For six months in 2021, the United States ran a quiet experiment in family policy. The expanded Child Tax Credit went monthly, became fully refundable, and reached families with little or no earned income. Census data showed child poverty falling from roughly 9.7 percent to 5.2 percent — the lowest on record. Then Congress let it expire, and child poverty more than doubled in 2022. The lesson many drew was about cost. The more important lesson was about design.
Means-testing creates the cliffs critics complain about
The CTC’s pre-2021 structure phased in benefits as earnings rose and excluded the poorest families because the credit wasn’t fully refundable. The 2021 version flipped that, sending the same baseline benefit to nearly every family with kids and phasing out only for high earners. Critics insisted on returning to “targeted” design to control costs and avoid sending money to people who “don’t need it.” But every layer of means-testing introduces administrative overhead, paperwork errors, missed eligibility, and political vulnerability. Universal programs — Social Security, Medicare — have proved durable because their constituency is everyone. Programs aimed only at the poor are easier to cut because the people benefiting have less political power.
Universal cash is administratively cheap
A universal monthly payment per child, delivered through the IRS or a successor benefits agency, runs at a fraction of the overhead of stacked tax credits with phase-ins, phase-outs, earnings tests, and dependent rules. Canada’s Child Benefit and similar programs in the UK, Germany, and Australia demonstrate that broad child allowances are administratively boring and reach the families who need them most. The 2021 IRS rollout, despite limited time and outdated infrastructure, delivered payments to roughly 61 million children. The argument that the government can’t write monthly checks to families is empirically false. It just did.
The targeting argument misses the point
The strongest objection — that high-income families don’t need the money — confuses the goal. A child allowance treats the costs of raising children as a shared social investment, the way public schools do. We don’t means-test kindergarten. The progressivity comes from the tax code, not from carving out who qualifies. Recovering some of the benefit at high incomes through marginal rates is cleaner than building a phase-out into the credit itself, and it preserves the political coalition that protects the program when budgets tighten.
The bottom line
The 2021 expansion produced one of the largest single-year drops in child poverty in modern American history. Letting it lapse was a policy choice, not a fiscal necessity. Rebuilding it as a stripped-down, near-universal monthly payment would lower administrative friction, deliver more predictable family budgets, and harden the program against future political swings. Means-testing was sold as making the credit fairer. In practice it made the program harder to access, easier to attack, and weaker at the one job — keeping kids out of poverty — that actually matters.
Leave a Reply