New U.S. Low-Income Housing Tax Credit Regulations on Average Income Set-Aside Finally Eliminate “Cliff Risk,” Clarify Compliance

After seven years, three rulemakings, and multiple rounds of public comment, the U.S. Department of the Treasury (Treasury) and the Internal Revenue Service (IRS) have at last delivered a durable framework for the average income set-aside. On September 30, 2025, the agencies published final regulations (the Final Regulations) addressing the average income set-aside for purposes of the low-income housing tax credit under Section 42 of the Internal Revenue Code of 1986, as amended (Code).¹ The Final Regulations replace the temporary and proposed rules published on October 12, 2022 (collectively with certain final regulations published at the same time, the 2022 Regulations) and are generally effective for taxable years beginning on or after the date of publication, though taxpayers may apply the Final Regulations to earlier years if they adopt the Final Regulations in full and on a consistent basis.

To their credit, Treasury and the IRS have been responsive to industry concerns throughout the rulemaking process. Proposed regulations published in 2020 (the 2020 Proposed Regulations) gave rise to the now-familiar “cliff risk” but offered only narrow and temporary mitigations. The 2022 Regulations introduced the “qualified group” concept and represented an important step forward, but as we noted in a prior alert, they did not fully address how projects could make sure they would not face disqualification in the face of isolated compliance failures. The Final Regulations now provide the missing piece: a clear mechanism for taxpayers to unilaterally correct noncompliance that they discover after the fact. The Final Regulations also provide improved reporting and correction rules that better integrate with established Section 42 compliance frameworks.

I. Background

Section 42 requires a residential rental project to satisfy one of three minimum set-aside tests to qualify as a “qualified low-income project”: the “20-50” test, the “40-60” test, or, beginning in 2018, the average income test. The average income set-aside requires at least 40% of the project’s units (25% in New York City) to be rent-restricted and occupied by tenants whose income does not exceed a taxpayer-designated imputed income limitation of 20%, 30%, 40%, 50%, 60%, 70%, or 80% of area median gross income (AMI). The average of all such designations may not exceed 60% of AMI.

The average income set-aside promised greater flexibility by permitting projects to serve a broader range of households across the income spectrum. But as interpreted by the 2020 Proposed Regulations, it also introduced a risk not present with the other two set-aside tests: If even a single unit was discovered after the fact to have been noncompliant, the entire project could fail the set-aside test if the average of the remaining compliant units’ designated income limitations exceeded the 60% AMI threshold, even if an alternative mix of units would have theoretically satisfied the requirements of the set-aside (albeit at a lower applicable fraction). This potential for wholesale disqualification became known as the “cliff risk.”

The 2020 Proposed Regulations acknowledged the cliff risk but relied on limited “mitigating actions” that were widely viewed as impractical. The 2022 Regulations sought to address these concerns. They adopted the concept of “qualified groups” — two taxpayer-designated sets of units that each had to meet the requirements of the average income set-aside but would be used for different purposes. One group, which could be constructed more conservatively, would be used just to test whether the project was considered to have satisfied the set-aside requirements. The other would be used to determine the project’s applicable fraction for credit calculation purposes. The purpose of this dual qualified group approach was to shift the burden of compliance determinations to taxpayers. The 2022 Regulations also granted to credit agencies broad authority to provide relief for noncompliance discovered after the fact. However, the 2022 Regulations still had three principal shortcomings:

1. Agency-Discretion Cliff. A taxpayer that discovered noncompliance in a reported qualified group after the fact had no unilateral ability to correct the error. The taxpayer could only preserve eligibility if the credit agency exercised discretionary relief authority. An agency could in theory deny relief even if an alternative mix of compliant units would have satisfied the statutory test, leaving taxpayers exposed to the same cliff-type outcomes.
2. Qualified Group Redundancy. The 2022 Regulations required taxpayers to designate both a “set-aside group” (to establish compliance with the minimum set-aside requirement) and a “fraction group” (to determine the project’s applicable fraction and thus the amount of credits the taxpayer could claim with respect to the project). The idea was that a taxpayer could designate a smaller set-aside group, thereby reducing the risk that unit-level noncompliance would cause the project to fail the minimum set-aside test. In practice, however, both groups were required to consist entirely of compliant low-income units in order for credits to be claimed, so a failure in the fraction group arguably still resulted in the loss of all credits for the year even if the set-aside group was fully compliant. Thus, the only real benefit of the two-group structure lay in the first year of the credit period: If the set-aside group was compliant as of the end of that year, the project would not be precluded from claiming credits in subsequent years in which the noncompliance was corrected. For years after the first year of the credit period, during which noncompliance might jeopardize credits for that year, but not in the remaining years of the credit period, the approach of using dual set-aside groups seemed to serve little or no purpose.
3. Delegated but Undefined Compliance. The 2022 Regulations assigned key recordkeeping and reporting responsibilities to state agencies but provided little guidance on timing, format, or cure procedures. Many agencies never issued clear policies or procedures for curing noncompliance, leaving taxpayers uncertain about how the rules would be implemented in practice.

II. The Final Regulations

The Final Regulations respond directly to the shortcomings of the 2022 Regulations. They retain the qualified group framework but provide mechanisms for taxpayers — not just agencies — to cure noncompliance. In doing so, they finally eliminate the agency-discretion cliff and clarify how compliance is to be demonstrated and corrected.

1. Corrected Qualified Groups. The most significant change in the Final Regulations is that taxpayers may now unilaterally submit a corrected qualified group when a previously identified group is discovered to contain one or more noncompliant units. This addresses the agency-discretion cliff by ensuring that a project will not be disqualified where an alternative mix of compliant units continues to satisfy the statutory average income test. To rely on this relief, the taxpayer’s corrective action must be timely: If the taxpayer discovers the failure, a corrected submission must be made within 180 days from the date of discovery. If the agency discovers it, the taxpayer must cure within the correction period provided in Treasury Regulation §1.42-5 (generally 90 days from agency notice, extendable up to six months). Agencies retain authority to waive failures in writing, but the taxpayer is no longer wholly dependent on the agencies’ granting discretionary relief.
2. Streamlined Reporting. The Final Regulations also partially address the redundancy of the two-group structure. Agencies may now permit taxpayers to report compliance using a single list of units sufficient to demonstrate both the minimum set-aside and the applicable fraction, although they retain discretion to require two lists if they choose. The rules include examples illustrating both approaches as well as an option for “differences only” reporting in subsequent years. This flexibility gives agencies the opportunity avoid redundancy introduced by the dual qualified group approach.
3. Integration With Existing Compliance Framework. The Final Regulations integrate average income reporting and cure procedures into the existing Treasury Regulation §1.42-5 monitoring regime. Taxpayers must contemporaneously record unit designations and qualified group identifications in their books and records, retain those records for the prescribed period, and communicate annual identifications to the agency in the manner the agency prescribes. This alignment with established compliance rules reduces uncertainty and ensures uniform application across agencies.

III. Example

Consider a 100-unit project in which 50 units are designated at 40% of AMI and 50 units are designated at 80% of AMI. The average of the designated income limitations is exactly 60%. If, however, one of the 40% units is later discovered to be out of compliance, the average of the remaining 99 designated units would rise above 60%, and the project would fail the test if evaluated solely on the basis of the remaining 99 compliant units. Under the Final Regulations, the taxpayer may restore compliance by submitting a revised qualified group that excludes both the noncompliant 40% unit and one compliant 80% unit. The revised group, consisting of 49 units at 40% and 49 units at 80%, for a total of 98 units, again averages to 60% of AMI and permits the taxpayer to satisfy the average income set-aside. The taxpayer would, however, need to recompute its applicable fraction on the basis of the 98 units in the revised set-aside group, resulting in a loss of credits attributable to both the noncompliant 40% unit and the compliant 80% unit that had to be removed from the set-aside group.²

A. If the taxpayer discovers the issue

Upon discovery, the 180-day correction clock begins. The taxpayer must take the following steps:

1. Identify and document the corrected group. The taxpayer identifies a revised qualified group. As noted above, in this fact pattern, removing the noncompliant 40% unit and one compliant 80% unit yields a 98-unit group averaging 60%.
2. Record the identification in books and records. The taxpayer records the corrected group in its books and records and retains that record for no less than the retention period required by Treasury Regulation §1.42-5(b)(2).³ This contemporaneous record is required for both the average income set-aside and the applicable-fraction determination.
3. Communicate the correction to the agency in the agency-prescribed manner. The taxpayer submits the revised group within 180 days of discovery in the manner prescribed by the agency.

Assume that one of the units designated as an 80% AMI unit happens to have been occupied by a tenant that would have qualified for a 40% AMI unit and that the unit is operated in a manner that would comply with the requirements of a 40% AMI unit. Could the taxpayer “fix” the issue by redesignating that unit as a 40% (rather than 80%) AMI unit, thereby preserving a qualified group that consists of all 99 compliant units? The preamble to the Final Regulations makes clear that the answer is no. The taxpayer may not “fix” the average by changing any unit’s imputed income designation retroactively after year-end; the permitted correction operates solely by changing which units comprise the qualified group, not by modifying their designations.

B. If the agency discovers the issue

When an agency identifies the noncompliance, the Final Regulations align the correction process with Treasury Regulation §1.42-5:

¹Treasury Regulation §1.42-5(b)(2) generally imposes a retention period of at least six years after the due date (with extensions) for filing the taxpayer’s tax return. However, records for the first year of the credit period must be retained for at least six years following the due date (with extensions) for filing the taxpayer’s tax return for the last year of the compliance period.
² Tenants in the removed compliant 80% unit would not need to vacate the unit, but the unit would no longer generate tax credits. 
³ Treasury Regulation §1.42-5(b)(2) generally imposes a retention period of at least six years after the due date (with extensions) for filing the taxpayer’s tax return. However, records for the first year of the credit period must be retained for at least six years following the due date (with extensions) for filing the taxpayer’s tax return for the last year of the compliance period.