LIHTC set-aside elections: 20-50, 40-60 and the average income test

The minimum set-aside is the floor a project must clear to be a “qualified low-income housing project” at all (IRC 42(g)(1)). It’s distinct from the applicable fraction, which determines how much credit the project earns. A project can meet the 40-60 floor with 40% low-income units, but it then earns credit on only 40% of its basis, so almost every LIHTC project is 100% low-income and the set-aside matters mostly for what income levels it allows.

The three tests

Test	Units required	Income and rent level	Notes
20-50	20% or more of residential units	Rent-restricted and occupied by households at or below 50% of AMI	Rarely used alone today
40-60	40% or more	At or below 60% of AMI	The default; not available in NYC
25-60 (NYC only)	25% or more	At or below 60% of AMI	Replaces 40-60 for projects in New York City under IRC 142(d)(6)
Average income	40% or more (25% in NYC)	Each unit designated at 20, 30, 40, 50, 60, 70 or 80% of AMI; the average of designations is 60% or less	Available for elections made after March 23, 2018

Source for all rows: IRC 42(g)(1) and the Form 8609 instructions (Rev. December 2025), line 10c.

In rural projects, as defined by section 520 of the Housing Act of 1949, income limits are measured against the greater of area median income or the national non-metropolitan median (42(i)(8)), for determinations made after July 30, 2008.

Choosing one

If your goal is…	Likely election
Simplicity, every unit at 60%	40-60
Higher rents on some units in a market where 70% to 80% AMI rents are below market	Average income
Deeper targeting paid for by higher-income units, or serving existing over-60% tenants in a preservation deal	Average income
A bond deal where the bond’s 142(d) test and the LIHTC test should match	The same test for both elections
A project with a few market-rate units and every low-income unit at 50%	20-50 (rare; most QAPs push deeper anyway)

The trade-off with averaging is administrative. Each unit has a fixed designation, the average must hold every year, and the next available unit rule has its own averaging version. Test a mix with the income averaging tool and read how the 60% average works in practice.

The average income test rules, now final

The IRS issued final average income test regulations in two steps:

1. T.D. 9967 (87 FR 61489, October 12, 2022) adopted 26 CFR 1.42-19: the project needs a “qualified group of units” making up at least 40% of residential units whose designated limits average no more than 60% of AMGI. If a unit fails, the owner can identify a different qualified group, which ended the “cliff” where one bad unit could sink the whole project. Applies to taxable years beginning after December 31, 2022.
2. T.D. 10036 (90 FR 46756, September 30, 2025) finalized the procedures that had been temporary regulations. They apply to taxable years beginning on or after September 30, 2025.

What the procedures require, from 1.42-19(c) and (d):

• Designate before occupancy. A unit’s imputed income limitation must be designated before the unit is first occupied as a low-income unit.
• 10-point steps only. Valid designations are 20, 30, 40, 50, 60, 70 or 80%.
• Record and report. Designations and each year’s identification of the qualified group (for the set-aside and for the applicable fraction) go in the owner’s books and records, kept for the 1.42-5(b)(2) retention period, and are communicated to the agency in the time and manner it sets.
• Changes only in listed cases: changes permitted in IRS guidance, agency-published policies applying to all averaging projects, changes required to enhance protections under the ADA, Fair Housing Act, VAWA, Rehabilitation Act and similar laws, tenant moves within the project (designations swap), and changes to restore compliance by designating a market unit or lowering other designations.
• Corrections. If the owner discovers a procedural failure, it has 180 days to submit a correction; if the agency discovers it, the agency’s correction period applies; the agency may waive a failure in writing.

How income limits for each designation are set

Each designation’s income limit is a multiple of HUD’s very low-income (50%) limit for the household size, per Rev. Rul. 2020-4: 20% is 0.4 times the 50% limit, 60% is 1.2 times, 80% is 1.6 times. Rents follow: 30% of the designated limit for the imputed household size. County figures are on the income limits pages.

Deep rent skewing

The deep rent skewed election (142(d)(4)(B); Form 8609 line 10d) is for projects that commit at least 15% of low-income units to households at or below 40% of AMI at all times during the compliance period, and keep market-unit rents at least twice the low-income rents. Its benefit is on over-income tenants: a continuing resident’s income may rise to 170% of the applicable limit, instead of 140%, before the next available unit rule applies (42(g)(2)(D)(iv)). It’s useful for mixed-income projects with long-term residents whose incomes rise.

Practical rules that catch people

1. Meet the set-aside by year-end of the first credit year. Lease-up schedules for 9% deals sometimes push this to the edge.
2. Multi-building projects. The set-aside is tested on the project. Each building is its own project unless the Form 8609 line 8b statement identifies the buildings in a multiple building project before the close of the first calendar year of the project period (8609 instructions). Missing the attachment means building-by-building testing.
3. Bond deals have a parallel test. The 142(d) set-aside for the bonds has its own qualified project period. See tax-exempt bond financing.
4. A failed set-aside means full recapture of the accelerated portion, not a partial one. See recapture.
5. State rules sit on top. Many QAPs require deeper targeting than the federal set-aside, or limit how many 80% units an averaging deal may have.