Skip to main content

Earnings Requirement Would Undermine Child Tax Credit’s Poverty-Reducing Impact While Doing Virtually Nothing to Boost Parents’ Employment

Legislation approved by the House Ways and Means Committee on September 15, part of the “Build Back Better” economic recovery package, would permanently extend an American Rescue Plan provision enabling children in families with little or no income to receive the full value of the Child Tax Credit for the first time. This is sometimes called making the credit “fully refundable,” and it is the key driver of the historic, more-than-40-percent reduction in child poverty achieved by the Rescue Plan’s expansion, which also raises the amount of the credit from $2,000 per child to $3,000 or $3,600, depending on the age of the child.

Taking away the full credit from children based on their parents’ earnings would needlessly leave [them] in poverty… while doing virtually nothing to boost parental employment.Recent suggestions that Congress should deny the credit to children whose parents don’t have earnings are misguided. Taking away the full credit from children based on their parents’ earnings would needlessly leave in poverty — or push deeper into poverty — the children who need help the most, injuring their long-term health and educational outcomes and reducing their earnings as adults, while doing virtually nothing to boost parental employment.

  • In more than 95 percent of families who benefit from making the credit fully refundable, the parent or other caretaker is working, between jobs, ill or disabled, elderly, or has a child under age 2.
  • Evidence from both the United States and Canada strongly indicates that giving the full credit to all children, including those whose families don’t have earnings in a year, won’t affect adults’ work participation to any large degree. An expert panel convened by the National Academy of Sciences projected that under a child credit policy similar to the expanded Child Tax Credit, 99.5 percent of working parents would continue to work, and few would substantially reduce their hours. And the reduction in child poverty from full refundability will likely have important long-term positive impacts on children, including on their educational outcomes and future health and earnings.
  • Adding an earnings requirement would exclude children in the most vulnerable families, where parents are least able to meet basic needs, exposing these children to serious hardship.

Denying the full Child Tax Credit to children whose families don’t have earnings would mark a step backward toward an approach that has failed the nation’s children for decades and undermined its economic strength. From the credit’s creation in 1997 until this year’s enactment of the Rescue Plan, federal rules denied or sharply reduced the Child Tax Credit based on parents’ work status and earnings. The federal government provided very little other income assistance to families with no or very low earnings. Taken together, these policies left this nation with a far higher child poverty rate than similarly wealthy countries that offer broad-based child allowances, while failing to raise U.S. labor force participation above most of those same countries.

Excluding the lowest-income children from the full Child Tax Credit would also undermine efforts to promote racial equity. Prior to the Rescue Plan expansion, about half of Black and Latino children received a partial credit or no credit at all because their incomes were too low, compared to about a fifth of white children. About half of children in rural (i.e., non-metropolitan) communities were also excluded from the full credit due to their family’s low income. These facts reflect sharp disparities in the educational and employment opportunities available to Black, Latino, and rural communities, driven in part by systemic racism. The nation’s high child poverty rate and gaping racial, ethnic, and geographic disparities in child poverty have translated into lost opportunities for millions of children and lost contributions to the nation as a whole.

More than 400 economists recently signed a letter endorsing permanent extension of the Rescue Plan’s Child Tax Credit expansion, including full refundability. The expanded credit “can dramatically improve the lives of millions of children growing up in the United States and promote our country’s long-term economic prosperity,” they explained, noting that it would have only a minimal impact on parents’ employment.[1] In addition, one study estimated that a credit similar to the expanded Child Tax Credit has a return to society of $8 for every $1 spent.[2]

Taking Credit Away From Children With Greatest Need Would Cause Them, and the Nation, Serious Harm

  • Full refundability drives the Rescue Plan’s large poverty-reducing impact. The Rescue Plan’s Child Tax Credit expansions are expected to cut child poverty by over 40 percent, lifting more than 4 million children out of poverty. More than 80 percent of this effect comes from the provision making the full credit available to families even if they have low or no income.[3] (See Figure 1.) Denying the full credit to children whose parents are out of work or have low earnings would leave millions of children in poverty.
  • Full refundability is critical to reducing racial disparities. Prior to the Rescue Plan expansion, 27 million children received a partial credit or no credit at all because their families’ incomes were too low.[4] This included about half of all Black and Latino children, compared to about 20 percent of white children, and about half of all children in rural communities.
  • Growing up in poverty causes lasting harm. Poverty and the hardships that it brings — unstable housing, frequent moves, inadequate nutrition, and high family stress — often take a heavy toll on children, leading to lower levels of educational attainment, lower earnings, higher likelihood of being arrested, and poorer health in adulthood.[5] The nation as a whole also loses out on what these children would have contributed as adults had poverty and hardship not shortchanged their futures.
  • Boosting family income helps children earn more as adults. Research has found that when parents with low incomes receive income supports, their children perform better academically,[6] which translates into higher earnings in adulthood.[7] Research around the initial rollout of SNAP (then called food stamps) in the 1960s and 1970s found that as the program gradually expanded nationwide, disadvantaged children born in counties with access to food stamps earned more and had greater self-sufficiency years later; they also were healthier at birth, were likelier to finish high school, and lived longer than peers who did not have access to food stamps.[8] Denying the Child Tax Credit to children whose parents (or other relatives) are out of work could therefore lead to lower earnings for the next generation by excluding some children from needed income support.

Vast Majority of Families Benefiting From Full Refundability Include Worker

  • More than 8 in 10 families who benefit from full refundability have a current or recent worker. In 72 percent of families who have earnings low enough to benefit from full refundability, the family head or spouse worked during the year.[9] And 82 percent of families benefitting from full refundability worked either in the current year or in the prior year or next year. (These data are for 2014-2016, the latest available source of this information.)
  • In more than 95 percent of families who benefit from full refundability, the family head or spouse worked in the current, previous, or subsequent year, is age 65 or older, has a work-limiting health condition or disability, or the family has a child under age 2. Additional parents have other compelling reasons for not working, such as attending school.

Fully Available Credit Would Not Stop Parents From Working

  • Giving families with low or no earnings full access to the Child Tax Credit is unlikely to meaningfully alter parents’ employment rate. As the economists’ letter supporting full refundability explains, “recent empirical studies suggest that the income provided through the [Child Tax Credit] program is unlikely to meaningfully reduce parental labor supply.”[10] As noted, an expert panel convened by the National Academy of Sciences projected that under a similar child credit policy, 99.5 percent of working parents would continue to work, and few would substantially reduce their hours.[11] A study of Canada’s implementation of its recent child benefit expansion found no detectable influence on employment for single mothers, the adults most likely to be affected.[12]
  • The structure of the credit doesn’t penalize work by low- or moderate-income parents. In order to constrain cost, programs benefitting low-income individuals often phase out at relatively low income levels, effectively reducing the return to individuals from working and earning more. The proposed Child Tax Credit expansion is different: it doesn’t begin to phase down until a family’s income exceeds $150,000 for married parents and $112,500 for unmarried parents. This means that as individuals with low incomes are able to work more hours or get a raise, they lose none of their tax credit.
  • Experience with unemployment insurance shows that cutting benefits doesn’t necessarily increase work. Governors in several states this summer ended federal unemployment insurance benefits early, believing that cutting aid would lead unemployed workers to get jobs more quickly. But a number of studies have found that employment growth was no faster in those states than in others,[13] and that financial hardship increased more in the states that cut benefits early.[14]

Weak Public Supports Contribute to High U.S. Child Poverty Rate

  • The U.S. child poverty rate exceeds that of other wealthy countries with more family supports, including supports that are not tied to parents’ work status. The U.S. poverty rate among children is higher than other similarly wealthy nations largely due to the weakness of public supports in this country.[15] Before counting income from government programs, many other countries’ child poverty rates are similar to or even higher than the U.S. rate. But these countries have markedly lower child poverty rates once benefits from government programs are included. (See Figure 2.)
  • Most countries in the Organisation for Economic Co-operation and Development with a child benefit have higher workforce participation than the United States. For example, France, Canada, the United Kingdom, and Germany all had higher labor force participation rates than the United States before the pandemic (see Figure 3), even though all have long had a child allowance. And all lift higher shares of their children above the poverty line than the United States does.

Low-Income Families Are Using Credit to Meet Basic Needs

  • Families use the credit for food, rent, clothing, and other essentials. The most common way people used their initial monthly Child Tax Credit payments this summer was to buy food.[16] Families with low incomes were particularly likely to spend the payments on necessities (as well as education, such as children’s school supplies). This is consistent with evidence from Canada, where parents spend their child allowances on essentials like food, rent, transportation, and education expenses.[17]
  • Food hardship fell after the monthly Child Tax Credit payments began. In the month and a half after the initial monthly Child Tax Credit payment was issued, the number of adults with children reporting that the household didn’t have enough to eat dropped by 3.3 million or nearly one-third.[18] The extension of the full credit to families with little or no earnings is likely an important reason for this improvement.

Credit Available to All Poor Children Has Lasting Societal Benefits

  • Research finds an 8 to 1 cost-benefit ratio from a fully refundable credit. Irwin Garfinkel and colleagues at Columbia University estimate that the societal benefits of a “child allowance” that is substantially similar to the expanded Child Tax Credit outweigh its fiscal costs by a ratio of 8 to 1, due primarily to the allowance’s economic and health benefits to future generations.[19] In the long run, the government would also recoup a portion of the upfront budgetary cost of the investment in kids because young people lifted out of poverty grow up to be healthier, more productive workers with higher earnings.
  • Temporary measures enacted in response to the pandemic reduced poverty markedly in 2020, showing what such policies can achieve. Despite the upheaval caused by the pandemic, the annual poverty rate fell dramatically — by 2.6 percentage points — in 2020, using the Census Bureau’s Supplemental Poverty Measure, which includes a broad range of government benefits. At 9.1 percent, that measure was the lowest on record in data going back to 2009.[20] While the response to the pandemic was far from perfect and assistance often took weeks or months to arrive, these data show that government policies can produce dramatic improvements in poverty.

End Notes

[1] Letter from Hilary Hoynes, Diane Schanzenbach, and other economists to Senate Majority Leader Chuck Schumer et al., September 15, 2021,

[2] Irwin Garfinkel et al., “The Costs and Benefits of a Child Allowance,” Center on Poverty and Social Policy at Columbia University, Poverty and Social Policy Brief, Vol. 5, No. 1, February 18, 2021,

[3] Chuck Marr, Kris Cox, and Arloc Sherman, “Recovery Package Should Permanently Include Families With Low Incomes in Full Child Tax Credit,” CBPP, September 7, 2021,

[4] Ibid.

[5] National Academies of Sciences, Engineering, and Medicine, A Roadmap to Reducing Child Poverty, National Academies Press, 2019,

[6] Greg J. Duncan, Pamela A. Morris, and Chris Rodrigues, “Does money really matter? Estimating impacts of family income on young children’s achievement with data from random-assignment experiments,” Developmental Psychology, Vol. 47, No. 5, 2011,

[7] Raj Chetty et al., “How Does Your Kindergarten Classroom Affect Your Earnings? Evidence from Project Star,” Quarterly Journal of Economics, Vol. 126, No. 4, November 2011,

[8] Martha J. Bailey et al., “Is the Social Safety Net a Long-Term Investment? Large-Scale Evidence from the Food Stamps Program,” National Bureau of Economic Research Working Paper 26942, April 2020,; Hilary Hoynes, Diane Whitmore Schanzenbach, and Douglas Almond, “Long-Run Impacts of Childhood Access to the Safety Net,” American Economic Review, Vol. 106, No. 4, April 2016, pp. 903-934. An earlier version of the paper is at

[9] Figures in this and the following bullet are from CBPP analysis of U.S. Census Bureau Survey of Income and Program Participation data for 2014-2016. Families in this analysis are Census families (including families of the household head and unrelated sub-families) with related children under 18 and with earnings low enough to benefit from full refundability; that is, 15 percent of the annual earnings (in 2021 dollars) of the family head and spouse are below $3,600 per child under age 6 plus $3,000 per older child. Elderly families are those where the family head and spouse (if any) are age 65 or older. Family health/disability means the family head or spouse had a work-limiting disability or reported not working due to a chronic illness/disability in December 2015.

[10] Letter to Senate Majority Leader Chuck Schumer et al., op. cit.

[11] National Academies of Sciences, Engineering, and Medicine, op. cit. The report estimates that 25.49 million people in families below 200 percent of the Supplemental Poverty Measure would be employed after enactment of a $3,000-per-child allowance, which is a similar proposal to the Rescue Plan’s expanded Child Tax Credit. That compares to 25.61 million people employed in such families prior to the enactment of a $3,000-per-child allowance. The report estimates that 4.55 million people in families below 200 percent of the Supplemental Poverty Measure would opt to work slightly fewer hours — an average of about one hour per week. See Danilo Trisi and Ife Floyd, “Benefits of Expanding Child Tax Credit Outweigh Small Employment Effects,” CBPP, March 1, 2021,

[12] Michael Baker, Derek Messacar, and Mark Stabile, “The Effects of Child Tax Benefits on Poverty and Labor Supply: Evidence from the Canada Child Benefit and Universal Child Care Benefit,” National Bureau of Economic Research Working Paper 28556, March 2021, Note that Canada’s child benefit phases out at much lower income levels than the Child Tax Credit.

[13] Peter Ganong, Twitter, August 20, 2021,; Kyle Coombs et al., “Early Withdrawal of Pandemic Unemployment Insurance: Effects on Earnings, Employment and Consumption,” August 20, 2021,; Jed Kolko, Twitter, August 17, 2021,

[14] Arindrajit Dube, “Early impacts of the expiration of pandemic unemployment insurance programs,” July 18, 2021,; G. Elliott Morris, “So far, ending pandemic unemployment aid has not yielded extra jobs,” The Economist, August 20, 2021,

[15] Janet C. Gornick and Timothy M. Smeeding, “Redistributional Policy in Rich Countries: Institutions and Impacts in Nonelderly Households,” Annual Review of Sociology, Vol. 44, July 2018, pp. 441-468,

[16] Claire Zippel, “After Child Tax Credit Payments Begin, Many More Families Have Enough to Eat,” CBPP, August 30, 2021,

[17] Lauren E. Jones, Kevin Milligan, and Mark Stabile, “Child cash benefits and family expenditures: Evidence form the National Child Benefit,” Canadian Journal of Economics, Vol. 52, No. 4, November 2019,

[18] Zippel, op. cit.

[19] Garfinkel et al., op. cit.

[20] Liana Fox and Kalee Burns, “The Supplemental Poverty Measure: 2020,” U.S. Census Bureau, September 14, 2021,