No Tenure Trends Report this Quarter – And Here is Why
This blog post was cowritten by Enterprise Public Policy Intern Seara Mainor.
Normally in mid-summer, we would update and release our quarterly Housing Tenure Trends Report with new data from the Census Bureau’s Housing Vacancy Survey (HVS) on the trajectories of national homeowner and rentership rates. However, as with many routines this year, normal is not what it used to be.
Specifically, the Census Bureau noted changes it made in its data collection process as a result of the Covid-19 pandemic, which may have impacted the accuracy of reported homeownership and renter rates last quarter. For this and other reasons described below, we have reservations about whether the Q2 2020 data appropriately represents actual trends in housing tenure.
So, in place of our usual report, this blog post will still describe what the Census Bureau found with respect to tenure rates last quarter, but place these observations in the context of altered data collection methods and other Covid-19 related factors that could be driving the divergence in the reported data.
What the Data Show
The HVS data released last week by the Census Bureau suggests that national homeownership grew dramatically last quarter, reaching a seasonally adjusted rate of 68.2 percent. If accurate, this would be the highest recorded share of homeowning households since 2007 and put the homeownership rate within a percentage point of the all-time record of 69.2 percent in 2004. The share of American households that rent their home, meanwhile, was reported at 31.8 percent.
The Q2 homeownership rate is nearly three percentage points greater than the 65.3 percent of homeowners reported in the first quarter of this year. This represents not only the largest quarterly increase in the seasonally adjusted homeownership rate in the HVS series (e.g. back to 1980), but also surpasses the previous high in a single quarter change (+0.7 percent in Q2 2004) by more than a factor of four. Considering tenure data tends to move at a glacial pace, this increase is surprising, to say the least.
Further pushing the bounds of credulity, every subset of households tracked by the HVS saw similar dramatic increases in homeownership rates last quarter, with the largest of these observed among the groups most negatively impacted by the economic fallout from the pandemic. For example, Hispanic households and those with incomes below the national family median income reported more than three percentage point swings in their tenure rates, putting them both at new record highs for homeownership. Black households also saw large percentage point gains in homeownership, raising their rate over seven percentage points relative to Q2 2019, which was a record low for this group since the HVS series started tracking tenure rates by race and ethnicity in 1994. Households below age 45 also saw notably larger gains relative to older households.
Sources of Possible Discrepancy in Reported Tenure Rates
Taken at face value, these data would represent dramatic tenure changes for every subset of household. A deeper dive into the methodology, however, suggests a change in data collection procedures due to the pandemic may be contributing to the reported increase in homeownership rates more than any actual growth in the number of homeowning households.
Specifically, the Census Bureau suspended in-person interviews for the HVS near the end of March 2020, relying instead on only phone interviews to collect survey responses. A result of this change was a lower response rate overall to the survey last quarter, down to 67 percent versus 79 percent in both Q1 2020 and in Q2 2019. Yet the HVS weighting procedure does not account for non-response and does not adjust for differences in response rates of different subgroups, so reported estimates of the number of owner and renter households would in part reflect the lower number of surveys conducted, potentially skewing the conclusions of the survey.
The Bureau further notes that homeownership rates recorded from in-person vs. phone interviews routinely show the former to be lower, with 60 percent of in-person respondents in 2019 owning, versus 67 percent of phone interviews. Therefore, with no in-person interviews conducted in Q2, only the higher rate phone responses were recorded. The Bureau acknowledges these discrepancies but says it has not been able to estimate their magnitude and adjust reported tenure rates accordingly.
Other Pandemic-Related Indicators of Owner and Renter Market Behavior
Of course, it was not just the Census Bureau survey procedures that shifted in the wake of the pandemic and beginning of a recession; other changes were observed that may indicate some very real and dramatic shifts in tenure rates. For instance, after an initial drop in mortgage applications and home purchases, both measures have since rebounded strongly, spurred in part by record low mortgage interest rates - 2.98 percent for a 30-year fixed rate loan as of June. The share of homebuyers purchasing their first home also rose during the quarter, as households able to make the move from renting sought to take advantage of these low rates. These trends suggest that some of the increase in homeownership rates reported by the HVS may not just be a function of changes in data collection.
At the same time, the pandemic has induced turmoil in rental markets. Many renters lost their jobs when businesses in several states were shut down, and have struggled to pay their rent even with receipt of enhanced unemployment and federal stimulus fund to supplement reductions in income.
A patchwork of eviction moratoria covering a subset of rental units allowed some renters to remain housed through the spring, though most of these protections have ended and eviction filings are on the rise. If, a wave of evictions does occur, as many advocates are predicting, the loss of renter households may further bolster the homeownership rate, simply by reducing the total number of households even if the number of homeowners remains steady.
What to Expect Going Forward
Pandemic-related impacts on reported tenure rates – driven both by methodological changes and by actual market behavior – are likely to continue through the remainder of 2020. The Census Bureau, for its part, reported that changes in data collection procedures will continue in the third quarter of 2020, though some areas of the United States may reintroduce in-person interviews as virus conditions allow. Data released from HVS will therefore remain circumspect until full resumption of in-person interviews – or a better process for adjusting estimates based on data collection processes – is achieved.
The outlook for near-term activity in housing markets, meanwhile, remains uncertain, As noted above, mass evictions would impact the observed share of households that own and rent (in addition to causing significant hardships for renters). A surge in the virus could also bring about a return of business closures and stay-at-home orders, introducing further economic turmoil to owners and renters alike. Finally, the actions of state and federal governments will go a long way towards determining whether out-of-work renters will be able to stay in their homes.
Stay tuned to our blog as we continue to monitor the Census Bureau’s quarterly data releases and other indicators of housing tenure trends. You can also check Enterprise’s Resilient Futures resource page for more information on housing and policy responses to the pandemic, and subscribe to our daily and bi-weekly policy newsletters for more information on Enterprise’s federal, state, and local policy work.