Hey readers,

Here in the US, we’ve been engaging in an unusual experiment over the last few months. Since July 15, parents have been getting monthly checks for at least $250 per child ($300 per child under 6).

The provision is an expanded version of the existing Child Tax Credit (CTC), which has been around in some form since 1997. But the Biden administration, as part of its stimulus efforts, passed a dramatic expansion of the credit, so it’s much larger (up to $3,600 a year, up from $2,000), is paid out monthly (instead of yearly), and has no work requirement.

That last part is key, as it meant ultra-poor families without any taxable earnings, families who were totally excluded from the CTC as it existed before 2021, could benefit for the first time.

I’ve been writing about this idea, called a “child allowance” and common internationally, since 2016, so I was incredibly excited to see it expanded as part of Biden’s stimulus plan. I cited an analysis suggesting that Biden’s child credit would cut child poverty by over 40 percent.

Now, as Congress debates extending the new credit so it doesn’t expire at the end of the year, a new paper suggests the credit might be less effective at reducing child poverty than I and others imagined.

Bruce Meyer, a veteran poverty researcher at UChicago who is particularly good at puncturing received wisdom in left-of-center circles, wrote the paper with Kevin Corinth, Matthew Stadnicki, and Derek Wu.

Their basic argument is that the old CTC encouraged parents to work. It phased in with income, meaning there was a de facto work requirement to receive it. By removing that incentive, the paper argues the expanded CTC will spur 1.5 million people to stop working, which dilutes its anti-poverty impact.

Using their preferred estimates, the new child credit only reduces child poverty by 22 percent, down from 34 percent before they considered work effects.

Meyer is a serious researcher, and this paper is no partisan hit job. CTC supporters should take these arguments seriously. But I don’t think this paper is the death knell of the proposal that some detractors are making it out to be.

Why Meyer’s estimates might be too large

The first reason is that a 22 percent reduction in child poverty is still not too shabby. It would be disappointing if the child tax credit were a less efficient means of reducing child poverty than we thought, but I’d be willing to bite the bullet and support the policy even if it “only” reduced child poverty by a fifth.

Secondly, after receiving the paper I canvassed other economists working in this area to get their takes. Many said the new paper’s estimate of how many people would stop working because of the tax credit's expansion was too high. That estimate drives the paper’s projection that recipients’ incomes from work will fall, which in turn drives the finding that the reduction in child poverty will be smaller than assumed.

Robert Moffitt is a professor of economics at Johns Hopkins University and a co-author of the 2019 National Academy of Sciences report on child poverty, which simulated the effects of a policy like the one Biden adopted this year. He told me that the NAS report didn't estimate an effect from removing the child tax credit’s phase-in, because “the bulk of estimates from economists show that the part we ignored is small in magnitude.”

“The estimates from the [Meyer] paper are wild,” Moffitt wrote in an email, noting that it suggests “16 percent of all families in the US would choose to stay in poverty because they don’t want to work rather than get out of poverty. All for a $4,000 payment. These estimates are really ‘out there.’”

The Meyer paper relies on an estimate of what economists call an “elasticity”: an estimate of how much the probability a given person will work changes, in response to the financial rewards of work increasing or decreasing. A high elasticity suggests that lowering the rewards to work, by de-linking the CTC and employment, could lead a lot of people to choose not to work.

Meyer uses an elasticity of 0.75, relying on the midpoint estimate of a Congressional Budget Office review of the evidence from 2012. UC Berkeley’s Hilary Hoynes, who also worked on the NAS report, argues that research in the ensuing decade suggests that 0.75 is too high, and that an elasticity more like 0.3 is reasonable. That would suggest far fewer parents quitting the workforce due to the CTC expansion.

Jacob Goldin and Kathy Michemore, economists at Stanford and University of Michigan respectively, and the Urban Institute’s Elaine Maag recently released their own paper that examined recent research on elasticities for poor Americans. They argue that the labor market effects of the expanded child tax credit would be quite small.

Goldin says that the Meyer paper is "assuming that hundreds of thousands of parents will walk away from jobs paying $30,000 and $40,000 per year, and tens of thousands of parents will walk away from jobs paying $70, $80, or $90,000 per year, all because they can now receive an extra $2,000 or $3,000 extra dollars of CTC. It just doesn't match common sense."

Meyer argues his large unemployment estimate matches common sense. In a phone call, he cited papers from the very people criticizing him now — like Hoynes or Moffitt — that found that the earned income tax credit (EITC), which functions similarly to the pre-Biden CTC, dramatically promotes work. Why, then, shouldn’t getting rid of the old CTC structure deter work?

Similarly, he notes that in the 1990s, the EITC was dramatically expanded, and old-school welfare, a cash program with no work requirement, was ended. Subsequently, more than a million single mothers joined the workforce. Doing the reverse, then — getting rid of an EITC-like program and replacing it with a cash program with no work requirement — might push more than a million single mothers out.

I get where Meyer’s coming from, but I still have some intuitive trouble with the story he’s telling. It’s easy for me to imagine that a single mom on cash welfare would, when faced with the loss of her benefits and a big tax credit if she works, join the workforce in the 1990s. It’s harder for me to imagine a single mom in 2021 earning $30,000 quitting her job because a work-based tax credit went away and was replaced with unrestricted cash. The situations don’t seem symmetrical in the way Meyer appears to assume.

As it stands, I still think the CTC expansion is a good idea, even if this new paper is right and it merely reduces child poverty by a fifth, not two-fifths. But it’s always important to interrogate your assumptions, and the Meyer paper is a service in that regard.

In any case, we’ll keep you updated on the latest research here. I suspect we’ll learn a lot in the coming year about the effect Biden’s policy had on the labor force — and whether Meyer’s predictions come true.

—Dylan Matthews, @dylanmatt