Stated Housing Preferences vs. Revealed Housing Reality

In March, the National Association of Realtors released the 2011 Community Preference Survey. Some greeniac urbanists are touting the survey results as evidence that smart-growth ideology is starting to align with Americans’ preferences for housing. For instance, the 2011 survey finds that 19 percent of Americans wish to live in dense central cities, up from 13 percent in 2004 when the survey was first conducted. Richard Florida declares victory. I remain unimpressed, and even generally pro-smart-growth urbanists are noting the contradictory stated preferences of survey respondents.

There is good reason to be generally skeptical of supposed evidence for or against a given position that is obtained through surveys. Talk is cheap. Action, on the other hand, is a far better measure of an individual’s preferences — in this case, a snapshot inventory of where people actually choose to live. According to the most recent American Housing Survey from the Census Bureau (an imperfect comparison, yes, but an interesting one), 29 percent of occupied housing units were located in central cities in 2009 (27.3 percent of occupied housing units were located in cities with at least 100,000 residents). I have yet to see anyone seize on the apparent disparity between the two reports and claim that this is evidence that many current residents of dense cities would rather live in less dense suburban, small town, or rural settings.