“Defense to Repayment” Protects Taxpayers with Defaulted Student Loans from Treasury Offset Program

Professor Michelle Drumbl who teaches at Washington & Lee University School of Law and who runs the low income taxpayer clinic there brings us a guest post on an interesting case regarding the intersection of the earned income tax credit, defaulted student loan debt and the Treasury Offset Program. I am privileged to work at the Legal Services Center of Harvard Law School with a group of amazing lawyers who represent individuals who have attended for profit colleges and who have not received the bargained for benefits of higher education. As Professor Drumbl describes and as they posted, my colleagues recently won an important case protecting the tax refunds of students of a for profit college. I hope that this case, and others, will help create a movement to protect the earned income tax credit from the offset program. For those interested in this issue look for a deeper discussion in Professor Drumbl’s book which will be published next year by Cambridge University Press. Keith

Individuals expecting a tax refund are sometimes unpleasantly surprised to learn the refund instead has been applied to another outstanding debt. Internal Revenue Code section 6402(a) authorizes the Department of Treasury to offset an “overpayment” (generally speaking, the amount of refund shown as due on the return) against any outstanding federal tax, addition to tax, or interest owed by the taxpayer. If the taxpayer does not have any outstanding federal tax debts, or if any amount of refund remains after those debts are paid, section 6402(c)-(f) provide that the overpayment is then subject to the Treasury Offset Program in the following order of priority: 1) past-due child support payments; 2) outstanding debts to other federal agencies, including federal student loan debt; 3) outstanding state income tax debt; and 4) outstanding unemployment compensation debt owed to a state.

read more...

Keith blogged about section 6402 nearly three years ago in a post that drew a few dozen comments, including many from return filers who had experienced various types of refund offsets. In my view, the refund offset rules are troubling because they capture the refundable portion of the earned income tax credit (EITC) and the child tax credit. These two refundable credits constitute important social benefits for millions of working Americans. Administered by the IRS and delivered as part of the tax return filing process, these credits are a critical part of the U.S. social safety net.

In its 1986 decision Sorenson v. Secretary of the Treasury, the Supreme Court held that the refundable portion of the earned income tax credit is an “overpayment” for purposes of section 6402(c). Sorenson involved a challenge brought by a married taxpayer who filed a joint income tax return with her husband; their expected tax refund was offset and applied toward her husband’s past-due child support obligation. Mrs. Sorenson protested, and because they lived in Washington state, the IRS determined she was entitled to one-half of the joint refund under the state’s community property laws. But Mrs. Sorenson was unsatisfied with that outcome and filed suit in federal court, arguing that Congress did not intend for section 6402(c) to reach the earned income tax credit. On appeal to the Supreme Court, Mrs. Sorenson made a statutory interpretation argument and also argued that “permitting interception of an earned-income credit would frustrate Congress’ aims in providing the credit.” The Court rejected both of these arguments. While the Court acknowledged the “undeniably important” objectives of the EITC, it also noted that the “ordering of competing social policies is a quintessentially legislative function.” In particular, it is for Congress, not the Court, to decide whether the goals of the EITC outweigh the offset program’s goals of “securing child support from absent parents whenever possible.” After all – as the decision alludes – securing child support from absent parents also reduces the number of families on welfare (just as the EITC does). Following Sorenson, it is clear that it would be up to Congress to explicitly carve out the EITC from the definition of overpayment for purposes of section 6402. In the meantime, taxpayers subject to refund offsets will continue to lose this valuable social benefit which they otherwise are entitled to receive.

In my forthcoming book, Improving Tax Credits for the Working Poor, I argue that Congress should indeed consider protecting the EITC from offset, at least in part, and at least with respect to certain types of debts. I acknowledge, though, that Sorenson presented the most morally troublesome argument for protection from offset, because the underlying debt at issue was past-due child support. It is difficult to argue that a taxpayer should receive the EITC in support of a child who currently resides with him or her if the alternative is to divert the EITC to a child for whom the taxpayer is delinquent on child-support obligations.

In contrast, I highlight student loan defaulters as a relatively sympathetic case for which to carve out EITC offset protection. My proposals are inspired in part by informative National Consumer Law Center (NCLC) reports available here and here, from which I learned that approximately 1.3 million individuals in student loan default were subject to tax refund offsets in 2017. We do not know how many of those 1.3 million individuals were also EITC recipients, but surely there is some significant overlap between low-income working families and student loan defaulters. Among the most vulnerable student loan borrowers are those who borrow to attend for-profit institutions. As NCLC attorney Persis Yu describes in her March 2018 report, some of these borrowers are denied the promised benefits of their education when a fraudulent school closes in mid-course.

Thus, I was thrilled recently to learn that Keith’s colleagues Toby Merrill and Alec Harris are succeeding in some of their consumer protection efforts against the for-profit college industry. Toby and Alec, who have also blogged on the issue of refund offsets previously, work for the Project on Predatory Student Lending, which is part of the Legal Services Center of Harvard Law School. Among other cases, the Project on Predatory Student Lending has represented individuals who borrowed money to attend Corinthian College, a now-defunct for-profit company that operated post-secondary schools around the country, including a school called Everest Institute in Massachusetts.

The Massachusetts Attorney General’s office spent several years investigating Everest Institute for its deceptive recruiting and marketing practices. On March 25, 2016, Massachusetts Attorney General Maura Healey and U.S. Department of Education Secretary John B. King announced that students who were defrauded by Corinthian campuses nationwide (including the two Everest Institute campuses in Massachusetts) would be eligible for forgiveness of those federal loans. While this sweeping announcement was great news for borrowers, the details remained to be seen as to how and when this relief would apply.

Darnell Williams and Yessenia Taveras were among the thousands of students who had attended programs at Everest Institute. Williams and Taveras each took out federal student loans to pay for their program. Both individuals defaulted on their loans in the fall of 2014, before Healey’s office had completed its investigation of the Everest Institute campuses. Following their default, in August of 2015 the Department of Education sent Williams and Taveras the required notice of intent to turn the defaulted debt over to the Treasury Offset Program (TOP). Once the Department of Education certifies to the TOP that the debt meets certain requirements, the debt becomes subject to section 6402 offset procedures in the manner I describe above. Neither Williams nor Taveras individually filed an objection to the Department of Education notice within the prescribed 65-day deadline.

After the 65-day window to file an objection, but before the Department certified Williams’ and Taveras’ student loan debts to the TOP, Healey wrote to the Secretary of Education to request immediate and automatic discharge of all federal student loans borrowed to attend Everest Institute in Massachusetts (note that this November 2015 request of Healey’s also predates the aforementioned joint announcement with the Department of Education). Healey referred to her written request as a “defense to repayment” application on behalf of the student borrowers. Healey’s defense to repayment application included, among other exhibits, a list of names of more than 7,000 student borrowers who had attended Everest Institute, including Williams and Taveras. The Department of Education nonetheless certified the Williams’ and Taveras’ debts for collection by the TOP without deciding on the merits of Healey’s letter.

The following spring, in April and May of 2016, Williams and Taveras each filed income tax returns showing refunds due. Because the Secretary of Education had certified the defaulted debts to the TOP, the taxpayers’ refunds were offset against their outstanding loans. The amounts they lost were significant: Williams’ offset was in the amount of $1,263, and Taveras’ offset was in the amount of $4,999.

At issue in Williams v. Devos is whether Attorney General Healey successfully raised a borrower defense proceeding on behalf of the thousands of individuals listed in her exhibit, including Williams and Taveras. The Project on Predatory Lending represented Williams and Taveras in the matter in federal court, arguing that the Secretary of Education improperly certified their student loan debts as legally enforceable for purposes of the TOP program.

Last month, the judge in this case ruled that Healey’s November 2015 submission did invoke a borrower defense proceeding as to Williams and Taveras, and that the Secretary’s certification of the debt to the TOP without consideration of Healey’s submission was arbitrary and capricious. The court order vacated the certifications for refund offset for Williams and Taveras and remanded the matter to the Department of Education for a consideration of the borrower defense asserted by Healey.

Congratulations to Toby, Alec, and all at the Project on Predatory Student Lending on this ruling in Williams v. Devos. This is a significant victory for student borrowers challenging the validity of their loans. Though not strictly speaking a tax case, this development has important collateral consequences for low-income taxpayers who are eligible for refundable credits. As Keith has written recently in the offer in compromise context, and as I contemplate in my book, the fact that the EITC is an anti-poverty supplement for working families provides a compelling argument to protect it from offset, at least in certain circumstances. While I would like to see Congress act to at least partially exempt the EITC from offset against any federal student loan default, this ruling is an important and tangible step forward, as it is precedent for protecting student loan defaulters from tax refund offset while a borrower defense proceeding is pending.

 

The Automated Substitute for Return Procedures

Today we welcome guest blogger Michelle Drumbl.  Professor Drumbl teaches tax at Washington and Lee University School of Law and runs the low income taxpayer clinic there.  She thanks clinic student Hollie Floberg for her assistance in writing this post.  One of Professor Drumbl’s suggestions concerns the use of the substitute for return procedures to pursue returns with little collection potential.  I wrote about the IRS policy of taking collection into account in the TFRP context.  That policy may have some play here.  Keith

The automated substitute for return (ASFR) procedure, authorized by section 6020(b) of the Internal Revenue Code, provides the Internal Revenue Service with a mechanism by which to use third-party information reporting to assess a tax liability for nonfilers. This enforcement mechanism is a relatively easy way for the Service to narrow the so-called “tax gap,” defined as “the amount of true tax liability faced by taxpayers that is not paid on time.” In the context of closing the tax gap, income that is subject to information reporting is low-hanging fruit. Unsurprisingly, studies show that taxpayer compliance rates are higher when income is subject to “substantial information reporting” (and highest if subject to both information reporting and withholding).

read more...

It is hard to refute that noncompliance undermines the tax system, or to deny the appropriateness of the Service pursuing assessment of unreported income it can easily verify. However, in her recently released 2015 Annual Report to Congress, National Taxpayer Advocate (NTA) Nina Olson identifies the selection criteria for cases in the ASFR program as Most Serious Problem #17. Olson insightfully identifies a number of ways in which the ASFR process, while easy for the Service, is imperfect for the taxpayer. Olson’s critique of the current ASFR selection criteria includes three points: 1) the IRS has a poor collection rate on these assessments; 2) the IRS has a high abatement rate on amounts assessed through ASFR; and 3) the ASFR program has a low return on investment.

I will address some of these points and the NTA’s recommended solutions from my perspective of a practitioner representing low-income taxpayers, and add some observations of my own. 

When Using its Third-Party Documentation, the IRS Should Take into Account Deductions and Credits, Not Just Income

In describing the abatement rate on ASFR assessments, Olson notes that the program could increase efficiency and return on investment if, as part of its case selection, it more carefully considered third-party documentation consistent with possible deductions and credits. She notes that selection of cases likely to result in abatement causes “rework for the IRS and potential harm for the taxpayer”. She suggests that the IRS refine its selection process by “considering third-party documentation that supports deductions or credits when determining which cases to select for the program”.

Olson’s point is well taken. The ASFR assessment process takes into account all income reported as earned by the taxpayer, but it ignores reported items that would reduce taxable income. For example, mortgage interest is reported to the IRS on Form 1098, state income tax withholding is reported on Form W-2, and student loan interest is reported on Form 1098-E. All of these reported payments potentially reduce a taxpayer’s liability.

I agree with Olson that the failure of the IRS to use or acknowledge this readily accessible information is inefficient, but the burden must remain on the taxpayer to claim these deductions. These deductions cannot be accurately calculated by the IRS during the ASFR process because the IRS has imperfect information. For example, the student loan interest deduction phases out at a certain income level, and it is possible that the taxpayer has additional income not subject to information reporting. Thus, it would be premature to propose the deduction based only on the income information reported by third parties. Accordingly, it is not unreasonable for the IRS to omit these deductions in calculating its proposed ASFR assessment. After all, the taxpayer sat on his or her rights by not filing a return; appropriately, the burden of filing and claiming deductions or credits rests with the taxpayer. But as I describe in my recommendation below, the IRS can and should communicate these possibilities to the taxpayer when known. Another issue the IRS should address more clearly in its communications is filing status.

Filing Status Presents Challenges for Taxpayers beyond What the NTA Report Describes

Olson is absolutely correct that filing status matters. Married taxpayers who file a joint income tax return are entitled to a number of potential benefits that are denied to married taxpayers who file separately. Most notably for low-income individuals, the earned income tax credit is not available to married taxpayers filing separately. But because joint filing is at the taxpayers’ election, an ASFR assessment must be based upon a filing status of single or married filing separately. Olson addresses this issue, but she does not describe the limitations that section 6013(b)(2) impose on married taxpayers who wish to file a joint return after filing separate returns.

Professor Kathryn Sedo has blogged here and here about one possible “procedural trap for unrepresented taxpayers” that arises under section 6013(b)(2)(B), namely that married taxpayers cannot amend from a separate return to a joint return once either spouse has filed a petition in Tax Court with respect to such tax year.

Another limitation, set forth in section 6013(b)(2)(A), is that if one of the spouses has filed a separate return, the taxpayers are prohibited from amending to file a joint return after the expiration of three years from the due date of the return for such tax year.

As I write about in a forthcoming article, in my low-income taxpayer clinic I commonly encounter married couples in which one spouse is a wage earner subject to information reporting and withholding and the other spouse receives nonemployee compensation subject to self-employment tax and no withholding. If a low-income couple with dependent children files a joint return, it may result that the personal exemptions, the standard deduction, the earned income credit, and the child tax credit combine to offset (in part or whole) the self-employment tax and the lack of withholding. However, a couple who lacks proper information and tax advice does not realize this; fearing a tax bill because of the self-employment tax and the absence of withholding, the taxpayers make a decision to file separate returns (or, quite commonly, a decision for one to file a separate return and one not to file).

Assuming the self-employed individual receives a Form 1099-MISC, the IRS will eventually discover the nonfiler and may pursue an assessment through the ASFR procedure. But this does not necessarily happen within three years of the due date of the return; as Olson describes, ASFR criteria require, among other things, that the module be “not older than five years prior to the current processing year.” But because of section 6013(b)(2)(A), the spouse who files a separate return has a limited window in which to amend and file jointly. Thus the nonfiler who comes into compliance more than three years late (perhaps in response to a proposed ASFR assessment) will have no choice but to file as married filing separately for those years if his or her spouse filed a separate return.

The IRS should (but currently does not) advise a nonfiler receiving a proposed ASFR of the time limitations imposed by section 6013(b)(2)(A) for amending to file jointly.

My Recommendation: Advise Taxpayers More Clearly of their Rights and Options, and Do So Sooner

The IRS can and should be much clearer in its communications with the taxpayer during the ASFR assessment process. My recommendation would be for the IRS to contact nonfilers within one year of the missed deadline and to include a letter explaining their rights and responsibilities more clearly. Currently, the 30-day proposed assessment letter lists the details of income reported by others. This letter should include all information received from third parties, not just income reported; it should notify (or remind, as the case may be) the taxpayer of any information reporting it received relating to possible deductions or credits. The letter could state clearly that the reported information may or may not be relevant in determining certain deductions or credits, note explicitly that the IRS has not included any deductions or credits in its calculation of the proposed assessment, and remind the taxpayer that it is his or her responsibility to affirmatively claim any deductions or credits on the return.

Moreover, the 30-day letter should be used as a reminder that the taxpayer may be eligible to use a more favorable filing status and is entitled to claim any qualifying dependents, either of which may reduce the proposed assessment. Specifically, it should notify the taxpayer that if he or she is married, there remains the option to file jointly with a spouse; it should further provide the deadline by which the taxpayer would need to file if the taxpayer’s spouse previously filed a separate return and wishes to amend and file a joint return.

Finally, the notice should recommend that the taxpayer visit a low-income taxpayer clinic if income eligible. 

In Conclusion: A Word About Those Low Collection Rates

Olson notes that the ASFR program has poor collection rates, citing statistics that the Service collected less than one-third of the amount assessed through ASFR in fiscal years 2011-2014. She suggests that the poor collection results are evidence of inefficient selection criteria, and notes that the ASFR program has a low return on investment relative to other IRS programs.

I feel strongly that the IRS should not simply choose to ignore nonfilers who represent a poor “return on investment” just because the program has “low collection rates”. Nonfilers degrade the tax system. They undermine the faith of the general public in a fair system, and they should not be given a pass because the relative dollar amounts are small or because they likely cannot afford to pay the liability. Olson speaks of the right to a fair and just tax system, but this concept cuts both ways. Compliant taxpayers expect the IRS to enforce filing requirements in an even-handed manner. Taxpayers who cannot afford to pay their tax liabilities have many avenues of relief available to them, including financial hardship status, a variety of installment agreement options, and the offer in compromise process. Taxpayers whose income exceeds the filing threshold must be expected to file a timely return and report their income, even if it is unlikely that the IRS will ever collect from them.

While the ASFR process can be modified in ways that would make it a more taxpayer-friendly (and perhaps a more efficient) procedure, it should not use collection likelihood (or the lack thereof) as part of its selection criteria.