Household Size & Apartment Demand

For some time now, apartment industry experts have been pointing to strongly positive demographic trends

For some time now, apartment industry experts have been pointing to strongly positive demographic trends that should increase apartment demand. Reasons cited include overall population growth, echo boomers entering the housing market, continued high levels of immigration and a continued shift in household composition away from the “traditional” married couple with children households toward single-person (as well as single-parent) households.

But there are some potential concerns in this forecast.  For one thing, a sluggish employment recovery for echo boomers may keep many of them “doubled up” with roommates (or in their parents’ homes) for some time to come. And there is the question of whether homeownership rates will rise in the intermediate term.

And there is still another concern beyond these: household size may be increasing. This raises the question of whether larger households will “soak up” much of the relevant population increase, resulting in fewer new household formations. Let’s investigate.

For most of the past 60 years, the average household size has been declining. Following World War II, there was an average of about 3.56 persons per household. Households began to get smaller at a rapid clip in the second half of the 1960s, dropping to an average of just 2.62 in 1989.

Beginning in 1990, household size seemed to rise a bit before tapering off again; by 2000 it had fallen back to the 2.62 mark.  Since then, it has inched lower, reaching 2.57 in 2009. It’s too early to conclude that household size is now increasing; we’ll know more when the 2010 Census is released next April.

Household size declines

One key factor behind the postwar trend in the average U.S. household size has been the drop in the number of adults (those at least 18 years old) per household. At the beginning of the postwar period there were 2.5 adults per household. By 1969, the number fell to 2.05.

Thereafter, the number of adults per household declined more slowly, edging down to 1.91 in 2003 before starting to rise a bit. That trend was offset at first by the 1946-1964 baby boom and the consequent increase in the number of children (under 18) per household: from 1.12 in 1947 to 1.24 in 1964. After that, however, the number of children per household dropped sharply, falling to 1.00 by 1973.

The baby boomers may have had another effect on the average size. At first, this decline in the number of adults per household was more evident in nonfamily households (unrelated individuals living in the same housing unit and people living alone) than in family households.

But beginning in the early 1970s, the average number of adults in nonfamily households began to rise—from a low of 1.16 to a level of 1.33 in 1994.  Since then, it has changed little. The increase was likely caused by the entry of baby boomers into the housing markets starting in the late 1960s—and the fact that young people are more likely to have roommates than are older people.

There is one other clear determinant of the average size of households: the number of people living alone.  For more than 60 years, the share of single-person households has risen sharply, tripling from just 9.2 percent in 1950 to 27.5 percent in 2008. By comparison, the share of married couples with children—once regarded as the “traditional” household—peaked at 44.3 percent in 1957 and has fallen to just 22.0 percent in 2009.

The share of single-person households dipped a bit in 2009 to 27.0 percent.  In the past, these dips were offset within a few years. There is good reason to think that there is once again pent-up demand for household formation, particularly among young adults this time, too.

Last year, 29.4 percent of adults 18-34 years old were living at their parents’ (or parent’s) home, the highest share ever for the years in which we have data.  This is well above the 27-28 percent range that prevailed during the past quarter century. The total number of young adults living with parents is a record 20.3 million. If an economic recovery brings the share down even to 28 percent, that still would mean an additional 950,000 households, most of whom would be renter households.

The increase in household size seen in 2009 seems to have been driven in large part by a pickup in the number of young adults living with a parent. That was surely caused by the Great Recession and is likely to be reversed as the economy recovers. To be sure, if the recovery remains weak and job gains small, this pent-up demand will manifest itself at a slower pace.  But there is every reason to think that a better economy will push household formation rates back up.

Thus, NMHC’s tentative conclusion: increasing household size may not be among the industry’s big concerns once the pent-up demand from the echo boomers is released into the market.

(Mark Obrinsky is chief economist and vice president of research at the National Multi Housing Council in Washington, D.C.)

To comment on this story, e-mail