Affordable Housing Isn’t One Problem. We Need to Stop Treating It Like One.

Share this article!

We use the term “affordable housing” constantly in this industry.

I use it. WRE News uses it. Developers, elected officials, housing agencies and advocates all use it.

The problem is that two apartments can both legally and accurately be described as affordable housing while serving households with dramatically different financial circumstances.

A home restricted to someone earning 80% of area median income is affordable housing. So is one targeted to 60% of AMI. So is housing intended for a household at 30% of AMI.

Those are not the same housing problem.

New reporting on vacancies in income-restricted apartments makes that distinction difficult to ignore.

The Associated Press reported that Austin has more than 4,500 vacant apartments classified as affordable, representing a vacancy rate approaching 16%. In Denver, the vacancy rate was reported at 13% for Low-Income Housing Tax Credit units serving households at 60% of area median income and 21% for units at 80% of AMI. Portland reportedly has more than 1,700 vacant affordable apartments.

At first glance, those numbers don’t make much sense.

The United States has an enormous shortage of affordable rental housing. The National Low Income Housing Coalition’s 2026 Gap report estimates that 11 million extremely low-income renter households are competing for only 3.8 million homes that are both affordable and available to them. That leaves a shortage of 7.2 million homes.

For every 100 extremely low-income renter households in America, there are only 35 affordable and available homes.

So how can we simultaneously have an extraordinary shortage of affordable housing and thousands of affordable apartments sitting vacant?

Part of the answer is that we are using one phrase to describe several different markets.

The National Low Income Housing Coalition defines extremely low-income renter households as those with incomes at or below the federal poverty level or 30% of their area median income, whichever is greater. A large share of subsidized rental production, however, operates at higher income levels.

The Low-Income Housing Tax Credit is a good example.

LIHTC has been enormously important to affordable housing in this country. Created by the Tax Reform Act of 1986, the program has helped finance millions of affordable rental homes. HUD describes LIHTC as the most important resource for creating affordable housing in the United States today.

I don’t think the lesson from these vacancy numbers is that LIHTC doesn’t work.

The lesson is that we need to be more precise about what we expect it to accomplish.

LIHTC rents are generally structured around area-median-income thresholds. Depending on how a project is structured, units may be targeted to households at 50%, 60% or other qualifying percentages of AMI. The rent restriction is tied to the income level assigned to the unit, not individually recalculated so that every tenant pays exactly 30% of that particular household’s income.

That distinction matters enormously at the bottom of the income scale.

A household can qualify to live in an income-restricted apartment and still struggle to afford the rent.

According to the Associated Press investigation, only about 12% of affordable units financed through LIHTC in 2024 were set aside for extremely low-income households. Most served households at 50% of AMI or above.

There is nothing inherently wrong with that. A family earning 50% or 60% of area median income can absolutely struggle with housing costs and need an affordable place to live.

But helping that household does not solve the same problem as housing someone earning 25% or 30% of AMI.

Austin provides a striking example.

The city established a goal of producing 20,000 homes for extremely low-income households between 2018 and 2027. According to AP’s review of the city’s housing data, only 543 had been completed as of 2024.

At the same time, Austin had already completed the 15,000 units it targeted for households earning between 60% and 80% of area median income.

That doesn’t mean Austin built 15,000 homes it didn’t need.

It means producing housing at one level of affordability can be considerably easier than producing it at another.

And that brings us to the part of this conversation that I think gets lost too often.

Someone has to make the numbers work.

Affordable housing still has land costs. It still has construction costs. It still requires insurance. It still needs maintenance. It still has utilities, management expenses, financing and property operating costs.

Reducing the rent does not reduce most of those expenses.

An affordable-housing developer interviewed by the Associated Press offered a simple example. One of the developer’s apartments serving households at 60% of AMI generated $1,715 in monthly rent while mortgage and operating expenses totaled roughly $1,575.

That leaves about $140.

Cut the rent substantially enough to serve a household at a much lower income and the project doesn’t suddenly become more affordable to operate. It develops a funding gap.

Somebody has to fill it.

Austin’s own affordable-housing documents show how significant that gap can become. Deeply affordable projects frequently require local Housing Development Assistance funding on top of LIHTC and other financing sources. A group of projects identified by the city required or was expected to require more than $68 million in city gap financing.

This is why I don’t think pointing fingers at developers gets us very far.

Nor does blaming the tax-credit program.

If a project can financially support rents affordable to a household at 60% of AMI but cannot support rents affordable at 30% of AMI without another subsidy, we are dealing with two different economic problems.

We should start talking about them that way.

There is another wrinkle appearing in markets that have added substantial apartment supply.

In Portland, the city’s official 2026 affordability requirements set the maximum monthly rent for a one-bedroom apartment at 60% of median family income at $1,444, assuming the owner pays all utilities. AP reported an average market-rate one-bedroom rent of approximately $1,581.

That’s a meaningful difference, but it may not always be large enough to overcome the additional income verification and qualification requirements attached to an income-restricted apartment.

If a renter can obtain a market-rate apartment relatively quickly for a modest amount more, some will choose it.

Meanwhile, the renter who desperately needs an apartment hundreds of dollars below that level may still be unable to afford either one.

That should tell us something.

Vacancy in an affordable development does not necessarily mean there is no demand for affordable housing.

It may mean the housing isn’t aligned with the income level where the greatest unmet demand exists.

I think this distinction matters even more now because we are putting substantial additional resources and policy attention into housing production.

In August 2025, FHFA doubled the annual LIHTC investment limit for Fannie Mae and Freddie Mac from $1 billion to $2 billion each. Together, the enterprises can invest as much as $4 billion annually, with half reserved for difficult-to-serve LIHTC markets and at least 20% of that portion reserved for Duty to Serve rural communities.

More investment in housing is welcome.

But the number of units produced cannot be the only scorecard.

We should also be asking who those units are affordable to, where the shortage is most severe and whether the financing structure actually matches the households we are trying to serve.

At some income levels, the answer may be increasing supply and reducing the cost of building.

At others, it may require deeper tax credits, rental assistance, vouchers, local subsidy or some combination of them.

Zoning reform can make land easier to develop. Faster permitting can reduce carrying costs. Lower regulatory costs can help more projects pencil.

Those things matter, and I have argued for many of them.

But none of them changes the basic math of a household that simply does not earn enough to pay the rent required to operate the property.

That household needs a different solution.

This is why language matters.

When we announce that another 300 “affordable homes” are being built, the natural assumption is that we have reduced the affordable-housing shortage by 300 homes.

Maybe we have.

But if those homes serve households at 60% or 80% of area median income while the community’s greatest shortage is among households at or below 30%, we have solved one housing problem while another remains largely untouched.

Both problems deserve solutions.

They just aren’t the same solution.

The United States does not have one affordable-housing problem. We have workforce housing problems, moderate-income affordability problems, extremely low-income housing shortages, homelessness, supply constraints and markets where the cost of producing new housing exceeds what local incomes can support.

Calling all of it “affordable housing” may make the conversation easier.

It doesn’t make the problem easier to solve.

If we want to know whether billions of dollars in housing investment are working, we need to get more specific about what success actually means.

Not simply how many affordable units we built.

Who can actually afford to live in them.