The Case for a Universal Child Tax Credit

Abstract

The United States has one of the highest child poverty rates of any developed nation. According to the US Census (2025), about 14.3% of children were poor in 2024—nearly 50% higher than the poverty rate among working-age adults and those 65 and older. Child poverty is associated with a host of negative outcomes for both the individual—including lower educational attainment (e.g., Duncan et al. 2011), poorer health, and lower economic mobility (e.g., Chetty et al. 2017)—and society, including higher crime and greater dependence on government programs (Hartley et al. 2022). There have been considerable debates over the years about the best way to reduce child poverty. Expanding the Child Tax Credit (CTC), a tax credit for families with children, has been discussed as one such solution (Shaefer et al. 2018). For instance, a 2019 National Academy of Sciences, Engineering, and Medicine (NASEM) consensus report included an expansion of the CTC into a near-universal child allowance of approximately $3000 per child as a key component of any policy package that could reduce child poverty by 50% (NASEM 2019). One critique of expanding the CTC into a near-universal child allowance is that the program would be very expensive and the vast majority of spending would go to higher income families, not children living in poverty. In fact, a follow-up NASEM report found that, under 2022 policy, just 11% of CTC dollars go to children living in poverty (NASEM 2026). This means that nearly 90% of federal spending on the CTC goes to families living above the poverty line. Rather than expanding a program where nearly all beneficiaries do not live in poverty, why not instead provide a more targeted benefit specifically to families in need? Whether to target benefits or provide universal benefits has been a longstanding debate in social policy. Targeting benefits reduces fiscal costs but also risks isolating recipients and increasing the stigma associated with participating in government programs. More targeted programs also tend to have less public support than programs that are universal. This puts targeted programs at risk, as evidenced by the recent cuts to food stamps and Medicaid—both of which provide benefits to those living in or near poverty—as part of the 2025 Big Beautiful Bill Act (BBBA). Universal programs, in contrast, tend to enjoy more public and political support and carry lower risks of fiscal cuts. One argument against providing universal benefits is that it is inefficient to provide benefits to individuals who may not need them. While there is little evidence that providing benefits to higher income families improves the outcomes of their children, there is a longstanding literature highlighting the importance of increasing economic resources to children growing up in poverty, particularly in the early years of life (e.g., Page 2024). A host of empirical studies have found causal evidence that increasing resources through exposure to the social safety net leads to improvements in education (e.g., Dahl and Lochner 2012), health (Goodman-Bacon 2021; Braga et al. 2020), and economic mobility (Bailey et al. 2021; McInnis et al. 2023). These improvements benefit not only the individuals themselves, but also society more broadly through lower dependence on government programs, lower crime rates, and higher tax revenue (Bastian and Jones 2021). The benefits produced by such interventions may therefore far outweigh the costs of providing benefits to children residing in higher income households (Ananat and Garfinkel 2023). In this Point-Counterpoint, I lay out an argument for delivering the CTC to a broader share of the population of families with children, highlighting the benefits of a universal approach over a more targeted approach that provides benefits only to lower income families. Since the current structure of the CTC excludes the lowest income families from receiving any benefit, I focus my argument on the case for expanding eligibility to families with little to no earnings while maintaining benefits for higher income families. I also provide suggestions for how to improve the progressivity, reduce the fiscal costs, and lower child poverty through changes to the benefit structure. Over the last several decades, the US social safety net has increasingly shifted away from unconditional cash benefits and toward work-contingent refundable tax credits such as the CTC. As of 2025, the credit was worth $2200 per child under the age of 17. Estimates suggest that approximately 90% of households with children claim at least some benefit from the CTC, at a fiscal cost of approximately $100 billion (Tax Policy Center 2025). Along with the Earned Income Tax Credit (EITC), the CTC lifts more than 3.7 million children out of poverty each year, the largest reduction in child poverty of any government program (US Census 2025).1 Despite its substantial impact on child poverty, the CTC was not originally intended to be an antipoverty program. Indeed, the original benefit structure included a $10,000 minimum earnings requirement before tax filing units could claim the first dollar of benefits. This threshold was imposed deliberately to coincide with the start of the phase-out of the EITC, which many low-income working families claim. The CTC was also not originally refundable, which meant that households with no tax liability could not receive any benefit from the CTC, excluding most families living in poverty. In the decades since the credit was implemented, there have been several expansions that have increased the size of the benefit, reduced the minimum earnings requirement, and made the credit partially refundable, all of which led to more low-income families gaining eligibility. The CTC is one of few policies that has enjoyed bipartisan support. The credit has been expanded under both Republican and Democratic administrations, with more conservative supporters highlighting its pro-family nature, and more liberal supporters highlighting its antipoverty impacts. Two recent reforms to the credit are worth highlighting to illustrate the broad political support for the CTC. In 2017, as part of the Tax Cuts and Jobs Act (TCJA) during the first Trump administration, the credit doubled from $1000 to $2000 per child and the refundable portion of the credit increased. The minimum earnings threshold modestly reduced (from $3000 to $2500), expanding the credit to slightly lower income families. At the other end of the distribution, the earnings threshold at which the credit began phasing out increased dramatically from $110,000 to $400,000 for married filers and from $75,000 to $200,000 for unmarried filers. This expansion meant that all but the richest 1%–2% of families with children could receive the credit (Goldin and Michelmore 2022). It was this increase in the maximum earnings limit that substantially increased the share of all children receiving the benefit, from 70% to approximately 90% of all children (Maag and Ramirez 2016), though the poorest children in the United States were still ineligible for the full benefit due to the minimum earnings requirement. This reform also substantially increased the fiscal cost of the CTC to approximately $100 billion. In 2021, as part of the American Rescue Plan Act (ARPA) during the Biden administration, the credit was expanded again, this time turning the credit into a near-universal child benefit akin to the recommendation put forth in the 2019 NASEM report. The benefit was increased to $3000 per child aged 6–17 and $3600 per child under the age of 6. In a marked shift from recent policy trends, the minimum earnings threshold was also removed, along with the phase-in structure. This meant that for the first time in the credit's history, families with no annual earnings could receive the full benefit. A 2025 NASEM report found that this expansion, alongside the EITC, reduced child poverty by more than half in 2021 (NASEM 2026). While Congress debated making this expansion permanent, ultimately it lacked political support and the credit reverted back to its TCJA structure in 2022, where it remains as of 2026. One of the concerns about making the ARPA CTC permanent was that it created disincentives for parents to work, since families could receive the full value of the credit even if they had no earnings in the prior year. Another central concern of the program was its cost—the 2021 ARPA expansion increased the cost of the CTC by approximately $100 billion, nearly doubling the cost of the credit under the TCJA. Further, the 2025 NASEM report found that only 6% of total federal spending on the 2021 CTC went to children living in families below the poverty line. This raises a question about whether we should continue to structure the CTC in such a way that the vast majority of children in the United States receive benefits, or whether we should target resources to those most in need, thus reducing the cost of the program and potentially increasing its antipoverty effects. When we target programs to those most in need, we risk isolating individuals who benefit from those programs. Programs that are more targeted tend to have higher stigma associated with them and enhance the differences between those in need of government support and everyone else. This is the argument made in Ellwood's (1989) Poor Support: Poverty in the American Family, which lays out three “helping conundrums” we face in structuring social policy. Of prime relevance for this question is the targeting/isolation conundrum: we would like to target benefits to those most in need, but in doing so we risk isolating individuals from society. Targeting also requires a mechanism to verify eligibility for benefits, which tends to increase administrative costs of the program as well as administrative burdens on potential beneficiaries. Such targeting can also create adverse incentives, as beneficiaries attempt to avoid “welfare cliffs,” whereby small changes in income could render them ineligible for benefits. Potential beneficiaries may therefore reduce employment to maintain eligibility for benefits. Such behavioral responses risk eroding the antipoverty impacts of the program, reducing overall efficiency. All of these features of targeted programs also lead to weak political support for such programs, increasing the risk of fiscal cuts. One historical example of this phenomenon is the traditional cash welfare program—what is now known as Temporary Assistance to Needy Families (TANF). When the original cash welfare program (Aid to Dependent Children) was established as part of the New Deal in the 1930s, the program was designed to provide economic support for widows raising children. Prior to the establishment of the cash welfare program, it was not uncommon for children to be removed from their homes and placed in orphanages if their father had died, since mothers were not expected to work and could therefore not support their children. It was later recognized that children would be better cared for in their own homes, and the federal cash welfare program was established. When the program originated, women were explicitly not expected to work, and thus the program was not designed to be combined with earnings. At its peak, more than 10 million families received benefits from cash welfare (Duvall et al. 1982). Fast forward several decades and the profile of cash welfare recipients began to shift. Along with the cash welfare program, the New Deal also established the old age pension system in the United States. Once Social Security began providing benefits to widows in the form of survivor benefits, which were based on the deceased or retired spouses’ earnings, the women receiving cash welfare were more likely to be separated, divorced, or never-married women rather than widows. As attitudes began to shift around the culture of women and work, the fact that the program actively discouraged women from working now became a liability of the program rather than a feature. The “welfare queen” persona—the idea of single mothers fraudulently collecting government benefits and living a life of luxury—emerged in the 1980s and reflected shifting public sentiment against the program. Because it no longer targeted benefits to a group perceived as “deserving,” public support for welfare eroded. This ultimately led to its substantial restructuring as part of federal welfare reform in 1996. As a result of these reforms, as of 2021, only about 1.1 million families received cash benefits from the program (Center on Budget and Policy Priorities 2022). While not nearly as dramatic as the cash welfare example, a more recent example of erosion of support for targeted programs is the EITC, a tax credit with a similar structure to the CTC but much more targeted in nature. While historically the credit—which provides a work subsidy mostly to families with children and earnings below approximately 200% of the federal poverty line—has been quite popular and has also enjoyed bipartisan support, there have been reforms in recent years that are indicative of eroding support for this targeted program. For instance, as part of the PATH Act in 2015, tax returns that include a claim for the EITC are held until at least February 15 so that the IRS can have additional time to verify eligibility. In 2025, the House-passed version of the BBBA included a provision to require all tax filers who claim the EITC to provide third-party precertification of eligibility for each qualifying child, a requirement typically used for those facing audits (this provision was ultimately cut from the final bill). Both reforms were introduced because of concerns of fraud. While not explicitly cutting the EITC, these reforms create additional hardships and burdens for those claiming the credit, which will likely reduce participation. At the other end of the spectrum, historically, programs that are more universal in nature tend to have much broader public support. The classic example is Social Security. Perhaps because the program is nearly universal for those with some work history, Social Security has one of the highest rates of public support among social safety net programs. Additionally, because individuals pay into the Social Security system through payroll taxes, there is a broad sense that individuals are entitled to benefits rather than perceiving benefits as a government “hand out.” In a 2024 Pew Research Survey, nearly 80% of respondents said that Social Security benefits should either be maintained or expanded (not cut) (Pew Research Center 2024). Perhaps even more remarkable than this high rate of support for the program is the similar levels of support among Republicans and Democrats—77% of Republicans said that the program should not be cut, compared to 83% of Democrats. This contrasts sharply with how Republicans and Democrats view support for the poor. In the same survey, two-thirds of Republicans stated that government aid for the poor did “more harm than good,” compared to just 23% of Democrats. While other factors certainly contribute to differences in support for public programs among Republicans and Democrats, this example illustrates how isolating the poor through targeted programs weakens political support. Just because a program is universal, however, does not mean that it cannot be progressive. Social Security achieves progressivity not by targeting who is eligible for benefits, but through the structure of the benefit schedule. Monthly benefits are calculated such that the first dollars of earnings have the highest replacement rates, with replacement rates falling as monthly earnings rise. This means that lower income workers receive a higher replacement rate of their wages than higher income workers. It is through these types of benefit schedule adjustments that could make the existing CTC more progressive in nature—improving targeting, reducing costs, and reducing child poverty. The 2025 NASEM report discussed 16 different policy options (PO) to modify the CTC to further reduce child poverty. The report simulated how these changes would impact child poverty, as well as their immediate, annual fiscal costs. I highlight three of those options to illustrate how to increase the progressivity of the credit while maintaining the spirit of a universal benefit and reducing fiscal outlays. I refer to these options as the “cost saver” (PO6 in the NASEM report), “the splurge” (PO11 in the NASEM report), and “the compromise” (PO9 in the NASEM report). I refer interested readers to the full report for more details on each of these options. All of the options discussed here include 17-year-old children as eligible for the credit, and benefits begin phasing out at the income thresholds in place prior to the TCJA reforms: $110,000 for married filers and $75,000 for unmarried filers. While not “universal,” even with these lower thresholds, estimates from pre-TCJA suggested that about 70% of families with dependent children were eligible for the credit (Maag and Ramirez 2016). Further, all the adjustments that I discuss could also be applied to a benefit structure that includes higher income households, though the total fiscal costs would differ. The cost saver (PO6) is most similar to the current CTC except that the credit begins phasing in at the first dollar of earnings, phases in on a per-child basis, and provides a minimum benefit of $1000 to those without earnings. I deem this first option “the cost saver” because its fiscal outlays are slightly lower than current policy, at about $101 billion, compared to $108 billion for current policy (all dollars in 2021$, as used in the NASEM report). In contrast, the splurge option (PO11) increases the size of the credit to $3000 per child aged 6–17 and $3600 per child under the age of 6; the same benefit level as the 2021 ARPA CTC. However, in contrast to the 2021 ARPA CTC, PO11 maintains a phase-in structure, but increases the phase-in rate to 30% beginning at the first dollar of earnings on a per-child basis and provides a minimum credit (worth half of the full benefit) to those without earnings. Additionally, PO11 exempts certain groups from the earnings requirement altogether, including the elderly, disabled, and those with children under the age of 6. I refer to this option as “the splurge” because its fiscal costs are substantially higher than current policy, at about $183 billion. Finally, the compromise option (PO9) is similar to PO6 in that it maintains the lower benefit level ($2000) and begins phasing in at the first dollar of earnings. It differs from PO6, however, in that it phases the credit in faster (30% per child rather than 15%) and maintains an earnings requirement to receive any benefit (i.e., there is no minimum benefit for those without earnings). As with PO11, the compromise does include an earnings exemption for the elderly, disabled, and those with children under 6. This policy is estimated to have a fiscal cost similar to current policy, around $108 billion. How do these policies compare to current policy in terms of their progressivity and impact on child poverty? As one might expect, PO11 (the splurge) would go much further in reducing child poverty than either the cost saver or the compromise, with estimates suggesting that the credit would reduce child poverty relative to current policy by nearly 40%, compared to just 12% for the cost saver, and 20% for the compromise (see Table 1). All three policy options are more progressive than current policy, with similar distributional impacts. Under current policy, children living in poverty represent about 18% of the population but receive only about 11% of fiscal outlays. Children living below 200% of poverty represent about half of all children in the United States but receive slightly less than half of fiscal outlays under current policy (46%). In contrast, each of the three policy options discussed here would increase the share of fiscal outlays going to the poor from 11% to roughly 20% and would increase the share of fiscal outlays going to those below 200% of SPM poverty from about 46% to at least 62%. What this exercise illustrates is that it is possible to improve the progressivity of the CTC, reduce child poverty, and keep fiscal costs down while continuing to provide the benefit to the majority of families with children. This last element—providing the credit to the majority of families with children—I argue is key to maintaining the public and political support for the credit, increasing its likelihood of remaining a key antipoverty program in the United States.