Can the U.S. Cut Child Poverty in Half in a Decade?

The Issue:

Child poverty is a serious problem for the United States. It compromises the health, learning and development of our children and their future employment opportunities and well-being. Can we find a feasible and cost-effective set of programs and policies that would generate substantial reductions in child poverty? Our recent experience serving on a National Academy of Sciences consensus panel, which was devoted to identifying ways to reduce child poverty by half within 10 years, suggests that we can indeed make a difference if we make reducing child poverty a priority.

Over 9 million children lived below the poverty line in 2015. Evidence-based solutions could make a substantial difference in the rate of child poverty.

The Facts:

• An estimated 13 percent of U.S. children — 9.6 million in all — lived in families with incomes below the poverty line in 2015. These figures are based on the Supplemental Poverty Measure, which amounts to an income of $26,000 for a two-adult, two-child family renting its housing. Income includes benefits from noncash programs like tax credits and housing vouchers. Nearly 3 percent of U.S. children are living in deep poverty — defined as below 50 percent of the poverty line. Poor children have weaker language and memory skills than their peers, and these problems continue into adulthood. When they grow up, poor children have lower earnings and income, are more dependent on public assistance, have more health problems, and are more likely to commit crimes. Robust causal evidence has shown that low income itself, rather than other conditions poor children face, is responsible for many of these negative impacts of poverty on child development.
• History shows that wealthy nations have been successful in lifting children out of poverty. The United States reduced child poverty by nearly half over the 36-year period from 1970 to 2016. The United Kingdom cut its child poverty rate by half in under a decade (between 2001 and 2008) with policies that made work pay, focused welfare reforms on employment but also raised the incomes of families with children regardless of the parents’ work status. Canada appears to be on course to achieve this goal even more quickly, after introducing its new child benefit in 2016.
• If it were not for existing anti-poverty programs and policies the current U.S. child poverty rate would be much higher. Examining the anti-poverty effectiveness of current government programs, we found that a number of them, in particular refundable tax credits — the Earned Income Tax Credit (EITC) and the Child Tax Credit (CTC) — and the Supplemental Nutrition Assistance Program (SNAP) have been effective in reducing child poverty. For instance, we estimate that without the EITC and the CTC the child poverty rate would have been 18.9 percent in 2015, almost 6 percentage points higher than it was (see chart below). Moreover, there is research evidence that directly links the refundable tax credits and SNAP with improving child health and achievement.

 

 

• The National Academies of Sciences, Engineering and Medicine tasked our committee with assessing what it would take to reduce child poverty in half in a decade. After soliciting ideas from a diverse set of policy analysts, the committee developed 20 ideas for enhancing anti-poverty programs and policies, none of which alone was estimated to meet the goal of reducing child poverty in the United States by 50 percent over a 10-year period. The committee then constructed four packages of policy and program ideas. Its choices were guided by the magnitude of potential reductions in poverty and deep poverty (defined by incomes less than half the poverty line), net cost, and by how they would impact people's participation in paid work. The committee used the best available evidence to simulate the poverty reduction and changes in employment and earnings generated by the programs and program packages.
• Two packages were estimated to achieve the 50 percent reduction goal, and both of these combined incentives for work with benefits to stabilize incomes. The means tested supports and work package would expand the Earned Income Tax Credit (EITC), concentrate the benefits of the Child and Dependent Care Tax Credit (CDCTC) on low-income families, and increase the generosity of two major income support programs — housing vouchers and SNAP. The universal supports and work package would include the same tax credits, a child allowance of $2700 per year, create a child support assurance program that would guarantee the payment of legally mandated child support, raise the minimum wage to $10.25 over three years, and eliminate some restrictions on access to benefit programs for legal immigrants. The child allowance component would replace the current Child Tax Credit, extend it down the income ladder to include all low-income families with children, and pay it out monthly. The EITC would continue to be paid once a year.
• At $90 billion to $110 billion per year, the cost of these two packages would be substantial, although much lower than the aggregate costs that child poverty imposes on the nation. Because these two packages combined policies that increased work and pay among low-income parents with policies that strengthened the safety net, they were estimated to cut child poverty in half while also increasing employment and earnings among low-income adults (see top chart). Annual employment increases achieved by these two packages were estimated to total 400,000 and 600,000, respectively. In spite of the increased participation in the workforce, the two packages were estimated to cost $90 billion to $110 billion per year. These budgetary costs should be placed along side of the cost associated with the continuation of child poverty. Based on worker productivity losses, costs to the healthcare system, and the costs of incarceration and the judicial system, child poverty has been estimated to cost the economy of the United States between $800 billion and $1.1 trillion each year.  
• Poverty reduction that prioritizes work incentives reduces the cost of combatting poverty but is not sufficient by itself to meet the goal of cutting existing child poverty in half. The committee’s work-oriented package was its attempt to meet the poverty reduction goal using employment-based policy and program ideas. It included the EITC and CDCTC expansions mentioned above, increased the national minimum wage to $10.25, and also scaled up a promising new sectoral training program – called Work Advance – which provides training for all heads of households (with children) under 200 percent of poverty. This package was estimated to increase employment by over 1 million workers and cost only $9 billion per year, but it reduced child poverty by less than 20 percent. The lesson we drew was that work enhancements alone, even when combined with current income support programs, are not enough to cut child poverty in half by themselves . Finally, we constructed a work-based and universal supports package that included the EITC and CDCTC expansions plus a $2,000 per year child allowance, as this is the same level as the current Child Tax Credit. This package is estimated to reduce child poverty by a third, rather than a half, while at the same time increasing employment among low-income adults by 570,000. And at $44 billion, it costs less than half as much as each of the two larger packages.
• Other factors beyond the programs considered can impact poverty reduction. A number of important contextual factors can greatly influence the impact and success of anti-poverty programs and policies. These factors include income stability and predictability; equitable access to programs; simple versus cumbersome enrollment procedures; bias and discrimination; neighborhood conditions; physical and mental health. In addition, while mandated employment programs and stable two-parent households have the potential to lower poverty rates, upon considering the research literature we determined that we lack rigorous evidence to suggest that mandatory work and marriage promotion proposals were sufficiently promising to include them in our simulations.