Data Dig : Lower Housing Vacancy May Not Mean Housing Is In Short Supply

In a recent post, we explained the common methodology underlying the widely varying estimates of the purported housing supply shortage. We questioned use of Census Bureau nationwide vacancy data for rental and for-sale housing as the foundational data for sizing whether housing supply is inadequate and, if so, by what quantity. In this post, we turn to other reasons to question the housing shortfall assertion that underlies nearly all housing policy discussion in 2026.

At shoveltokeys.com, we offer interactive versions of the charts in our posts. On Substack, the charts are static images.

2005 Vacancy Rate Data As The Benchmark

Most of the reporting about housing supply relies on two premises:

• First that housing vacancy rates are a proxy for the balance (or imbalance) of housing supply and demand; and
• Second that the vacancy rates measured by the Census Bureau American Community Survey between 2005 (when the Census Bureau began collecting vacancy data) and 2010 are the benchmark for a balanced housing market.

From the below-average homeowner vacancy rate since 2021, researchers and housing advocates conclude that:

• Housing is in short supply;
• The supply shortage causes elevated home prices; and
• Building more homes would lower the price of homes.

In the remainder of this post, we examine the two premises, cast doubt on their validity, and set the stage for a future discussion of the three conclusions regarding housing supply and prices.

The Impact of the Subprime Mortgage Fiasco on Vacancy

Most of us will remember a housing dark age coincident with the early years of the Census Bureau vacancy data. From 2000 to 2005, the number of completed foreclosures (title to the home was transferred away from the defaulted borrower) averaged 238,586 per year. In 2005 (the first year of vacancy rate data from the Census Bureau), the owned-home vacancy rate was 1.7%.

Foreclosures and the Vacancy Rate

In the five years after 2005, the number of completed foreclosure climbed 400% from 293,541 in 2005 to 1,178,234 in 2010. From 2006 to 2010, foreclosure added 4,172,000 vacant homes to the market, 336% higher than the five-years from 2001 to 2005. By 2016 (the last year for which comparable foreclosure data is available), the housing market collapse had added over 8,165,000 vacant homes to the for-sale market — a total equal to 34 years of the 2000 to 2005 average.

A foreclosure-induced vacancy may last briefly if the home is in good condition and is purchased by a new occupant or rental investor. The vacancy may endure for many months if the property is badly damaged or lands in the real-estate-owned inventory of a bank. No matter the duration of the vacancy, the result of foreclosure necessarily is a rise in the vacancy rate.

The Census Bureau vacancy data appears to reflect the impact of foreclosures. From 2005 to 2008, the owned-home vacancy rate rose from 1.7% to 2.7%. The vacancy rate hovered at 2.5% through 2010 and declined slightly to 2.4% in 2011. The absorption of foreclosed homes did not materially drop the vacancy rate until 2012, when the rate declined to 2.0%. The vacancy rate did not return to the 2005 level until sometime in 2016 and stabilized through 2019 at 1.5%, just a bit lower than the 2005 rate. This history may suggest that a “normal” owned-home vacancy rate in the neighborhood of 2005’s 1.7% is the baseline that researchers should be using today to measure housing supply.

The Vacancy Rate from 2005 to 2016 Is Not a Good Baseline

We surmise that the vacancy rate in owned homes from 2005 to 2016 was driven upward by the enormous number of foreclosures during that time. If our inference is accurate or even reasonable, the utility of the vacancy rate in those 12 years as a baseline for housing shortage claims is undermined if not defeated entirely. We posit that the appropriate baseline vacancy rate likely is between 1.5% and 1.7%, the range that prevailed before and after the calamitous aftermath of the subprime free-for-all.

Institutional Investors Likely Moderated the Owned-Home Vacancy Rate

As an aside, the decline of the owned-home vacancy rate from 2013 to 2019 to a rate proximate to the 2005 rate suggests that some participants in the housing market stepped up to absorb the astonishing volume of foreclosures. Institutional investors likely were those participants. With their capital and operational capacity, those investors played a crucial role in quickly returning foreclosures to habitability and use.

At least in the context of recovery from the subprime fiasco, we think the criticism of large investors as buyers and lessors of single-family homes is unfounded at best. Given the millions of homes that sped from default to foreclosure from 2006 to 2016, we doubt that individual buyers could have absorbed the flood of inventory and returned the homes to the market nearly as quickly as well-resourced investor buyers. Even if large investor-buyers were competing with some individuals, most individuals had neither the financial resources nor the construction expertise to accomplish the post-foreclosure turnaround feat for one home much less millions. The rapid return of the market to the 2005 vacancy rate gives good reason to believe that investors played a vital role in the country’s recovery from the subprime mortgage debacle.

Foreclosures Likely Drove Down the Rental Vacancy Rate

Another aside, while we are on the topic of side-effects. The subprime mortgage mess almost certainly drove down the rental vacancy rate from the 8.0% range of 2005 through 2009. Foreclosed homeowners lose a home but not the need for a place to live. The logical path from repossession for a home-owning family is to a rented home. The Census Bureau data show that the rental vacancy rate dropped from a high of 8.4% in 2009 to roughly 6.0% in 2015. Like the owned-home vacancy rate, the rental vacancy remained stable through 2019. The migration of foreclosed-home families to rental is another side-effect of the subprime mortgage mess that undermines the utility of the vacancy rate from 2006 to 2015 as a baseline for housing supply or supply shortage in 2026.

And Then Came Covid

If the housing supply story ended with a return to vacancy rate stability after the crush of foreclosures ended, we might be hearing much less about a housing shortage. Covid had a different idea. As the first chart in this post reveals, the societal disruption of Covid and probably the enormous infusion of cash into the economy by the federal government pushed the rental and owned-home vacancy rates down rapidly.

Because of Covid, no data is available from the Census Bureau for 2020 vacancy rates, but from 2019 to 2021, the rental vacancy rate fell from 6.0% (a stable rate from 2015) to 5.2%. The owned-home rate dropped to 0.9% from 1.5% during the same pandemic timeframe. The reason simply was a sudden and large increase in household formation from 2019 to 2021 provoked by:

• Enormous fiscal stimulus by the federal government;
• The sudden freedom to work at home;
• Concern for personal health safety;
• A demand backlog created by intensely tightened mortgage lending standards after the errors of the subprime profligacy.

The flood of household formation drove up rapidly demand for housing. But for the unprecedented influence of Covid, household formation and housing demand likely would have climbed much more slowly. Rental and owned-home vacancy rates certainly would not have plummeted, as they did from 2019 to 2021. Housing supply as inferred from the vacancy rate would not seem so constrained.

Up Next

We call into question in this post the assertion that low rental and owned-home vacancy rates demonstrate an inadequate supply of housing. Were it possible to imagine a 20-year history without the subprime mortgage mess and Covid, the purported inferential power of vacancy rate would disappear, we believe. In such a world without calamities, the owned home vacancy rate likely would have been stable, and housing researchers, advocates, and policy-makers would lose the data foundation for the claim that the housing market lacks adequate supply today.

Challenging the assumed connection between vacancy rate and supply adequacy is not simply an academic pursuit, although we enjoy the research and chart creation. By undermining the vacancy - supply foundation of the supply shortage argument, we also destroy the corollary that rents and home prices are currently elevated by scarcity. As we will explore in future posts, scarcity is not sustaining rents and home prices at anti-competitive prices.

The housing affordability problem is a function of the cost of producing the housing that Americans want to rent and buy and that builders and investors find appropriately profitable. We must lower the cost, change the demand, lower acceptable profit, or devise some combination of the three.