Where Does Location Affordability Drive Residential Mobility? An Analysis of Origin and Destination Communities

Abstract

Despite an overall decrease in residential mobility after the 2007 housing crisis, many households, particularly those that are low income, continue to move in pursuit of a better life. Traditional theories of residential mobility suggest that mobility will occur when housing and transportation costs are cumulatively greater than the cost of moving to a new location. At the same time, the influence of these factors is not likely to be uniform across geographic contexts or for moves up or down the metropolitan hierarchy. Our analysis examines how well affordability measures explain patterns of county-level residential mobility. Specifically, we contrast conventional measures of affordability focused on the ratio of income to housing expense with measures of location affordability that factor in both housing and transportation costs. We find that whereas households tend to move from lower to higher cost locations, transit affordability at the destination plays an important role in mobility decisions.

Keywords:

• Residential mobility
• population change
• location affordability

Notes

1. Whereas, ideally, our model would examine mobility patterns at as small a geographic unit as possible (such as the census block groups for which LAI data are available), the most reliable national source of data on residential mobility flows contains data only at the county level. We recognize that the inherent size and heterogeneity of counties represents a significant limitation of our modeling strategy; however, we see important benefits to be gained from developing an aggregate model of mobility that could be refined with more disaggregate data.

2. For each of the regression models, 500 permutations were used to estimate standard errors.

3. The 2013 USDA Rural–Urban Continuum Codes OMB definitions of metropolitan and micropolitan areas provide a finer grained characterization of these areas’ counties. Metropolitan counties are divided by size, and nonmetropolitan counties are divided by size and whether the county is adjacent to a metropolitan area. The resultant classification scheme includes nine categories. For the sake of simplifying our analysis, we consolidate the six nonmetropolitan and rural classifications (which are subdivided by county size) into three categories—nonmetropolitan, adjacent to a metropolitan area; nonmetropolitan, not adjacent to a metropolitan area; and rural. These three categories represent slightly less than 15% of the United States population.