Pathways to reduce child poverty: Impacts of federal tax credits

On February 5, 2026, the Center for Economic Security and Opportunity (CESO) hosted a briefing on a new report by the National Academies of Sciences, Engineering, and Medicine (NAS), “Pathways to Reduce Child Poverty: Impacts of Federal Tax Credits.” The report examines the implementation and effects on child poverty of the temporary expansions of the Earned Income Tax Credit (EITC) and Child Tax Credit (CTC) under the American Rescue Plan Act (ARPA) of 2021.

Introduced by CESO Director Tara Watson, V. Joseph Hotz of the University of Chicago summarized the report’s key findings. His presentation was followed by a panel moderated by Diane Whitmore Schanzenbach of Georgetown University, featuring Dolores Acevedo-Garcia of Boston University, Marianne Bitler of UC Davis, Michael R. Strain of the American Enterprise Institute, and James Sullivan of the University of Notre Dame.

Hotz began by summarizing the core features of the EITC and CTC and the major changes enacted in 2021. While ARPA made adjustments to the EITC that mainly benefitted childless workers, the CTC operated more like a “child allowance” in 2021: it became fully refundable (meaning families without earnings or tax liability could claim the full amount), benefits increased, and payments were delivered monthly. The NAS report finds that the combined 2021 EITC and CTC lifted more than 2 million children out of poverty, reducing child poverty by 2.9 percentage points—more than halving it relative to a scenario with no credits in place. These effects were similar after accounting for the weaker work incentives in 2021 relative to other years.

The NAS report also includes an analysis of 16 alternative policy options compared to the existing EITC and CTC, which capture potential variations in eligibility, refundability, credit amounts, and phase-in and phase-out structures. The analysis reflected the idea that employment incentives are likely to matter more in the context of long-run policy adoption as opposed to a temporary expansion. Hotz emphasized a consistent theme across each policy option: there is always a tradeoff between poverty reduction and fiscal costs.

The panel discussion built on these findings by examining how policy design and implementation affect different populations and broader outcomes. Acevedo-Garcia highlighted how policy tweaks have unequal effects across racial, ethnic, immigrant-status, and family-structure lines due to variations in baseline child poverty rates. She noted that while one in four U.S. children lives in an immigrant family—most of them U.S. citizens—disparities in program access persist, and Social Security number requirements under the One Big Beautiful Bill Act may worsen them.

The panelists highlighted a need for better data infrastructure to guide policy. Sullivan pointed to gaps in current reporting, noting that in 2021, only 28 percent of unemployment insurance benefits and 42 percent of SNAP benefits were captured in population surveys. Bitler argued for investment in a microsimulation model that could integrate information on means-tested safety net and social insurance programs with the tax system.

This need for better data framed a broader debate over the fundamental purpose of these credits as sources of tax relief or long-term upward mobility. Sullivan argued that effective poverty reduction requires targeting the bottom of the income distribution, where income transfers yield the most substantial improvements in outcomes. Strain cautioned against overemphasizing the “magic threshold” of the poverty line, suggesting that policy goals should be focused on fostering broader self-sufficiency. Bitler bridged these perspectives by citing evidence that EITC exposure produces long-term effects on outcomes beyond just poverty and workforce participation, including improved education and higher future earnings for children.

The discussion concluded with a consensus that these credits differ fundamentally from traditional “cash welfare” due to their structural emphasis on work. While Strain pointed out that the 2021 CTC did not have the same reward to work as the current version, the panelists agreed that today’s EITC and CTC operate are distinct from the safety net before 1996 welfare reform. Acevedo-Garcia noted that conflating these credits with “welfare” reinforces stigma and effectively limits access.