The Child Tax Credit Ambiguity in the GOP Tax Plans That Will Make or Break Tax Reform for 6 Million Parents

Parents haven't been the main focus of either the House and the Senate tax plans thus far. The bulk of the benefit in both plans has consistently come from business tax relief, even as they have evolved to stay within fiscal constraints. By contrast, while both plans expand the child tax credit, they have fallen short of ideas from both parties to direct more tax relief to parents, such as the Bennet-Brown and Rubio-Lee plans. The House bill even initially repealed the adoption tax credit, before restoring it later.

Both plans will help many parents early on before individual tax relief expires and the slower chained CPI inflation adjustment has time to bite. Exactly how many however hinges on a key ambiguity around the Child Tax Credit that arises from the way they expand the child tax credit. Six million primarily working- and middle-class parents could get either a tax hike or a tax cut from the Senate bill in 2018 depending on how this expansion is interpreted. Millions more will see larger refunds or smaller tax bills.

To be clear, this ambiguity isn't relevant in the later years in the Senate bill, since that plan reverses virtually all of its individual tax changes after 2025, and the number of parents who are worse off rises substantially.

To be equally clear, JCT seems to already interpret the House and Senate bills in the more generous way. But given the lack of clarity in the legislative language, this is not necessarily conclusive for how the final tax plan, should it become law, gets interpreted by the IRS or others later. More clarity in the final bill is needed.

The Additional Child Tax Credit Under Current Law

The ambiguity arises from the Additional Child Tax Credit (ACTC). Under current law, the regular Child Tax Credit is $1,000 per qualified child, but it's also nonrefundable. This means it can only offset any federal tax you owe. If you owe no tax, the regular CTC itself cannot give you any more benefit.

Enter the ACTC, which takes the regular CTC and makes some of it refundable for certain families. Sometimes this is called the "refundable portion" of the CTC. To simplify it: if you have any "unused" portion of the regular CTC -- meaning you can't claim the full $1,000 per child regular CTC benefit because your federal tax liability falls to $0 first --  then that unused portion spills over into the ACTC and gets refunded back to you, even if you don't have any federal tax liability.

There are several restrictions on this however; for example, the ACTC benefit is capped at 15% of your earnings, and your first $3,000 of earning don't count for this calculation. Other restrictions apply as well. In practice then the total benefit of the CTC plus the ACTC may be lower than $1,000 per child for some families.

Despite these limitations -- and again, simplifying here -- it might be helpful to our discussion of the House and Senate plans in a moment to think of the CTC/ACTC in tandem as essentially being a single (potentially) refundable credit of up to $1,000 per child under current law. As I mentioned before, this isn't strictly true in all cases, but it illustrates the key take-away that any dollar used under the regular CTC to reduce federal tax liability reduces the potential amount of the ACTC 1-for-1. Because of this, the two pieces are in alignment with one another.

Illustration of the House and the Senate Plans

The House and Senate bills however bring the two out of alignment.

The Senate bill raises the regular CTC to $2,000 per child while the House bill raises it to $1,600. Both bills meanwhile only raise the ACTC to $1,100 (though they do allow the ACTC to gradually grow over time with inflation until it converges with the regular CTC, a process that is likely to take well beyond 10 years).

Interestingly -- and counterintuitively -- even though both the CTC and ACTC have been expanded under both plans, it's still possible under some interpretations that a family could end up with less benefit. I'll use an illustrative example to show how.

Take a single mom whom I'll name Janet with 9 year old twins in 2018, earning $36,000 in wages and filing as Head of Household (I crafted this example specifically to avoid some of the complexities of the ACTC mentioned above that aren't relevant to my core point).

Under current law, which is the first column in the table above, Janet takes her $36,000 in wages and deducts $4,150 each for herself and her two kids in personal exemptions (for a total of $12,450). She also takes the standard deduction of $9,550. So her taxable income is $14,000. Most of this is within the current 10% income tax bracket with a small amount in the 15% bracket. Janet's initial federal income tax liability before credits is $1,420.

The nonrefundable regular CTC offsets all of this $1,420, and because she doesn't get to claim the full $1,000 per child of regular CTC before her tax liability gets to $0, the "left over" portion -- $580 -- becomes ACTC and gets refunded to her. Adding in the $2,085 of Earned Income Tax Credit (EITC) she gets from working, Janet's bottom line refund is $2,665.

Now let's look at the Senate plan, which has the more generous CTC expansion.

The Senate plan changes Janet's math in several ways. Gone are the personal exemptions, but the standard deduction almost doubles. The 15% bracket becomes a 12% bracket. The net effect of these changes is that her federal income tax liability before credits rises to $1,888.

Since the child tax credits are larger in the Senate plan, this higher before-credit liability may be offset. But it depends on the order in which the CTC and ACTC are applied.

Put another way, the Senate plan is the equivalent of changing our current CTC, which we can think of in an oversimplified sense as a single credit potentially refundable up to $1,000 per child, into a $900 per child nonrefundable credit and an $1,100 per child refundable credit in 2018.

Thought of in this way, the question then becomes which credit "goes first" in bringing down Janet's $1,888 in federal income tax liability.

If the nonrefundable credit gets drawn down first, as in the second column above, then that means the nonrefundable credit offsets $1,800 in tax liability on its own before it's all exhausted. But Janet can then also get the full $2,200 of the refundable credit, since it doesn't exceed 15% of her income less $2,500 (the income adjustment was also changed in the Senate plan).

But watch what happens if the refundable credit gets drawn down first, in the third column. Janet uses the first $1,888 of her refundable CTC to offset her tax liability, and can then still claim its last $312 as a refund so that she gets all $2,200. But notice how now there's no tax liability left over for the nonrefundable CTC to cover. So Janet gets none of the $1,800 nonrefundable credit.

This ends up making a substantial difference for Janet: her bottom line tax refund either grows by $1,532 or shrinks by $268 depending on how this question is answered.

The $10 Billion Question

Nor is Janet unique. The difference in interpretation here is significant, particularly for working- and middle-class parents. Using the excellent OSPC TaxBrain model [1], I find that the more generous nonrefundable-first interpretation of the Senate bill means roughly $10 billion more in tax relief flows to parents in 2018. More than 96% of this added benefit goes to parents in the $20K-$75K income range. The average tax change among all parents is an almost-$200 larger cut versus the less generous refundable-first approach; for parents between $30K-$50K, their average after-tax income goes up by two percent more under the more generous approach. Roughly 5.6 million fewer parents overall (11% of all parent filers) get a tax hike. For parents making between $40K-$50K, roughly half will go from a hike to cut depending on the credit sequencing.

The "Right" Interpretation

The "nonrefundable-first" interpretation is the more generous one, and based on my own conversations seems to be the one that JCT uses in their scores of the House and Senate bill. For my money, it's also the one I think makes more sense.

But JCT's is not the only interpretation, and other economists have been approaching it the opposite way. OSPC for its part models the refundable credit as being drawn down first by default. One can make either case for the "right" interpretation here: the current law CTC / ACTC sequencing on the IRS Form 1040 suggests the nonrefundable credit should get drawn down first, but this is an untested assumption since hitherto the CTC and ACTC have been in alignment. The fact that greater draw-down of the CTC under current law brings down the ACTC in a zero-sum manner lends itself to a "refundable-first" interpretation.

Either way, the ambiguity here is significant for parents, and greater clarity is needed in the final bill. 

[1] My analysis includes the indirect benefits of cutting the corporate income tax, distributed in a manner consistent with JCT. It does not include any of the primary or secondary effects of the repeal of the ACA individual mandate, as I lack sufficient data to analyze this at the individual level in OSPC's model.

More on why so many parents lose under the TCJA

As a followup to my post earlier this morning, I've created a table below detailing how each policy in the TCJA affects the number of parental families in 2027 seeing a tax hike.

Using refined policy parameters that have been added to the OSPC model since yesterday, I find that under the TCJA as written, 22 million families with children would see a tax hike in 2027 versus current law. This includes the indirect effects of the corporate income tax cut, but assumes that the filer credits expire in 2023.

That 22 million figure is almost exactly half of the 44 million total families with kids projected for 2027.

I've stacked the policies left to right in roughly the same order as in JCT's score of the TCJA. You can see all the way to the right in bold the 22 million number at the end of the running sum row. So this table is showing how each policy in the TCJA incrementally changes the number of parents who lose out, until we get to the 22 million bottom line.

The story I would tell goes something like this:

1. Right off the bat, 25 million parental families get hit by the new income tax bracket structure.

2. The higher standard deduction and the expansion of personal & child credits are a little more than enough to offset the hit to parental families harmless from the repeal of the personal exemptions: the hike count falls by 2.5 million on net from these three policies combined. If the child tax credit expansion were more generous & refundable, this would be more.

3. Chained CPI adds another million losers, and the miscellaneous limits and repeals of deductions and credits add another 4 million.

4. The corporate tax cuts, AMT repeal, and passthrough rate caps are enough to offset the raw number of losers from #3 above, but their benefits skew more upper income. There are 2.3 million more families making $100K-$200K who get a tax hike from these policies together, for example.

Many parents face higher taxes under TCJA, even if Congress makes its credits permanent

This quick post looks at the effects of the Tax Cut and Jobs Act (TCJA, the House GOP's proposed tax plan introduced last week), but (selfishly) I focus on a specific segment of the population: families with children under 18.

It turns out that parents do far worse under the TCJA than the population as a whole, and making the expiring credits in the TCJA permanent only modestly changes this story. More than 40% of families with children face a hike under TCJA in 2027, even with the security of permanent filer credits and assumptions about the benefit of corporate tax cuts.

As always, I'm using the excellent Open Source Policy Center (OSPC) TaxBrain model for this analysis.

To set the table: under the TCJA as written (what I'm calling "TCJA law"), 72 million filers or about 37% of the total get a direct tax increase in 2027. This number falls to 50 million, or 26%, if I distribute the benefit of TCJA's corporate tax cut similar to how JCT did in their score of TCJA last week. [1]

But filers with children under 18 fare much worse than the overall population. In 2027 I estimate that there are 24 million families with kids who see a direct tax hike, 55% of all such families. Distributing the corporate tax cut lowers this by 4 million to 20 million, or 45% -- still a far higher proportion than all filers writ large.
In fact, both with and without the corporate income tax, the median tax change among all families with kids in 2027 is roughly $0 under TCJA law, though there are wide disparities in outcomes even within the same income band.
The rebuttal made by supporters of TCJA is that there are benefits phased out in the legislation to save money that, in reality, would not expire, specifically the filer and nondependent credit. Under TCJA these expire in 2023, and doing this rather than keeping them permanent saves around $200 billion over 10 years. But the claim here is that Congress will feel compelled to extend them before they expire, as it does with many other ostensibly "temporary" provisions in the tax code.

I wanted to test the substance of this argument, so I also simulated what would happen if Congress made these two credits permanent. Would families with children be that much better off under the TCJA?

The results below show that the answer is a modest "yes". About 1.9 million more families would get a tax cut in 2027 if Congress made these two credits permanent ("TCJA Policy"), an additional 4% of all families with kids.

We can see by taking the difference between the TCJA Law and TCJA Policy dispersions to find out who the net beneficiaries are of making these credits permanent. Unsurprisingly, families making under $100,000 in 2018$ see almost all the benefit.
What this tells us however is that a lot of parents will be worse off under TCJA even if the filer and nondependent credits are made permanent. That's because the law makes several changes -- elimination of the personal exemptions, a rise in the lowest bracket to 12%, a slower inflation adjustment for benefits like the Earned Income Tax Credit (EITC) -- that hurt families.Often the changes the law makes in support of families such as the increases in the standard deduction and the child tax credit, and the introduction of the filer and nondependent credits, are not enough to make families whole.

In other words, even if these credits become permanent and even giving the plan the benefit of its corporate tax cuts, more than 40% of parents face a tax hike under the TCJA.

The good news is that members of both parties have introduced plans that could serve as templates for shifting the TCJA towards helping parents, including the Bennet-Brown plan from Democrats and Rubio-Lee among Republicans. The details of these plans differ but the broad strategies involved -- an expansion of the child tax credit and/or EITC with some mechanism for making the child tax credit more refundable -- are compatible.

What's clear though is that the TCJA as written poorly serves parents. As the House and Senate negotiate, they should overhaul it to make it a true pro-family and pro-child piece of legislation.



[1] For the purposes of this post, I'll graph the results of the corporate-tax-inclusive analysis, but it's important to remember that the benefits of a corporate tax cut to families are 1) indirect (i.e. in the form of higher wages or investment returns rather than lower taxes owed on a 1040), 2) the subject of fierce disagreement in economics, and 3) likely greater than zero but sensitive to how the tax cut is financed.