WASHINGTON — The US faces a housing shortage of millions of homes, and closing that gap will require a dramatic and sustained increase in apartment construction, according to a new report from the Center for Public Enterprise that argues federal financing policy—not just zoning reform—must play a central role in solving the crisis.
In Raising the Housing Investment Level: The History and Future of Multifamily Investment Policy, Executive Director Paul Williams contends the nation cannot end its housing shortage without increasing annual multifamily production from roughly 350,000 homes to at least 500,000 homes per year and maintaining that pace for eight to 10 years.
The report says that level of construction has only been achieved during periods when the federal government actively made apartment development financially attractive.
“America has a housing shortage,” the report states. “Depending on how you count, we’re short between 2 and 7 million units, and the gap is widening.”
The report arrives as housing affordability continues to strain households nationwide. According to the findings, half of all renters now spend more than 30% of their income on housing, a commonly used threshold for cost burden.
Meanwhile, annual apartment construction has remained near 350,000 homes for four decades despite population growth, job creation, economic booms and recessions.
“This report rests on a simple claim: closing the housing gap requires multifamily housing production rising from its current level of 350,000 units to at least 500,000 units per year, and sustaining that elevated level for 8-10 years,” the report states.
Williams argues that public debate has focused heavily on regulatory barriers such as zoning restrictions, environmental reviews and permitting delays, but financing constraints have received far less attention.
While many local governments are now pursuing land-use reforms, the report says those changes alone are unlikely to produce enough homes if developers cannot secure workable construction financing.
The study identifies a large inventory of approved but stalled housing developments.
It estimates that approximately 750,000 multifamily homes nationwide have already cleared zoning and land-use approvals but have not yet applied for construction permits. That total is roughly six times the number of homes with permits that have not broken ground and slightly exceeds the number currently under construction.
“Many of these projects have completed architectural drawings and engineering work,” the summary states. “What they lack is financing on terms that make them viable.”
The report cites major metropolitan areas as examples of the stalled pipeline. San Francisco alone accounts for an estimated 52,000 such homes, Boston has 23,000, and New York City has roughly 100,000 homes in its broader active development pipeline.
Because no federal agency tracks the entitlement stage of development nationally, the report describes its 750,000-unit figure as a reasonable estimate rather than a precise count.
Williams also draws on postwar housing history to argue that federal policy has repeatedly proven capable of driving large-scale apartment construction.
“The only times in modern American history we’ve achieved anything close to that were periods when coordinated federal investment policy made apartment construction unusually attractive for investment,” the report states. “For the past 40 years, we’ve had no such policy. And for 40 years, multifamily production has flatlined.”
The first major boom followed creation of the Section 236 program in 1968. That policy used mortgage insurance, subsidized interest rates and federal liquidity tools to lower financing costs and encourage apartment construction. According to the report, it helped produce roughly 3 million apartment starts between 1970 and 1973.
“The HUD-driven boom of 1968-1973 leveraged mortgage insurance, liquidity, and interest rate subsidy, contributing to 3 million apartment units started over four years,” the report states.
A second boom occurred in the early 1980s under favorable tax treatment for rental housing. But after the Tax Reform Act of 1986 repealed key depreciation incentives, multifamily starts plunged from nearly 600,000 to fewer than 140,000 by 1991, according to the report.
“Both programs had some flaws, and both eventually ended,” the report states. “But importantly, both worked: when government made it financially attractive to build apartments, large numbers of apartments were built.”
The report is especially critical of the current performance of the Federal Housing Administration’s 221(d)(4) construction loan program, which it describes as highly competitive in theory but hobbled by delays in practice.
The program can provide fixed-rate loans covering up to 90% of project costs and can shield borrowers from personal liability if projects fail. Yet it finances only about 4% of new apartments.
“What Williams calls the best construction financing product in the country on paper” has become difficult to use because approvals take 270 to 360 days, compared with 60 to 90 days for conventional lenders, according to the summary.
The report attributes those delays to the absence of a codified single-underwriter model, outdated information technology systems, and multiple layers of environmental and compliance paperwork. At the same time, private banks have tightened construction lending standards since 2022, pushing many builders toward higher-cost private debt funds.
As a result, the report says developers are often bypassing the government program intended to support housing production.
To address the financing bottleneck, Williams outlines five federal policy options, several of which he says would require modest public spending or administrative changes rather than major new appropriations.
The first recommendation is to streamline FHA multifamily lending by accelerating processing times and expanding categorical exclusions for infill development. The report says those changes could increase production by tens of thousands of units annually “at essentially no fiscal cost.”
The second proposal would create a one-time $2 billion federal match for state housing finance agency acceleration funds. Those state programs provide subordinate construction loans that recycle capital into new developments as loans are repaid. The report estimates such an initiative could support roughly 30,000 additional apartments per year with no continuing cost.
A third recommendation calls for allowing government-sponsored enterprises such as Fannie Mae and Freddie Mac to expand into construction lending, including construction-to-permanent loans, forward commitments or a secondary market for mezzanine loans. According to the report, that could unlock a major new source of development capital.
Additional proposals include accelerated depreciation schedules for rental housing and renewed use of the Federal Home Loan Bank system to support affordable housing supply through a larger share of housing investments.
“These levers are neither particularly expensive nor are they particularly partisan,” the report states. “Taken together, they could allow us to quickly realize the potential of ongoing zoning reform efforts by adding hundreds of thousands of units to annual multifamily production.”
The report’s broader argument reflects a growing consensus among many housing economists that the nation’s affordability crisis is rooted in persistent underproduction. In many metropolitan regions, homebuilding has lagged behind population growth, household formation and employment growth for years, placing upward pressure on rents and home prices.
Rather than framing the debate as regulation versus markets, Williams argues the nation should also confront the role of capital markets and public finance. If zoning reforms expand what can be built, the report suggests financing reform may determine whether those homes are ever delivered.
“The housing shortage is serious but not intractable,” the summary concludes, pointing to prior eras when federal policy drove apartment construction above today’s levels and to the large pipeline of already approved projects waiting for financing conditions to improve.
Follow the Vanguard on Social Media – X, Instagram and Facebook. Subscribe the Vanguard News letters. To make a tax-deductible donation, please visit davisvanguard.org/donate or give directly through ActBlue. Your support will ensure that the vital work of the Vanguard continues.
Two of the local challenges to building apartments in Davis are:
1) the local apartment operators have almost all deployed a low-rise apartment model … two or three stories. We saw that play out in the Nishi 2018 proposal where the 67 units per acre approach of Nishi 2016 was scaled back to only 27 units per acre so that all the buildings were 3 stories.
2) Core construction materials for buildings over 4 stories are believed to be beyond the structural carrying capacity of wood, and need to have a steel/concrete core (Note: I don’t know how Davis Live on Russell just past Trader Joe’s was built, but it would be interesting to know). However, while listening to NPR last week there was a segment on how “Cross laminated timber” is allowing builders to go up well over 10 stories without needing steel rather than wood. That could be a real game changer in Davis. The segments can be read/listened to at https://www.marketplace.org/story/2024/12/30/clt-sustainable-building-material-wood-timber-construction-housing-high-rises
and
https://www.wbur.org/hereandnow/2026/04/15/laminated-timber-construction
Because as everyone says, “The best neighborhoods in Davis are the ones that are all apartments”. Oh yeah, no one says that.
In other words, they now (also) want “free fries and a coke” with that meal.
If these things aren’t “penciling out” on their own, that (by definition) means that market demand isn’t sufficient. Alternatively, it’s a sign of unequal wealth distribution.
Either way, there is no housing shortage. The article itself has an absurdly wide range (2-7 million). If they can’t even narrow THAT down, why would anyone believe them at all?
There are already newly-built apartments across the country that are sitting empty, because they can’t rent them out. They’re also having trouble selling them, for that same reason.
1.6 kids per couple, these days, and immigration essentially reversed. Don’t be afraid of change.
“Alternatively, it’s a sign of unequal wealth distribution.”
The most likely answer given what is happening in the housing market and why VF is pushing for more large houses.
As I’ve pointed out, the number of people per household in California increased from 2.9 in 2010 to 3.2 in 2020 despite a falling birthrate. That shows that we are 10% short of housing at least. Also, the house price premium in Davis shows that we are at least 9,000 homes short as I showed earlier.
Neither of those figures has anything to do with a housing shortage, and you created the latter figure from your imagination.
But since you’re advocating for “smaller” units, it’s likely to result in more occupants per square foot. Exactly the opposite of what “families” (as we usually define that term) seek out. (Then again, I don’t see why pursuing families in particular is a goal in the first place. For one thing, they usually have at least two cars per household.)
Also, there’s this:
“Since 1900, he said, housing construction in the United States has outpaced population growth, even when you account for the reduction in household size.”
https://48hills.org/2026/04/the-best-and-worst-of-ca-housing-policy-on-display-at-ucla-conference/
How does an increase in household size driven by more adults living in the same abode not reflect a housing shortage? And how does a scarcity pricing premium not reflect a shortage? Here’s a definition of scarcity value for your perusal: https://en.wikipedia.org/wiki/Scarcity_value I used a standard economic tool to calculate the excess demand (and/or supply shortage) that is creating the scarcity price premium for Davis. No imagination, just professional training and experience.
And I have posted several times the NAHB study showing the younger families are looking for smaller homes at a lower price. They understand the tradeoffs to get into the bottom step of the housing ladder.
Also cost of living especially housing are linked to people delaying and reducing the number of children they have. Ron keeps acting like that number is set in stone and destiny.
“The most likely answer given what is happening in the housing market and why VF is pushing for more large houses.”
That’s evidence that there’s a difference in wealth, not a housing shortage.
They wouldn’t be proposing them (and spending a bunch of money for an EIR, campaign, etc.) – if there wasn’t a market.
$1 million or so is “chump change” for the population that can actually buy a house, these days. And no – it’s not based on income.
“Also cost of living especially housing are linked to people delaying and reducing the number of children they have.”
The 1.6 kids number is nationwide – which therefore includes places like Topeka, Kansas (which I assume has pretty low housing prices). Not to mention the entire state of Utah, where the Mormons live. (I just read that one of the multiple “Osmonds” died. But since it wasn’t Donny or Marie, it’s barely news.)
I believe that the birthrate has consistently declined over time.
Ron O
Unlike you, I try to read the links that you post. And yet again I found that you don’t fully understand the citations that you provide. Tim Redmond is relying on a fatally flawed study to make his assertion about housing supply. Here’s TWO different critiques that found the same flaw in that study:
https://papers.ssrn.com/sol3/papers.cfm?abstract_id=5227968
https://michaelwiebe.com/assets/supply_constraints/supply_constraints.pdf
Here’s an abstract from the first one:
Louie, Mondragon, and Wieland (2025) use total income to measure demand for housing. But total income is mostly determined by population growth, which is co-determined with housing supply growth, which is jointly determined by supply and demand for housing. Thus, what they call demand is in fact the equilibrium of supply and demand.
Because their model is poorly grounded, its core mechanism cannot match the data, which show rising prices in highly-constrained metros. Their regressions imply that those metros have unexplained downward supply shocks and upward demand shocks together causing a price increase of at least 0.5 percent per year.
Here’s a simpler explanation of why the study is wrong: https://www.econlib.org/never-reason-from-a-population-change/
You’re citing abstracts, rather than complete studies.
In any case, what exactly does this mean to you? It sounds like something Kamala Harris would say:
“But total income is mostly determined by population growth, which is co-determined with housing supply growth, which is jointly determined by supply and demand for housing. Thus, what they call demand is in fact the equilibrium of supply and demand.”
Put a number on that (other than “somewhere between 2-7 million”, as this article does without any explanation or calculation whatsoever).
And it’s not Tim Redmond who is increasingly challenging the so-called housing shortage. It’s professors at universities (other than the resident law school YIMBY professor at UC Davis). The same professor who ironically lives in San Francisco, and commutes to “much cheaper” Davis for his employment.
“I used a standard economic tool to calculate the excess demand (and/or supply shortage) that is creating the scarcity price premium for Davis. No imagination, just professional training and experience.”
Skimmed your reference (took all of 10 seconds), and no – you didn’t use any tool referenced on that website.
One of the problems with your calculation (which you haven’t actually shown) is that if, for example, developers were allowed to run amok just outside of Davis (and somehow “suddenly” built 9,000 houses) is that the value of houses in surrounding communities would also drop, if your assumption was correct.
Meaning that the price differential would continue to exist.
Your assumption completely disregards alternatives (not just with housing, but with job opportunities elsewhere, etc.).
There aren’t 9,000 families clamoring for houses in “only” in Davis. If there were, they’d be bidding up the price of existing housing a lot higher than it is. And there’d be a lot more existing homeowners willing to sell.
Not sure if you realize this, but Davis isn’t heaven on earth. It’s barely even better than Spring Lake – and buyers are “proving” that by buying houses there.
Though the best/most-solid deal is a pre-existing house in Davis, which is also usually not subject to very much (if any) Mello Roos.