The Household Mystery: Part II

Last year I wrote The Household Mystery and I noted:

Recently we’ve seen strong demand for both owner occupied and rental units. We can see this demand in rapidly rising home prices and in rents. This suggests a significant increase in household formation in 2021 (demand is increasing faster than supply).

It is no surprise that demand for homeownership has increased recently. As I noted in 2015, in the 2020s, a large cohort would be moving into the 30 to 39 age group (a key age for ownership). (See: Housing and Demographics: The Next Big Shift). However, the aging of Millennials suggests a preference for homeownership over renting, and not total demand for housing.

So, what is driving demand for both homeownership and rentals? It must be household formation!

Recently there have been two papers that shed some light on this mystery.

From SF Fed economist John A. Mondragon and Professor Johannes Wieland: Housing Demand and Remote Work. The authors showed that areas with more work from home (WFH) saw larger house price increases.

In this paper we show that the shift to remote work caused a large increase in housing demand. In turn, this increase in housing demand caused house prices and rents to increase sharply. Based on our cross-sectional estimates controlling for migration spillovers, we argue that remote work accounts for at least one half of the 24% increase in house prices from December 2019 to November 2021.

emphasis added ©Mondragon and Wieland

And from Federal Reserve economists Daniel García and Andrew Paciorek: The Remarkable Recent Rebound in Household Formation and the Prospects for Future Housing Demand

This note updates our previous work on household formation and living arrangements from the summer of 2020. At that early stage in the pandemic, the data showed a dramatic decline in headship rates as millions of Americans changed their living arrangements, many by remaining with or moving back in with parents and older relatives. …

In contrast, over the past year and a half there has been a remarkable rebound in the headship rate, driven in large part by a return to the pre-pandemic rates at which younger adults lived with parents or older family members. This rebound has been an important contributor to a huge increase in housing demand.

Putting these two papers together - it is very possible that work-from-home drove some of the likely sources of household formation over the last 18 months. For example, young adults working from home might have been more motivated to move out of their parent’s homes. Or roommates working from home might have decided to split up instead of sharing the same space all day.

And there were probably more separations and divorces. On divorces, we only have data through 2020, and divorces were down sharply at the beginning of the pandemic from 2.7 per 1,000 population to 2.3 per 1,000 population. When the data for 2021 is released, we will surely see an increase in divorces.

According to the Census Bureau, the number of households decreased in 2020, and the number of people per household increased to 2.53 from 2.52 in 2019. In 2021, the number of households increased sharply, and the number of people per household decreased to 2.51. Although there are questions about the accuracy of these Census surveys, this suggests strong household formation in 2021 (as indicated in the second paper).

Was this a one-time surge in household formation as the pandemic eased? Or will we see further increases in household formation even with little population growth (perhaps due to more work-from-home)? This is a key question for housing.

My suspicion is household formation will slow significantly, taking pressure off of demand.