a $2,700 child allowance, which would single-handedly cut child poverty by a third. In second place? A $2,000 child allowance.
The next two after that are expanded housing vouchers — which are a quite cash-like benefit, replacing an expenditure that poor families would otherwise have to pay out of cash income — and expanded food stamps, which are basically a cash benefit that the government semi-arbitrarily bars recipients from spending on stuff other than groceries.
The bigger the program is, the more it does to reduce poverty. The more cash-like the program is, the more it does to reduce poverty. It’s not hard to see why, given these realities, so many poverty experts have coalesced around an even larger $3,000 to $3,600 per year child allowance.
And encouragingly for fans like myself of such a policy, that idea is being rapidly adopted within the Democratic Party. Since 2017, Sens. Michael Bennet (D-CO) and Sherrod Brown (D-OH) have been pushing the American Family Act, a bill that would create a $3,000 child allowance for older kids and a $3,600 allowance for younger kids, directly modeled after the poverty experts’ plan. Almost every major Democratic senator running or considering running for president in 2020 has cosponsored the bill, including Bennet and Brown themselves but also Kamala Harris, Cory Booker, Kirsten Gillibrand, Amy Klobuchar, and Elizabeth Warren. Bernie Sanders is the sole holdout.
Democrats could easily pass the American Family Act or a bill similar to it through budget reconciliation, so long as it’s either paid for or expires after 10 years. And as Canada’s recent experience suggests, the results would be profound. After Justin Trudeau introduced a vastly expanded child benefit there, the share of children living in families below Canada’s low-income threshold fell by about a third. There’s no reason US children can’t receive the same kind of help.
Sign up for the Future Perfect newsletter. Twice a week, you’ll get a roundup of ideas and solutions for tackling our biggest challenges: improving public health, decreasing human and animal suffering, easing catastrophic risks, and — to put it simply — getting better at doing good.