Millennial renters in Washington, D.C. are six times more likely to live in a group house than their peers in Boston-Cambridge, MA.
Does housing demand create supply, or does supply create demand? There are no simple answers to this longstanding debate—the relationship can go in both directions. But the underlying dynamics of how local housing supply and demand interact can have sweeping implications for how people live, particularly people at the margins of the market such as young adults and low-income families.
Consider the diverging cases of Boston-Cambridge, Massachusetts and Washington, D.C. They are both versions of the archetypal, walkable cities that many associate with the modern 21st century knowledge economy. And both have large populations of young adults in school or just starting their careers. In this analysis, we compare the housing choices of millennials—young adults ages 18 to 33—in these two cities.
In many respects, the two cities are very similar. Millennials have similar employment rates in the two cities—68 percent in Washington, D.C. versus 66 percent in Boston-Cambridge—and in both cities, about two-thirds of millennials rent. Similarly, roughly half of those who rent in both cities live alone (or, live with only a spouse or child).
However, in other respects, the two cities are very different. Among renters who live with roommates, the choices of homes vary widely. Washingtonians are much more likely than Bostonians to live in group houses—that is, single family residences with multiple bedrooms shared by many unrelated individuals. By contrast, Bostonians who have roommates are much more likely to live in smaller apartments with fewer roommates.
As the figure below illustrates, 17 percent of millennial renters in Washington, D.C. lived in a group house compared to just 3 percent of their peers in Boston. By contrast, nearly half of millennial renters in Boston lived in shared apartments compared to under one-third of their peers in Washington.
The average number of “families” in a household, defined as the number of unrelated groups of individuals residing in the same household (a proxy for the number of roommates), is much higher in Washington than in Boston-Cambridge. In Washington, 31 percent of millennial renters lived in a house with four or more unrelated “families,” compared to just 19 percent in Boston.
The difference in the average number of “families” per household is driven overwhelmingly by the difference in percentages between millennials residing in group homes (higher number of “families”) versus residing in apartments with roommates. In Washington, 64 percent of millennials group houses had four or more unrelated “families” compared to 13 percent of multifamily units with roommates. In Boston-Cambridge, the shares were 58 percent and 17 percent respectively.
A number of factors explain these differences. One commonly held idea is that rent controls in Washington, D.C. have reduced incentives for landlords to convert large single-family residences into several separate apartment units. By contrast, single-family homes that have been converted into multiple, smaller apartment units are much more common in Boston-Cambridge, where voters eliminated rent controls in 1994. While local zoning regulations such as rent controls are frequently intended to assist households at the margins of the housing market, the opposite can also occur. As the data above suggest, these regulations can play an important role in shaping the household structure decisions of young adults.
 The data are drawn from the U.S. Census Bureau’s 2012 American Community Survey (ACS) made available by Steven Ruggles, J. Trent Alexander, Katie Genadek, Ronald Goeken, Matthew B. Schrowder, and Matthew Sobek, Integrated Public Use Microdata Series: Version 5.0 [Machine-readable database], Minneapolis: University of Minnesota, 2010.