JC Craig of JC Craig Consulting submits this comment concerning REG-119882-25, RIN 1545-BS06, Docket No. IRS-2026-1057, Application of the Personal Responsibility and Work Opportunity Reconciliation Act of 1996 (PRWORA) to the Refunded Portion of Certain Federal Refundable Tax Credits.

Interest of the commenter

I have volunteered with the IRS Volunteer Income Tax Assistance (VITA) program for 25 years and prepared returns for thousands of low- and moderate-income taxpayers, including many mixed-status families. I have also developed and led training for volunteer preparers, providing technical assistance to organizations offering VITA service nationwide. I have served as a volunteer liaison between VITA/TCE practitioners and the IRS through the Security Summit and as a member of the IRS Electronic Tax Administration Advisory Committee (ETAAC) from 2017 to 2019. These comments reflect my experience as both a preparer and a leader in this field.

The proposed regulations should be withdrawn

These proposed regulations are poor tax administration. They would direct the IRS’s limited compliance, taxpayer-service, and legal resources toward an eligibility question with no documented history of erroneous claims or fraud, while leaving unaddressed the two problems the agency’s own data identifies as the actual drivers of improper payments on the refunded portion of these credits, discussed below. I urge the Department of the Treasury and the IRS to withdraw proposed §§ 1.23-2, 1.24-3, 1.25A-7, and 1.32-4 in their entirety.

As proposed, these regulations would apply PRWORA’s “qualified alien” test to the refunded portion of the adoption credit, the Child Tax Credit (through the Additional Child Tax Credit), the American Opportunity Tax Credit, and the Earned Income Tax Credit. Treasury’s own regulatory analysis estimates that, of the 24 million taxpayers expected to claim the refunded portion of these credits for tax year 2026, between 200,000 and 700,000 would become ineligible under the proposed rule, disallowing between $0.7 billion and $2.6 billion in refundable credits.

Whatever one thinks of the underlying immigration policy, a change of this magnitude should not rest on reinterpreting a 1996 welfare statute that Treasury and the IRS did not previously believe applied to the refunded portion of these credits. As the preamble itself acknowledges, “[p]rior to 2018, the Treasury Department and the IRS had not viewed tax benefits, including refundable credits, as constituting Federal public benefits under PRWORA.” That view changed only after the Office of Legal Counsel revisited the question in a 2020 opinion and reaffirmed it in a November 19, 2025 opinion, citing Loper Bright Enterprises v. Raimondo, 603 U.S. 369 (2024), as grounds to revisit the earlier reading. Loper Bright means a court reviewing this rule will not defer to Treasury’s interpretation merely because it is reasonable; the court must decide independently whether it is the single best reading of PRWORA. If this interpretation is genuinely the best reading of a statute enacted in 1996, it is difficult to explain why Treasury and the IRS held the contrary view for the next twenty-two years, and why it took two Office of Legal Counsel opinions over five years to arrive at it. If Congress intended the refunded portion of these four credits to be governed by PRWORA’s alien-eligibility categories, it has had roughly thirty years to say so in the Internal Revenue Code itself.

The proposal creates an administrability problem the current statute already avoids

Congress has already written an objective, verifiable eligibility test into three of the four affected credits, tied to whether the taxpayer (and, for the Child Tax Credit, the qualifying child) has a valid, work-eligible Social Security number:

• Section 32 requires an SSN issued by the Social Security Administration, not one issued solely to receive a federally funded benefit;
• Section 24(h)(7), applicable to taxable years beginning after December 31, 2024, requires a work-eligible SSN for the qualifying child and for the taxpayer (or at least one spouse on a joint return); and
• Section 25A requires a work-eligible SSN before the return’s due date.

Congress revisited the Child Tax Credit’s SSN requirement as recently as the 2024 filing-year statutory change and chose not to layer a separate PRWORA “qualified alien” determination on top of it. The proposed regulations would nonetheless impose exactly that additional test — one that turns not on a verifiable Social Security Administration data field, but on an immigration classification that does not appear on any IRS form and that preparers have no reliable way to confirm.

The adoption credit contains no SSN-based or immigration-status eligibility requirement; it requires only that the taxpayer report the adopted child’s name, age, and taxpayer identification number. Proposed § 1.23-2 would create the only immigration-status eligibility gate anywhere in section 23 — not by implementing a term Congress used, but by grafting PRWORA onto a credit Congress never conditioned on immigration status.

The proposed regulations compound this problem by creating a new timing rule solely for immigration status, layered onto credits whose existing eligibility tests already have their own timing rules — a work-eligible SSN by the return’s due date, for example. The preamble states that each taxpayer must be a U.S. citizen, U.S. national, or qualified alien “on the date of filing an initial or amended Federal income tax return first claiming the credit for the taxable year,” reasoning that “the date on which the taxpayer first claims the credit is the most appropriate date” because, in the preamble’s own words, “the nonrefundable and refundable portions of a single credit cannot be claimed at different times.” In practice, this locks in a taxpayer’s qualified-alien status as of whenever they first file a return claiming the credit at all — including the nonrefundable portion, which most taxpayers claim regardless of PRWORA status. A taxpayer who is not yet a qualified alien on that date has no apparent path to recover the refundable portion later by amending once their status changes, even within the ordinary three-year period to file a refund claim under 26 U.S.C. § 6511(a), because the preamble treats the two portions of the same credit as impossible to claim at different times. Eligibility for these credits is already fully established by existing law; immigration status, to the extent it is tested at all, should determine only whether an already-earned refund can be paid, based on the taxpayer’s status whenever they properly file a claim — not on whichever return happened to be the first to mention the credit. No other eligibility requirement for the refunded portion of these credits works this way, and administering a filing-date-dependent test will require new systems logic, forms, and preparer training devoted to a distinction the current statute does not draw. It also creates an incentive to delay filing altogether: a taxpayer expecting to become a qualified alien later may choose to wait rather than file on time and forfeit the refundable portion permanently — the opposite of the timely compliance the tax system depends on.

In my VITA experience, the portion of a credit that reduces tax liability and the portion that is refunded have always been calculated the same way, using the same eligibility test, on the same line of the same form. The proposed regulations would require preparers to apply two different eligibility standards to two slices of a single credit on a single return: the existing statutory SSN test for one slice, and an unverifiable, date-locked immigration classification for the other.

Identity theft and falsely claimed dependents — not immigration status — are the documented fraud risks in these credits

To the extent the proposed regulations are meant to protect the refunded portion of these credits from fraud, they target the wrong risk. The IRS’s own reporting identifies two dominant sources of improper claims: unintentional errors relating to the qualifying-child residency test, and intentional fraud involving stolen identities and falsely or duplicately claimed dependents. Neither is caused by, or meaningfully addressed by, a taxpayer’s immigration status.

On the unintentional-error side, Treasury’s own analysts attribute most EITC overclaims to the qualifying-child residency requirement — whether a child lived with the taxpayer for more than half the year — in part because the IRS receives no administrative data to verify where a child resided. In my experience, taxpayers responding to IRS disallowances struggle to find documentation showing a child lived with them for the required time, because children are generally not listed on leases or mortgages.

On the fraud side, the dominant and well-documented risk is identity theft: a return filed using a stolen Social Security number, or a child claimed by someone before the legitimate taxpayer files. A Treasury Inspector General for Tax Administration report issued in May 2026 found that the IRS stopped roughly $7 billion in fraudulent refunds tied to identity theft in calendar years 2024 and 2025 alone — out of 7.5 million returns flagged by the IRS’s identity-theft filters over that period. That same report found that a substantial share of the flagged returns, roughly half in each of the two most recent years reviewed, actually belonged to legitimate taxpayers whose refunds were delayed while they proved who they were. That is the real administrability trade-off this rulemaking should be weighed against: a screening system that already produces large numbers of false positives among legitimate filers, now paired with a second, harder-to-verify eligibility test layered on top of it.

The IRS already has a targeted response to the false-dependent problem, and it has nothing to do with immigration status. Beginning with the 2025 filing season, the IRS started accepting electronically filed returns claiming a dependent who was already claimed on another return, provided the legitimate taxpayer includes a valid Identity Protection PIN — resolving in minutes an issue that previously required paper filing and months of delay. Where no IP PIN is available, the IRS’s existing process requires the legitimate taxpayer to paper-file, wait for the agency to contact both filers (typically around two months after filing), and then substantiate the claim with documentation such as birth certificates and proof that the child lived at the taxpayer’s address for more than half the year. Immigration status plays no role anywhere in that process, because it has no bearing on whether a dependent claim is fraudulent.

Layering a PRWORA “qualified alien” determination onto this system closes neither documented gap: it does not help the IRS verify where a child lived, and it does not help the IRS or preparers detect a stolen identity or a duplicated dependent claim. It only adds a third, unconnected eligibility question that will consume examination and taxpayer-service resources without addressing either of the fraud and error patterns the agency has actually documented.

The required declaration carries felony exposure for an honest mistake

The proposed regulations would require every taxpayer claiming the refunded portion of one of these credits to sign, under penalty of perjury, a declaration that they are a U.S. citizen, U.S. national, or qualified alien eligible to receive the credit. The preamble is explicit about the stakes: it cites 26 U.S.C. § 7206, which makes willfully providing false information on a return a felony punishable by a fine of up to $100,000 and up to three years in prison; 18 U.S.C. § 1015(e), which separately makes a knowing false claim of citizenship or national status a felony; and 18 U.S.C. § 1001, which makes a knowing and willful false statement to the federal government a felony.

Asking taxpayers to certify their own PRWORA status under threat of felony prosecution assumes that status is simple to determine. It is not. The Office of Legal Counsel itself took two opinions and roughly five years to settle how PRWORA applies to these credits. VITA volunteers and paid preparers alike will now be asking taxpayers, many of whom have never heard the term “qualified alien” and cannot readily classify their own status against the eight categories in 8 U.S.C. § 1641(b), to attest to that classification under oath. A taxpayer who guesses wrong in good faith, or who is misled by someone else, has signed the same felony declaration a person who lied deliberately would have signed. The absence of a verification method cuts both ways: the same uncertainty that puts an honest taxpayer at risk of an unwitting false declaration gives a dishonest preparer or filer confidence that a false declaration will not be caught.

I recommend that the final rule state expressly that a good-faith, non-willful error in this declaration will not be treated as a violation of 26 U.S.C. § 7206, 18 U.S.C. § 1001, or 18 U.S.C. § 1015(e), and will not be referred for criminal investigation absent evidence that the taxpayer actually knew the declaration was false. Without that assurance, this requirement will not just deter accurate filing by eligible and ineligible taxpayers alike — it will generate additional correspondence, disputes, and taxpayer-service contacts for the IRS to resolve, on top of a verification burden the agency’s own data does not show is necessary.

The rule will cause eligible taxpayers to forgo refundable credits, without a workable way to catch the taxpayers it targets

I have repeatedly encountered taxpayers who conflate PRWORA benefit eligibility with the separate “public charge” inadmissibility test under INA § 212(a)(4). Even though the two are legally distinct, the confusion is real, and it changes behavior. Research on the 2019 public charge rule found that one in five adults in immigrant families with children (20.4 percent) reported avoiding public benefits altogether because of immigration-related fear or confusion — even though tax credits were not among the programs targeted by that rule. There is every reason to expect a comparable chilling effect here, this time affecting U.S. citizen children whose parents are uncertain about their own status or simply afraid of making a mistake.

The eligibility test also attaches to the wrong person. Under the proposed regulations, it is the filer’s own status — not the qualifying child’s — that determines whether the refunded portion of the Child Tax Credit is paid. A U.S. citizen child, whose own eligibility Congress addressed directly in section 24(h)(7), can lose access to that credit’s refundable value entirely because of a parent’s immigration classification that has nothing to do with the child’s own citizenship. Congress did not write the Child Tax Credit’s SSN requirement this way; it tested the child’s and the taxpayer’s own SSN status, not a separate immigration classification of the parent.

Paid preparers, meanwhile, face due-diligence obligations under 26 U.S.C. § 6695(g): a penalty, currently $650 per failure for 2026 returns (up to $2,600 if a single return involves all four credits), for failing to adequately determine a taxpayer’s eligibility for the EITC, CTC/ACTC/ODC, AOTC, or head-of-household status. Without a verifiable method to confirm “qualified alien” status, preparers are left to choose between declining to prepare returns for taxpayers they cannot definitively clear — denying eligible taxpayers a credit they are entitled to — or relying on the taxpayer’s own representation and risking a penalty if that representation later proves wrong through no fault of the preparer. As a VITA preparer, I rely on the taxpayer’s own representation for every other eligibility question on a return; there is no reason PRWORA status should be different, yet unlike every other eligibility fact a taxpayer represents, it now carries the felony exposure described above.

The regulation invites costly, resource-intensive litigation over an unresolved legal question

Even apart from the concerns above, finalizing this rule is likely to draw exactly the kind of prolonged, multi-front litigation that consumes the same scarce IRS and Treasury resources this comment describes elsewhere, without producing a lasting result. As discussed above, Loper Bright eliminated the deference courts previously gave agencies interpreting ambiguous statutes; a reviewing court must now decide independently whether Treasury’s reading of PRWORA is the single best reading of the statute, not merely a reasonable one. That is a harder standard to meet, particularly given that Treasury and the IRS took the opposite position for twenty-two years before 2018.

This rule also has the features that have drawn sustained legal challenge to other agency actions in recent years: a claim of new authority over a matter of substantial economic and political significance — hundreds of thousands of taxpayers, billions of dollars — based on reinterpreting a statute other than the one the agency ordinarily administers. The closest precedent is the 2019 public charge rule, which was challenged within weeks in federal courts across multiple states, produced conflicting rulings and preliminary injunctions, and was ultimately rescinded before the underlying legal question was resolved on the merits, after years of litigation cost to the government.

If this rule follows a similar path, the IRS should expect to defend it simultaneously in multiple courts while also building the implementation — new forms, systems changes, preparer guidance, and examination procedures — that a nationwide injunction could freeze or unwind at any point. That is a substantial commitment of the same limited legal, compliance, and administrative resources this comment describes elsewhere, spent on a rule that may never take lasting effect, in an area where the agency has not identified a documented compliance problem to justify the expenditure.

Recommendations

If the Department and the IRS do not withdraw the proposed regulations, I recommend, at minimum:

1. Remove the adoption credit (proposed § 1.23-2) from the rule’s scope, since section 23 contains no immigration-status eligibility test Congress enacted for the IRS to implement.
2. Do not create a separate timing rule for qualified-alien status. Eligibility for these credits is already fully established by the existing work-eligible SSN test; immigration status, if retained at all, should determine only whether the refundable portion can be paid, based on the taxpayer’s status whenever they properly claim it — including by filing an amended return within the normal refund limitations period under 26 U.S.C. § 6511(a) if their status changes after their original filing.
3. Publish the specific, auditable method the IRS and preparers are expected to use to verify “qualified alien” status, or state plainly that the rule will rely on taxpayer self-certification alone.
4. Provide education to taxpayers regarding specific statuses that are not eligible for refundable portions of tax credits, with clear directions on how to get more information. In that outreach and education, make clear that affected taxpayers remain eligible to use the credits to reduce tax liability.
5. Confirm, in the final rule, that VITA/TCE volunteer preparers and paid preparers satisfy their due-diligence obligations under 26 U.S.C. § 6695(g) by relying on a taxpayer’s valid, work-eligible SSN, without independently verifying immigration classification.
6. State expressly that a good-faith, non-willful error in the required declaration will not be treated as a violation of 26 U.S.C. § 7206, 18 U.S.C. § 1001, or 18 U.S.C. § 1015(e), and will not be referred for criminal investigation absent evidence of actual knowledge of ineligibility.
7. Provide a reasonable-cause cure period, rather than automatic disallowance or preparer penalty, when a good-faith declaration later proves incorrect.
8. Delay the applicability date beyond “taxable years ending on or after” the final rule’s publication, to give the IRS, tax software providers, and preparers at least one full filing season of lead time.
9. Before finalizing the rule, address how Treasury and the IRS intend to prevent the kind of chilling effect documented after the 2019 public charge rule, which caused eligible U.S. citizen family members to forgo benefits out of fear or confusion unrelated to their actual eligibility.
10. If Treasury and the IRS retain a fraud-prevention rationale for this rule, explain in the final rule how the proposed eligibility test improves on the identity-verification tools already in use, such as the IP PIN process for duplicate dependent claims, since immigration status is not a factor in either of the documented error and fraud patterns — qualifying-child residency and identity theft — that the IRS itself has identified as the primary drivers of improper payments on the refunded portion of these credits.
11. Consistent with the cost-benefit analysis Executive Orders 12866 and 13563 already require, publish an estimate that accounts for implementation, examination, taxpayer-service, and reasonably anticipated litigation-defense costs, and weigh that estimate against the absence of documented evidence that immigration status is a meaningful driver of improper payments on the refunded portion of these credits.

Conclusion

The refundable credits at issue already have workable, statutorily defined eligibility rules that VITA volunteers, paid preparers, and the IRS itself rely on every filing season. The proposed regulations would replace that clear standard with a second, harder-to-verify test grafted onto a 1996 statute that Treasury and the IRS did not previously apply to the refunded portion of these credits, for reasons that, by the preamble’s own account, trace to two recent OLC opinions rather than any change Congress made to the Internal Revenue Code. Every time Treasury reinterprets a non-Code statute to change tax administration in this way, it invites a future administration to do the same for unrelated purposes — for example, to revisit eligibility for corporate or high-income tax provisions through a similarly creative reading of some other federal statute, rather than through legislation. Doing so is not only legally vulnerable; it is poor use of the IRS’s limited resources at a moment when those resources are better spent on the credits’ documented sources of error and fraud. I respectfully urge the Department and the IRS to withdraw the proposed regulations or, at minimum, adopt the changes described above before finalizing them.

Respectfully submitted,

JC Craig

JC Craig Consulting

johncharlescraig@gmail.com