What Ross Douthat Misses On The Child Tax Credit

Ross Douthat has responded to my point that the Mike Lee tax plan is extremely distributionally unfair to poor parents. My underlying critique of the plan is that a poor family that has a child will receive far less of a boost from the two child tax credits and personal exemption than a richer family, and this is true all the way up the income ladder until a phaseout is hit. All three of these benefit programs individually have this favor-the-rich effect, and then of course collectively as well.

Some poor parents will receive literally nothing from these benefits. But in the below-graph, which I featured in a prior post, I take a better-off poor family who receives $1500/yr for its twin children and a richer family that receives $8000/yr for its twin children (both of which can and will happen under the plan) and multiply those sums by 18 years. The rich family receives, over the course of the childhood of their twins, $117,000 more than the poor family in benefits. I don't care for that.

Douthat concedes this basic distributive description. This is an improvement over my prior interlocuter Patrick Brennan at the National Review who seemed at first to deny this distributive reality, and then was willing to inadequately and narrowly concede on twitter that this distributive differential would be present for "some" parents. But it's not just "some" parents understood as a distinct and discrete class. The more apt description is that the whole system is designed and patterned so that the more money you make, the more benefits you get, up until the various phaseouts trigger. It's not like only some set class of parents slip through the cracks. Poor families will receive the least, near-poor families the next least, lower-middle class the next least, and so on until we get to families whose market income is high enough to grab all the benefits. So it's a sliding scale of benefit deprivation.

Let's dive into Douthat now. I will only address two of the points because they are the ones I keep pushing and have yet to get satisfactory responses for.


First he says:

I don’t see, and don’t think its architects see, the child tax credit exclusively in terms of its poverty-fighting potential. In policy terms, it’s also about reducing anti-family distortions across the income spectrum; in political/strategic terms (which any policy agenda has to take into account), it’s more about addressing the major concerns of middle and working class voters (the cost of living squeeze, the work-life juggle, the pressures on family stability those create) than it is about crafting Compassionate Conservatism 2.0. 

It's clear that the architects are not concerned with fighing poverty, else they wouldn't have excluded the poor in the way that they have. But I have stressed in my rebuttals again and again that their own non-poverty justifications also render unequal benefits incoherent.

Look at the one Douthat provides here: "reducing anti-family distortions across the income spectrum."

I find this point odd and I don't think it motivates anyone, but I am not going to pursue that thread here. Take the anti-family distortion point for what it is, straight up and on its face. Now ask yourself: does this policy reduce the supposed anti-family distortions caused by Social Security and Medicare across the income spectrum? Does it reduce them for poor parents? No. It doesn't.

The distortion argument is supposed to be that, because of SS and Medicare, people don't feel as pressed to have kids because they don't need them as much as a security against old age. The cost-benefit calculation of having a kid therefore gets skewed by these programs because the benefits of having a kid decline somewhat while the costs remain the same. But, with Mike Lee's child benefits, the scales get tipped back in the other direction. Except not if you are poor and don't get the benefits (or get way less of them)!

Depriving the poor of the benefits is especially problematic on this particular rationale. A poor person in a non-SS and non-Medicare world is especially motivated to have kids because they know they aren't going to be able to put much money away for retirement (unlike rich adults who could plausibly think they don't need a kid because they can save enough). This would mean SS and Medicare especially tips the scales in a "distortionary" manner for them (doubly so when you recognize that raising a kid while poor is much harder than when rich). Yet, the supposed distortion-correcting benefits don't flow to them! So the distortions are not being equally reduced across the income spectrum. Only equal child benefits would do that.

Parent Tax Penalty

Douthat continues later:

As for whether a flat child allowance might be preferable, I am open to the possibility, and might become more open in the future depending on economic trends … but for now, I’m still extremely wary of making a commitment to parents that isn’t tied in some way to work and wage-earning. Linking a child tax credit to the payroll tax makes sense given that that’s where the “parental tax penalty” (which I think Bruenig slightly misunderstands, but I’ll bracket that somewhat theoretical debate) explicitly takes its bite.

I wish he had not bracketed it because I do not think I misunderstand it. I have read it out of a number of mouths and it always says the same thing: the point of these two child tax credits and personal exemption are to compensate parents for the future payroll taxes of their children. This also entails that the current income of the parents should be irrelevant as a basis for determining benefit levels.

Robert Stein's chapter in Room to Grow is the most explicit on this front I have seen so far (Mike Lee's materials are comparably much more vague). He gives two separate ways of calculating appropriate benefit levels. When you are reading them, what you are looking for is whether the rationale depends at all on the income of the parents (hint it doesn't). Here is number one:

The Department of Agriculture says the cost of raising the typical child is $13,600 per year through age 17—and that doesn’t include the cost of saving for college. Considering that Social Security and Medicare will absorb about 25 percent of the labor income of a child born today, sharing the direct financial costs of raising children to the same extent that the benefits of their future labor income will be shared suggests reducing the annual tax bill of parents by $3,400 per child (25 percent of $13,600). 
This argument observes that SS and Medicare will absorb 25 percent of the future labor income of children. This 25 percent figure is then applied to the cost of raising a child. Where does parental income come in? It doesn't at all. If a poor parent has a child, they will undertake the costs of raising the child. That child will also pay 25 percent of the future labor income into SS and Medicare. So surely this $3,400 should go to everyone! The parent's income level is irrelevant to child-rearing costs and the percent of future income their child will pay in payroll taxes. So why is it being used to determine benefits?
Here is Stein's second basis to determine benefit levels:
Another way of determining the appropriate amount of tax relief for parents is to consider the present value of future Social Security and Medicare contributions for a typical worker born today, which is about $160,000.
Rewarding parents for making these future contributions possible suggests annual tax relief of about $9,000 per child. (The contribution figure is what matters, because today’s children will get benefits only if they as a group raise children, regardless of whether they’re “promised” benefits under current law.) 
Here, Stein takes the present value of the child's future payroll tax contributions and says that is what should be kicked back to parents. He specifically says this is "rewarding parents for making these future contributions possible" and even parenthetically says "the contribution figure is what matters." Once again then, the parents' current income is irrelevant. A poor parent will raise a kid that pays payroll taxes in the future, just like a rich parent. If the appropriate relief is the "present value of future Social Security and Medicare contributions for a typical worker born today" (meaning the kids of parents), then every parent should get equal benefits.
If we return to those hypothetical couples with twins at the top and use these future payroll tax contributions to stand in for the amount of the parent tax penalty (as Stein does), we find that this is what both sets of parents "pay" in the parent tax penalty:
The same amount, that is.
Child Allowance
Douthat says he could be open to a child allowance. I really think this is a good idea that conservatives should be able to get behind. All of their arguments for their plan support equal benefits for all families not the unequal benefits they have oddly settled on. It's also the case that red states are poorer states and so their child benefit scheme is a bit politically strange as well. As to its practicality, it's important to know that a lot of European countries have it already and they do fine. In fact, they do more than fine. Here are 6 such countries (there are more) with a child allowance and their relative child poverty rates just for fun:
  1. Denmark — 3.7% — Nordic Social Welfare State
  2. Finland — 3.9% — Nordic Social Welfare State
  3. Norway — 5.1% — Nordic Social Welfare State
  4. Iceland — 7.1% — Nordic Social Welfare State
  5. Sweden — 8.2% — Nordic Social Welfare State
  6. Austria — 8.2% — The Vienna System

Keen observers will note that these are the lowest child poverty rates in the world. I've calculated a $300/mo child allowance here would cut child poverty in the US by 42 percent. This is in line with the child benefits Mike Lee is proposing, except it reduces poverty significantly because it doesn't exclude the poor so much.