Lawler: Likely "Dramatic shift" in Household Formation has "Major implications" for 2023

Early Look at the 2022 CPS/ASEC Household Data

Tom Lawler’s conclusion, Nov 21, 2022:

“There is strong reason to believe that household growth is no longer getting the “boost” from behavioral changes as that suggested from March 2021 to March 2021.  If that turns out to be true, then household growth is or will return back to growth more consistent with underlying population changes by age group, a development that would imply household growth over the next year of about half that apparently experienced from early 2021 to early 2022.  Such a dramatic shift has major implications for projections of rent growth and home price growth next year.”

This is from housing economist Tom Lawler:

Last week Census released data on America’s Families and Living Arrangements for March 2022, based on the results of the 2022 Current Population Survey Annual Social and Economic Supplement (CPS/ASEC).  Based on a relatively small sample (though larger than the regular monthly CPS), the CPS/ASEC attempts to estimate the number of and characteristics of households and families.  While the CPS/ASEC estimates have not generally been consistent with Decennial Census results, especially with respect to young adults, it is still useful to look at the survey data.

Note that the March 2021 estimate is 687 thousand lower than that shown in the 2021 CPS/ASEC report, reflecting the incorporation of the 2020 Decennial Census results (the 2021 CPS/ASEC estimate was based on “Vintage 2020” population estimates, which did not incorporate the Decennial Census results.). 

The CPS/ASEC results suggest that households grew by almost two million from March 2021 to March 2022, despite very modest growth in the adult population, with especially rapid % growth in young adult households. 

Young Adults Living at Home and Living Alone

Below is a table showing the CPS/ASEC estimates for the number of young adults (18-24 and 25-34) living at home (or a child of householder); the number living alone; and the number who were a “head” of household. (Numbers in thousands)

According to the CPS/ASEC results, from March 2021 to March 2022 the number of young adults living at home fell sharply, while the number of young adults who “headed” households increased significantly, with much of that increase coming from young adults living alone.

Below are charts showing some recent history on the share of young adults living at home.  Note that the CPS/ASEC data prior to 2021 have NOT been adjusted to reflect the 2020 Decennial Census results.

According to CPS/ASEC estimates, the % of 18-24 year olds living at home increased significantly right at the onset of the pandemic and stayed high in 2021 but fell back significantly in 2022.  And the % of 25-34 year olds living at home increased at the onset of the pandemic but by March 2022 had fallen to the lowest level since 2015.

Below is a chart showing the CPS/ASEC estimates of the % of 25-34 year olds living alone.

According to CPS/ASEC estimates, the % of adults (18+) living alone jumped to 14.9% in March 2022 from 14.4% in March 2021, and the % of households with just one member climbed to 28.9% from 28.2%.  Both “living alone” measures (share of population and share of households) were at the highest levels ever recorded by the CPS/ASEC.  The jump in one-person households was the major reason why the average household size as measured by the CPS/ASEC plunged to 2.50 in March 2022 (the lowest ever recorded) from 2.54 in March 2021. (Unrounded the average household size fell by .0341).   In terms of share of total households, the one-person share increased the most for young adults (up to 39 years) and for 65-74 year olds.

If the CPS/ASEC estimates were an accurate reflection of household changes from March 2021 to March 2022, then the almost two million increase in households over this period was to a very large extent driven by the surge in the % of adults living alone – that is, by “behavioural” changes rather than population changes by age group.  For example, if so-called “headship rates” households as a % of population) had been constant for each age group from 2021 to 2022, household growth would have been about half as large as that shown by CPS/ASEC estimates.

As I alluded to before, however, there are several reasons to question whether the CPS/ASEC results represent an accurate depictions of actual household changes.  For one, CPS/ASEC results have not historically been consistent with Decennial Census results, especially with respect to young adults.  For another, over the last few years the CPS/ASEC has been plagued by unusually low response rates, and there is some evidence that the characteristics of respondents differ from those of non-respondents.  (See: How Has the Pandemic Continued to Affect Survey Response? Using Administrative Data to Evaluate Nonresponse in the 2022 Current Population Survey Annual Social and Economic Supplement)

But if the CPS/ASEC results were accurate, then some obvious questions emerge:  why was there a big decline in the number of young adults living at home, and why was there such an large increase in the number of people living alone?  To what extent was the pandemic and the dramatic changes in the labor market (including the increase in telecommuting a factor in these changes?  Did declining college enrollment play a role?  How will the outsized increase in the cost of housing (both rents and the costs of buying a new home) since 2021 affect new potential household formations, and possibly existing householders living at home?  How will the public’s diminished concerns about COVID and the recent trend toward more companies requiring workers to return to the office affect behavior?

I’ll have more on these topics in later reports, but there is strong reason to believe that household growth is no longer getting the “boost” from behavioral changes as that suggested from March 2021 to March 2021.  If that turns out to be true, then household growth is or will return back to growth more consistent with underlying population changes by age group, a development that would imply household growth over the next year of about half that apparently experienced from early 2021 to early 2022.  Such a dramatic shift has major implications for projections of rent growth and home price growth next year.

Note: Previous articles on Household Formation:

McBride Sept 2021: Household Formation Drives Housing Demand

Lawler Nov 2021: The "Household Conundrum"

McBride Dec 2021: The Household Mystery

McBride May 2022: The Household Mystery: Part II

Lawler Oct 2022: Update on the Household “Conundrum”