Holzer provides a compelling argument for why the CTC should be more targeted. On many points, we are in agreement. We both agree that we should increase the value of the credit and provide more generous benefits to low-income and nonworking families. To improve the anti-poverty effects of the CTC and limit costs of the program we should take steps to increase the progressivity of the credit. I believe that all of the recommendations that Holzer makes are very sensible. One issue that Holzer raises—how a universal child tax credit would disincentivize work—deserves a bit more discussion. While it is true that any reductions in employment driven by a more universal CTC would mitigate its antipoverty effects, any simulations or cost–benefit analyses should also consider several complicating factors: there is a lack of consensus regarding the magnitude of labor supply elasticities in the literature, the complexity of the US social safety net and its associated work incentives, and circumstances that make it challenging for parents to meet work requirements. The National Academies of Sciences, Engineering, and Medicine (NASEM 2026) report includes three different labor supply elasticities in its simulations precisely because there is no consensus on the magnitudes of labor supply elasticities in the academic literature. While economists generally agree that labor supply elasticities are larger for low-educated, unmarried mothers, particularly those with young children, the precise magnitude of these elasticities is not well understood. Much of the evidence in the literature comes from studying expansions to the EITC that took place in the 1990s, more than 30 years ago and in the context of a very different social safety net compared to present day. Even within this literature, there is wide variation in the estimated labor supply elasticities, ranging from 0.42 to 2.05 (see the NASEM 2026 report for a more detailed discussion of this literature). More recently, the 2021 expansions to the CTC during the Biden administration led to a flurry of studies examining the labor supply responses to these changes. Most of that evidence suggests that there were little to no effects on employment in the aggregate, and modest, negative employment effects for unmarried mothers with young children (see, e.g., Schanzenbach and Strain 2023). Still, even the largest estimates from this more recent literature (implying a substitution elasticity of approximately 0.50) pale in comparison to some of the estimates from the older EITC literature (which were as high as 2.05; see NASEM 2026 report). While the 2021 reform was temporary and the longer-term employment effects of a more permanent expansion are unknown, this only further underscores the uncertainty regarding potential labor supply responses to a permanent universal CTC. The uncertainty about labor supply elasticities is exacerbated by the fact that the US social safety net is a complex web of programs, each with different income eligibility criteria, phase-out structures, and benefit generosity. How the labor supply incentives of a universal CTC would interact with existing tax credits that increase the return to work, such as the EITC, as well as other programs that are slated to require employment in the near future (e.g., SNAP and Medicaid), is unclear. It is therefore difficult to say with any certainty how a universal CTC might alter parental employment in the context of the broader social safety net. Even if a universal credit does lead to modest reductions in parental employment, these effects should be weighed against their potential benefits. There are many downsides to tying receipt of benefits to employment. Parents with good intentions of working may face hurdles, such as childcare constraints and other caregiving responsibilities, or challenging labor market conditions that make it difficult to find work. During times of high unemployment, requiring individuals to work to receive benefits undermines any countercyclicality of such programs that could serve as automatic stabilizers against economic shocks. For instance, Bitler et al. (2017) find that during times of high unemployment, low-income families risk losing EITC benefits, as reductions in employment either reduce the amount of benefits they are eligible for or render them ineligible altogether, depending on the extent of the employment shock. In contrast, many higher-income families might find themselves newly eligible for the EITC as their earnings fall below the maximum earnings threshold required to receive benefits. This phenomenon highlights a sub-optimal consequence of a targeted benefit that is tied to employment: while it can serve as an automatic stabilizer for middle-income families, it exacerbates economic shocks for low-income families. Finally, there is also an argument that if the goal of the CTC is to provide a credit to offset the high costs of raising children (and potentially incentivize individuals to have more children), then the credit should not be tied to employment at all. Instead, pro-work goals should be achieved through other programs (e.g. the EITC). There are some parents that do not work for reasons that might be in the best interest of their families. For instance, there could be negative consequences of maternal employment for those with young children (e.g., Herbst 2017), and many individuals perform valuable caregiving that would be otherwise difficult to replace (Michelmore et al. 2024). Incentivizing or requiring that these individuals work to receive child benefits could have unintended consequences for the broader family unit. Because of these concerns, in some of its simulations the NASEM (2026) report carves out exceptions to work requirements for certain groups, such as families with young children and those with work-limiting disabilities. While Holzer is correct in pointing out that it is difficult to differentiate the “deserving” and “undeserving” nonworkers, these types of carve outs could be more politically palatable (compared to a true universal credit) for conservative policymakers who value pro-work incentives, while addressing the concern among progressive policymakers that the CTC does not provide enough support to the most disadvantaged families. In sum, I am hopeful that the arguments put forth here and in Holzer's essays provide a promising roadmap for policymakers seeking to find common ground on how to improve the antipoverty impacts of the CTC, increase progressivity, and rein in costs. I believe that it is possible to achieve these goals through a thoughtful restructuring of the credit.
Katherine Michelmore (Sun,) studied this question.