In my last blog post, I identified some of the challenges that arise when the Child Tax Credit is converted from an annual pay-out to a monthly pay-out. In this blog post, I will explore some alternative approaches that may address, if not resolve, those challenges. I have to confess that I am muddling through this – none of this is easy, and there are trade-offs. Before I explore some alternative approaches, it may be helpful to lay out some key design considerations.
Second, the program has to be administrable – both from the IRS’s and the taxpayer’s perspective. The minimal amount of administrative burden should be placed on taxpayers, while maintaining program integrity – i.e., we can’t design a process that is so easy to administer that it results in a significant amount of payments to the wrong individuals, because that will erode public and political support for the program. There are a lot of levers here to avoid that result – e.g., refining the definitions of eligible individual or qualifying child; establishing safe harbor amounts; paying out only a percentage of the full credit. Administrability also has to take into consideration the IRS’s technology and staffing capabilities, taxpayer’s ability to navigate whatever systems and procedures are developed, and the external support networks necessary to ensure taxpayer participation in the program.
Some have expressed concerns that going from an annual payment to monthly payments will be a heavy lift for the IRS. I don’t think the issuance of monthly payments in and of itself is a problem. The challenge for the IRS is getting the payments to the right household. The IRS will do what the law instructs it to do; it needs clear guidance, and if guidance is vague it will resolve the ambiguity in the manner that makes it easiest for the IRS to administer and that minimizes the risk of incorrect payments. Because the IRS is an enforcement-minded agency, it will always err on the side of not issuing a payment where ambiguity exists, unless the law clearly instructs it otherwise. Design features such as safe harbors and bars on offset against other tax refunds are clear legislative statements that it is understood and accepted some level of error is unavoidable when administering the program. By minimizing (by not eliminating) clawbacks, these features allow the IRS to direct its resources to assistance, education, and outreach to minimize avoidable errors, as well as focus its enforcement actions on actual abuse and fraud.
And to avoid unnecessary errors, the monthly advance CTC should align with other tax provisions – the EITC, the American Opportunity Tax Credit (AOTC), head of household status. There will have to be differences between the programs because some are administered annually and one monthly, but the concepts underlying eligibility rules of all these programs should be fairly consistent in order to avoid confusion and illogical claims. For example, we shouldn’t design a program in which a taxpayer can receive the EITC for a given child on the tax return but be barred from claiming the CTC for that child on the return because irrebuttable presumptive eligibility awards the monthly CTC to another person.
Third, the program’s goals must be clearly defined. Is the point of the Child Tax Credit to alleviate child poverty, or, as I discussed in my earlier blog, is it still to reflect the impact of family size on a taxpayer’s ability to pay tax? Are we trying to do both, and if so, are we prioritizing one goal over the other? If the goal of the program is to alleviate child poverty, then avoiding clawbacks of the credit and smoothing the amount of the credit (consistent payments throughout the year, without fits and starts) will be high priorities.
Fourth, the design of the program should accommodate the characteristics of the population that will avail themselves of the CTC. Elaine Maag, Elizabeth Peters, and Sara Edelstein show that between 1996 and 2008 the composition of US households changed significantly. By 2008, although the majority of children were in households with two parents, their share of households had declined from 70.9 percent to 67.3 percent. The percentage of children in households with cohabiting couples and non-parent households increased over that period. Single parent households, regardless of income levels, remained relatively constant. These patterns hold true for low income households. Thus, the CTC needs to work for a wide variety of households; its eligibility rules should reflect that diversity.
Fifth, and this seems to me to be very important, how significant is the behavioral change we are asking of individuals who are currently claiming or will be eligible for claiming the CTC? Are we introducing discrepancies into the current system that will confuse currently eligible individuals or increase the likelihood of errors? We will need to consider how the introduction of advance monthly payments affects taxpayers already engaged in the tax system.
In Tax Year 2018, according to IRS Statistics of Income, the CTC was claimed on almost 39.4 million tax returns for almost $81.5 billion. Over 26 million taxpayers claimed the Additional CTC, for about $36 billion; 19.5 million of these taxpayers received $19.5 billion as refunds (i.e., not offset against taxes owed). For that same year, over 26 million taxpayers claimed the EITC, totaling almost $65 billion; over 23 million of these EITC taxpayers received $56 billion in refunds. That is a lot of human beings who are currently claiming the CTC and the EITC and who will be affected by any changes in the design and delivery of the credit. On the other hand, making the CTC fully refundable with no income requirement clearly benefits a significant number of families. Jacob Goldin and Katherine Michelmore estimate this change makes 5.9 million children, primarily from Black and very low income households, eligible for the credit. Some of these households are currently not engaged in the tax filing process (although some may have filed returns through the nonfiler portal in 2020 and 2021 in order to receive economic impact payments).
Currently, under the IRC, the CTC is awarded to an eligible individual with respect to a qualifying child (QC) under IRC § 152(c) (the rules for claiming the dependency exemption). In addition to having a specific relationship to the child, and other rules relating to age, support, etc., there is a residency requirement – the QC must have lived with the eligible individual for more than half the year. Essentially, the relationship and residency tests are proxies for the “care and control” standard found in many benefit programs and the “main carer” standard found in some tax credits in other countries, including Australia. The result of the 6 months-and-a-day test is that it is all or nothing – if you meet the test, you get the entire credit; if you don’t meet the test, you get nothing. Under the test, if the child moves around a lot, it is possible no carer gets the benefit of the credit.
If we base monthly eligibility on where the child resides for that month, how is residency determined? Do we determine monthly residency based on where the child is on the first day of the month, or the last day of the month, or the majority or most days of the month? What if the child splits her time 50-50 between parents? What if the split is 51 percent/49 percent?
We could allow both “main carers” to allocate the month’s credit between themselves and notify the IRS that they are making the election in that way (Australia allows this). The IRS has said it will split the Advanced CTC between parents who filed Married-Filing-Jointly in 2020 (or 2019) if they notify the IRS that they have separated and will file separate returns for Tax Year 2021. While under the current law one of these payments would be considered incorrect, the law could be amended to incorporate this concept of two (consensual) main carers. The Ways and Means Committee Working Draft of the Building an Economy for Families Act here adopts this approach. The IRS would only get involved where one main carer will not agree to split the credit, in which case the all or nothing residency test for that month applies (in addition to all the other IRC § 152 eligibility requirements – relationship, support, etc).
Regardless of which test we apply, there will be changes of circumstances. Maag et al found that while most children live in the same household within the tax year, 4.6% change parental households at least once in a tax year and 1% change 2 or more times within the tax year. Perhaps because of this volatility in households, presumptive eligibility proposals shift the burden of updating for change of circumstances to the newly eligible household to apply; the household currently receiving the monthly payment has no responsibility to update. And because there is an irrebuttable presumption that once you’ve received a payment for the month, you are eligible for it (absent fraud), there are no erroneous payments. This violates my cardinal rule – that we should try to pay the AdvCTC to the person who actually cares for the child in that month.
I understand the concern that when children move out of a home it is often because that home is in turmoil or under deep stress, so it may not be possible to expect an immediate update of circumstances. I fully understand the challenges low income taxpayers face every day – after all, representing and advocating for low income persons has been my life’s work. Still, it seems to me that the responsibility to update change of circumstances should be borne by everyone, regardless of income level. Then it becomes a question of how easy we can make the updating process; and whether we need to accommodate the exigencies and stresses of poverty by having safe harbors in order to avoid serious clawbacks.
There also seems to be a belief that the new claimant household will navigate the administrative burden of filing a claim for the QC immediately upon becoming eligible. I personally think that is a flawed assumption. If a child moves into a new household, there are many things that need to be addressed, and it may be a few months before the new household learns about or applies for the monthly CTC. Yet presumptive eligibility is not retroactive to when the child actually lived with the taxpayer; it will only go back to when the taxpayer applies for it. And in the meantime, the original household continues to receive the monthly CTC for a child that does not live with them.
So here are my thoughts about how we could reconcile all these somewhat competing considerations. I put this out here so that others can comment and make suggestions. I know there are issues I haven’t addressed, or I’ve made choices that others might not make. But I have tried to think through how to implement a monthly tax credit through an annual tax system without doing too much violence to that system. I may be splitting the baby here (with apologies to King Solomon), but I am continuing to come down on the side of having the monthly CTC paid out as an advanced CTC, with year-end reconciliation required. Also note that I am calling persons who apply for the AdvCTC “taxpayers” – this is a deliberate reminder that we are administering this program through the tax system. These individuals have now entered a different system from the traditional benefits system. We cannot, nor should we, ignore that fact.
- The monthly AdvCTC is based on where the child has a principal residence, so long as other CTC eligibility requirements are met (relationship, age, etc.). Principal residence would be defined as where the child resides for more than half the month (or for the greatest number of days in the month where the child does not reside with one household for more than half the month).
- Taxpayers may apply for the AdvCTC on the tax return for the prior year or during the year via the change-of-circumstances portal. If they do not want the AdvCTC for their qualifying child, they must affirmatively state this, so the IRS can identify competing claims. If there are no competing claims and the claim passes the usual questionable refund/identity theft/dependent database screens, the monthly AdvCTC will be paid out to the electing taxpayer during the year.
- The presumption that the electing taxpayer is the eligible taxpayer is not irrebuttable, however. The electing taxpayer has the obligation to update the IRS when the household is no longer the principal residence of the child for any given month, other than temporary absences. (There are many definitions of temporary absence in the IRC, but regulations may need to define this for purposes of the AdvCTC.)
- Taxpayers who successfully elected on the prior year tax return to receive the AdvCTC would receive the first credit on or about May 15, for the month of April. Payments would continue throughout the year (unless the taxpayer notifies the IRS circumstances have changed or a dueling claim arises) until January 15, which is the payment for the month of December. The monthly AdvCTC would not span tax years. The taxpayer could receive up to 9/12ths of the annual CTC in advance.
- By January 31 of the next year, the IRS would be required to send every taxpayer receiving the monthly AdvCTC a form showing which monthly payment was received for a qualifying child (Form 1099-CTC?).
- The electing taxpayer would use this form to file a reconciliation schedule with their return (not just to allow for claiming the CTC retroactively but because more than 6 months of the monthly CTC creates potential eligibility for the EITC, CDCC, and AOTC). For very low (or no) income taxpayers, the reconciliation could be done through a “simplified filing portal” with the IRS populating certain fields from internal information (e.g., the reconciliation form). This reconciliation would be no more difficult than recertifying eligibility for public benefits once every 12 months.
- If eligible, the taxpayer can claim the remaining 3/12ths of the CTC on the tax return; if the return passes IRS refund filters, the taxpayer can expect to receive his full refund, including the remaining CTC, the EITC, and other credits by the last two weeks in February. (Recall that by law, the EITC and CTC cannot be paid out before February 15 of any given year so IRS can validate income.)
- Where there is income volatility, taxpayers would be able to elect to use prior year income instead of current year income when reconciling and claiming the CTC on the current year tax return.
- Where a taxpayer is eligible for the EITC and has an overpayment of monthly AdvCTC, the IRS would be prohibited from applying the EITC refund (or limited to 15% of the EITC refund) toward that overpayment.
- There would be an expeditious administrative dispute resolution process throughout the year, with immediate notice to the household currently receiving the AdvCTC, advising that a competing claim has been filed and reminding that household to update its circumstances if appropriate. The eligible new claimant will receive the monthly AdvCTC retroactively to when the qualifying child resided with the claimant for more than half the month (or the most days in the month). There would be a safe harbor of 3 months protecting the original claimant from clawback, and there would be reasonable cause exceptions (e.g., for sickness) and first-time abatement provisions for late change-of-circumstances updates, similar to penalty relief already available under the IRC and IRS procedures.
- To ease the administrative burden of both parties in situations of dueling claims, the IRS would be required to develop an affidavit form, similar to Forms 8836 and 8836 Schedule A which were developed and successfully tested in 2003 and 2004 to establish residency for purposes of the EITC.
- Judicial review by the Tax Court would be available when the income tax return is filed, reconciling the credit for the year. This allows for all benefits pertaining to family status and children to be determined in one proceeding.
- The combination of the 3-month safe harbor, paying only 9/12ths of the CTC in advance, the income look-back rule, the EITC collection prohibition, the reasonable cause and 1st time failure-to-update relief, routine nudges regarding the responsibility to update, and IRS systemic error and fraud detection will minimize clawbacks while protecting program integrity and reputation.
One final point: by applying the principal residence test of § 152 on a monthly basis, and amending the definition to establish principal residence by the greatest number of days in one household in a month, you may actually make more taxpayers eligible for the monthly AdvCTC than if you measured residence by the majority of days. And once you have established principal residence for a given month under this definition, you may make more taxpayers eligible for the EITC and all the other family benefits that require 6 months-and-a-day residency.
In my final blog for the week, I will discuss the Family and Worker Benefit Unit that Congress should require the IRS to establish in order to administer all the family status proposals fairly and as seamlessly as possible. In closing, I just want to reiterate that the above proposals are my attempt to think through very challenging questions of tax policy and administration. I welcome people’s thoughts and alternative recommendations. I have put my thoughts forward here because I believe we will all benefit from a full and transparent discussion of these complex issues.