In 2019 the nonprofit Urban Land Conservancy held a series of public meetings to help shape its plans for Viña, a proposed affordable rental complex in North Denver’s Elyria-Swansea neighborhood.
The community engagement process confirmed the importance of the project in the face of billions of dollars of public investment in the neighborhood that threatened to send property values and rents soaring and force out many low-income Latinx residents. It also highlighted the urgent need for homes that extremely low-income (ELI) households could afford.
“A lot of what folks in the community were saying was, we’ve got to have units that are [affordable] at 30 percent of area median income and below,” says Aaron Miropol, ULC’s president and CEO. “A lot of families in the community make . . . 30 percent or, to be frank, even below 20 percent of AMI.”
To accommodate those families while ensuring the project’s financial stability, ULC needed to find a way to offset the very low rents the families could pay. The project was being financed with Low Income Housing Tax Credits (LIHTC), and traditionally, the only way LIHTC units have been able to house ELI households is when those ELI households also have a housing voucher—but there aren’t enough of those to go around and they are not available to undocumented people.
ULC wanted instead to make up the difference by setting substantially higher rents for other tenants who could afford them, and cross-subsidizing. Typically this hasn’t worked for LIHTC units because the upper income limits for covered units was 60 percent of AMI, not leaving enough flexibility to support the low-enough rents.
To thread that financial needle, however, ULC took advantage of income averaging, an alternative LIHTC option authorized by the 2017 tax reform law. The rule lets landlords accept some tenants who earn up to 80 percent of AMI, as long as all the households together average out to no more than 60 percent of AMI.
Along with enabling ULC to offer rents affordable to extremely low-income community members without vouchers, income averaging can help projects qualify for additional funding streams, such as loans the city of Denver offers to developers of deeply affordable housing.
Income averaging projects have other advantages as well: they create affordable units for moderate-income residents who wouldn’t otherwise qualify for assistance, and provide social, health, and educational benefits for residents.
Mixed-income housing “creates extremely strong communities,” says Diana Stoian of Columbia Ventures, ULC’s development partner on the Viña project. “In order to fully achieve equitable development and serve multiple bands of income, income averaging is the solution.”
Filling a Need
When the federal Tax Cuts and Jobs Act passed in 2017 it was criticized for broadly favoring the wealthy. Among other effects, the resulting lower tax rates for corporations threatened to reduce incentives for investing in housing tax credits. At the same time, advocates were pleasantly stunned by the last-minute addition of what the IRS calls the “average income test” option for LIHTC-backed projects, says Jen Brewerton, vice president of compliance for Dominium, a large Minnesota-based affordable housing developer and operator.
“We were all shocked, absolutely shocked, when it did get released,” she says.
Housing advocates had lobbied for federal income averaging for years. It’s particularly important in rural areas, where low AMI brackets and sparse populations can make it difficult to find enough qualifying prospective tenants to allow the financing and construction of LIHTC complexes.
At the same time, some urban areas have more affordable units in the 50-to-60 percent of AMI range than they can fill and simultaneous shortages of homes for ELI and moderate-income households, says Peter Levavi, executive vice president at Brinshore Development, a large affordable housing developer based in Chicago.
“We have been building for this unicorn of people between 50 and 60 percent for 30 years,” he says. “The program, when it was originally designed, was meeting a need that was enormous. That need was filled two decades ago. So the program doesn’t work well for people who make less than 50 percent AMI and it certainly doesn’t work at all for those who make more than 60 percent AMI.”
Brewerton says Dominium quickly embraced the new test, making income averaging its default choice for LIHTC projects. It published explanatory guides, worked with housing agencies to develop state-level regulations, and helped tax-credit investors build their comfort level in the new option. One of its first projects converted former public housing units in Georgia to project-based Section 8 through the Rental Assistance Demonstration program. The company used income averaging to avoid displacing long-time residents whose incomes had risen above the old limit of 60 percent AMI, Brewerton said.
In some places, Dominium found that even with income averaging, most tenants still ended up being in the 60 percent AMI bracket because the local market did not support higher rents, she said. To pull in more tenants who were closer to the 80 percent AMI limit, such as police officers and teachers, the company built an affordable complex in a desirable section of downtown Minneapolis. Identical two-bedroom units rent for $885, $1,357, or $1,830 depending on the household’s income.
“It’s this beautiful community that looks no different from the communities around it. It’s across the street from the convention center there in downtown Minneapolis. It’s just absolutely stunning. And now it’s serving people at 40 percent AMI, 60 percent AMI, and 80 percent AMI, all living together . . . able to work and live downtown,” Brewerton said.
Dominium has built or renovated seven developments using income averaging, but in the last two years it has also had about 15 others that were canceled or changed to a different tax credit option, she said. That’s because in October 2020 the IRS published a proposed rule on income averaging that appears to make it troublingly easy for a development to suddenly fall out of compliance with the tenant AMI requirements.
If that happens, the development loses all its tax credits for the year and possibly for the full 10 years of credit. The investors who acquired the credits suffer a major loss in expected tax reduction, potentially in the millions of dollars. After the rule’s release, investors’ interest in income averaging plummeted and such projects became extremely difficult to finance.
Income Averaging’s Elevated Risk
Brewerton believes the IRS did not understand or was hostile to the idea that the federal government should expand LIHTC to assist moderate-income households. “The IRS came out with this whopping draft rule that scared the bejesus out of everybody,” she says. “Based on everything that we’ve seen, they feel like there’s these big bad developers that are only going to lease their 80 percent units and charge 80 percent rents, and just do 100 percent of the property at 80 percent,” though that wouldn’t have actually been possible under the averaging provision.
“Unfortunately, the tax credit program is run out of Treasury. It’s not run out of HUD,” Levavi says. “It’s not run by people who understand housing or care about housing. They care about taxes. They created a series of rules that made it very, very difficult for investors to monitor and do compliance to make sure that the tax credits could flow.”
Because of the perceived risk, the few investors who still agree to income-averaging deals end up paying less for the tax credits than they would have otherwise, reducing the amount of funding available to developers, he says.
One problem with the IRS proposal is that it conflicts with other federal policies by not allowing a unit’s AMI designation to change. Brewerton gave the example of a third-floor, 80 percent AMI tenant who becomes disabled and, under federal rules, must be offered a more accessible first-floor unit if one is available. If the first-floor apartment is a 30 percent AMI unit, the IRS says the landlord cannot just swap the unit designations to accommodate the disabled tenant; as a result, if the person does make the move, the building is suddenly out of compliance and could lose all of its tax credits.
Another big issue is the proposal’s unprecedented strictness about the impact of individual units going out of compliance, compared to the rules for other LIHTC options.
For example, one of the older LIHTC tests requires that at least 40 percent of a building’s units must have qualifying tenants. This is called the minimum set-aside. To maximize the tax credit, the percentage of qualifying tenants may actually be much higher; many LIHTC buildings are 100 percent qualifying tenants, but the program only requires that 40 percent of them be. In that scenario, if at some point some tenant incomes start exceeding the limit, the developer can simply stop claiming credits for those units. The project loses a few hundred dollars worth of credit, but as long as it satisfies the minimum set-aside, most of the millions in tax benefit remain.
Under the IRS draft rule, however, the 40 percent minimum no longer applies to income-averaging properties. A developer cannot just stop claiming credit for a few units in order to rebalance the AMI average. If incomes go up and the building average even slightly exceeds 60 percent of AMI, the development can fall off a “cliff,” going fully out of compliance and losing all its tax credits.
As a result, investors may require developers to maintain lower tenant AMI averages in an effort to lessen the chance of non-compliance, Levavi says. That reduces the amount of rent that buildings bring in to cross-subsidize rents and cover expenses.
The Biden administration has promised that the IRS will issue a revised final rule by the end of this month that will address the problems with the 2020 proposal. Brewerton said it is essential that the government fix its implementation of income averaging if the option is ever to fulfill its potential in the long run.
“If any part of that draft rule that came out becomes a final rule, then average income will go away—will completely go away,” she says. “No one’s going to want to do it.”