Residential mobility shapes both the experiences of individuals and the characteristics of neighborhoods. Approximately 12 percent of U.S. households relocate yearly, and residential mobility rates are higher among low-income households, renters, and younger families. Neighborhoods vary in their levels of residential mobility as well, and housing units turn over more frequently in neighborhoods: (1) with low homeownership rates; (2) with larger shares of households without children or stable employment; and (3) undergoing changes in the built environment or local economy. While residential mobility is neither uniformly positive nor negative, the implementation and outcomes of community change initiatives are affected by the level and type of mobility in their target areas. As such, practitioners and policymakers need to make better use of available data and analytic methods to understand mobility in the communities they serve.
Why We Need to Understand Residential Mobility
Community initiatives and organizations strive to benefit people and places, and data can measure the progress and effects of these efforts. However, the fact that residential mobility is one of the primary factors driving neighborhood change makes the interpretation of community indicators challenging. For example, a neighborhood may seem as if it is improving based on point-in-time indicators, but these “snapshots” of neighborhood measures cannot distinguish whether conditions are improving for current residents or whether change is being driven by an influx of newcomers who are better off than residents they are replacing. By digging deeper into the data, we can shed light on the questions of gentrification and displacement that often come up in community change initiatives.
An incomplete understanding of mobility can also lead to misinterpretation of data that indicate a lack of change in community conditions. Households, for example, may take advantage of neighborhood-based services and opportunities to improve their situation and then move, but the neighborhood socioeconomic profile may seem unchanged if these households are replaced by newcomers with similarly high levels of need. Relying only on point-in-time measures, one could incorrectly conclude that the assistance programs available in a neighborhood were not effective. A study of residential mobility and neighborhood change in 10 cities concluded that some neighborhoods that remained poor were really very dynamic, serving as launch pads for many residents who moved to better areas only to be replaced by poor newcomers. Other neighborhoods that also appeared poor in consecutive snapshots were more like traps in which numerous poor residents were stuck in place for extended periods with few opportunities to move up, either residentially or economically.
Mobility also has implications for the success of community initiatives, since too many or too few moves can affect whether individuals are able to benefit from interventions intended to help them. Frequent residential moves resulting from stress or yielding no improvement in circumstances have negative consequences, particularly for children. Community programs may create new housing or economic opportunities in a neighborhood, but if households are forced by circumstances to move frequently, they may not remain in the same area long enough to benefit from the programs. The goals of the community initiatives may also be thwarted if minority individuals have difficulty leveraging their new skills or assets to move to better neighborhoods because their freedom of movement is constrained by racial and economic segregation. By understanding the many ways that mobility can affect people and neighborhoods, practitioners can tailor programs to their particular context and knowledgably interpret indicators of neighborhood change.
Perspectives on Residential Mobility
Gauging residential mobility as a positive or negative force, and then shaping interventions to suit, depends on understanding a number of factors such as the reasons for moving, the frequency and timing of moves, and the results of relocation for people and places. For individuals, moving to a new home can reflect positive changes in a family’s circumstances, such as buying a home for the first time, moving to be close to a new job, or trading up to a better house or apartment. But mobility can also be a symptom of instability and insecurity, with low-income households making short-distance moves because of problems with landlords, creditors, or housing conditions. Similarly, staying in place can reflect a family’s security and satisfaction with its home and neighborhood surroundings, whereas in other cases, it may reflect lack of resources to move to a better home or neighborhood. A study of disadvantaged neighborhoods in 10 cities found that only about one-third of movers left for better places, while two-thirds simply moved nearby to similarly distressed circumstances. Likewise, approximately two-thirds of those families who remained in place did so reluctantly, and only about one-third were satisfied with their situation.
At the neighborhood level, residential mobility can be problematic but can also be a source of neighborhood vitality. As a negative force, excessive residential turnover can diminish neighborhood social ties, and weaken neighborhood institutions by disrupting neighbors’ participation. High residential instability, combined with concentrated disadvantage, undermines the ability of neighbors to take collective action, and in turn limits the ability of residents to prevent crime and maintain safety and order in their neighborhoods. On the positive side, increased housing turnover is key to maintaining housing market strength and preserving homeowner equity as areas attract newcomers. An optimal level of turnover can bring talent, vitality, and enrichment to the neighborhood. Moreover, mixed-income development, one important policy option for reducing persistent poverty and social exclusion, requires the replacement of at least some low-income households with middle-income households.
Finally, data on residential mobility must be interpreted in light of what is known about the structural influences shaping neighborhood selection. For low-income households, particularly those that are African American, the neighborhood selection process is fraught with roadblocks and constraints. A history of racial discrimination in housing markets and a lack of affordable housing make many neighborhoods simply out of reach. Studies of household residential mobility show African Americans are disadvantaged compared with whites regarding access to better neighborhoods. This adverse neighborhood selection process tends to reinforce neighborhood inequality, often across generations, where families may be mobile but only within low performing neighborhoods that offer limited educational and economic opportunities. Because the cumulative effect of various neighborhood contexts affect individual well-being, structural barriers to residential mobility tend to reproduce social inequality over time.
Measurement Issues and Data Sources
Measuring residential mobility is not a simple matter, and a range of indicators is needed to capture different aspects of the process: residential mobility, residential instability, housing unit turnover, and neighborhood change. Researchers should carefully select data collection and analysis methods, paying attention to aspects of residential mobility that are of interest and practical to measure. For example, should the focus be on individuals, households, housing units, or neighborhoods? Should the measurement be based on a cross-sectional or longitudinal perspective? What is the definition of moving with respect to time and space? What data sources are available, such as individual or household surveys, administrative records, or interview data? The common approaches to measuring residential mobility differ regarding these questions, and these nuances must be taken into account in interpretation.
Cross-sectional measures of individual residential mobility
One of the most commonly used measures of residential mobility is from the U.S. Census Bureau’s American Community Survey (ACS). The survey asks each individual in the sampled households whether he or she lived in the current housing unit for at least one year. If the answer is no, the survey asks for the respondent’s previous location. The residential mobility rate for the neighborhood is calculated as the percentage of the population that was not in the sampled housing unit in the previous year. This ACS−based residential mobility measure is cross-sectional and can be interpreted as a one-year residential mobility rate for individuals. For small areas such as census tracts, the one-year individual residential mobility rates must be derived from the ACS five-year sample estimates. The summary measure reflects how many people surveyed each month during the five-year period had moved within the year. Although most residential moves are local, the ACS data also identifies the prior year’s move as from outside the county, state, or country. Numerous studies of neighborhoods use this cross-sectional measure of residential mobility. It has played an important role in research on community social organization.
Household level measures of residential mobility
The measurement of residential mobility for households is a more complicated matter than for individuals because it is difficult to disentangle household moves from changes in household composition. Community initiatives are often interested in the latter because household turnover has different implications for their programming than does the movement of individuals. Households may be doubling up, for example, or children may be placed with relatives in new homes. The census measures do not distinguish between individuals who return to or become part of an established household, such as a child returning from college or the addition of a spouse to the family, and an entirely new household moving in. Capturing these distinctions requires survey data on the same households over time. The survey must contain information on the individuals who compose the household at each survey point, allowing analysts to compare household membership to determine whether individuals or the entire household moved.
Data collected in low-income neighborhoods in 10 cities for the Annie E Casey’s Making Connections program provides an example of this kind of analysis. By matching the individuals in the household rosters gathered during different waves of the survey, researchers were able to determine whether the entire household made a residential move, or whether specific individuals left or joined the household. Their analysis found that household compositional changes were much more frequent than residential moves that involved the entire household relocating. This type of analysis was also able to characterize the household changes based on the ages and relationships of household members.
Frequency of mobility measures
Although most families move infrequently, community initiatives often want to estimate the level of recurrent mobility among residents since frequent movers face particular challenges. But cross-sectional mobility rates do not capture this information. Instead, frequent mobility can only be calculated using information on the number of residential locations in which individuals have lived during a specified period. For example, a longitudinal study of low-income families in Michigan asked mothers to provide their residential movement history at each wave of data collection. Using this fine grained information, researchers have been able to measure the frequency of moves for children at various ages to determine the patterns of movement that put children at risk at certain developmental periods.
Even longitudinal studies do not provide all of the answers. It is difficult to measure the extent of residential mobility at the community level because longitudinal surveys seldom have enough respondents to provide reliable indicators for neighborhoods. However, the advent of Integrated Data Systems is making this measurement of frequent mobility more feasible. These systems link administrative records from multiple agencies at the individual level. The data can cover states, counties, or cities, and increasing evidence suggests that they are cost effective data sources for longitudinal research and policy evaluation. These systems can be a source for measuring frequent mobility if they capture address histories for individuals from the various administrative records. Such address history data can be used to estimate numbers and frequencies of moves for the population covered in the data system. With these types of measures, stakeholders can identify pockets of frequently mobile families or individuals and explore interventions, reduce mobility, or at least minimize the harmful consequences of frequent mobility.
Measuring housing unit turnover and neighborhood change
Housing unit turnover and the role that it plays in neighborhood change reveals another aspect of mobility. The housing stock of a neighborhood typically changes slowly, whether through construction, demolition, or rehabilitation. The dynamic movement of households into and out of housing units, though, is a continuous flow that can affect the neighborhood in many ways. Housing unit turnover can have positive or negative consequences for the community depending on its magnitude, velocity, and the characteristics of those moving in and out. Community initiatives have the potential to manage these shifting population dynamics so that neighborhoods thrive rather than crumble under the momentum, but only if they have information on the patterns of housing unit turnover and the resulting changes in the population.
Examining residential mobility through the lens of housing units enables an understanding of point-in-time neighborhood composition and the flow of households that shapes it. Investigating the stock and flow requires longitudinal data on housing units and their occupants, but such data sets are relatively uncommon. One exception is the American Housing Survey (AHS), which tracks a panel of housing units in selected cities. A limitation of the AHS for communities is that the sample of housing units is selected to be representative of metropolitan areas rather than neighborhoods, so the data cannot be used to describe small areas. Nevertheless, this data can be used to understand the dynamics of displacement and gentrification. One study using a national sample of housing units in high-poverty neighborhoods suggested that increases in average income of occupants were a combined result of richer households moving in, the exit of some poorer households, and income increases for low-income households that stayed in place. Thus, wholesale displacement of the poor was not the norm but, instead, the socioeconomic status of occupants gradually changed.
The Making Connections survey mentioned previously is an example of a data source that tracks a representative sample of housing units within specific neighborhoods. Therefore, it can also be used to look at the contributions of housing unit turnover to neighborhood change. An analysis of these data shows that residential mobility levels are related to housing unit and neighborhood conditions. For example, single-family rental homes are more likely to turn over to poorer residents, even though the rate of turnover is higher in multifamily buildings. This may be caused by the tendency of owners of single-family homes to defer maintenance when they convert the properties for rental occupancy. In addition, housing units tend to turn over to poorer occupants when the surrounding neighborhood has low levels of social cohesion and safety.
Having data on housing unit transitions and how those shape the mix of households in neighborhoods can be useful to community planning and development. This type of information can help communities monitor issues of concern such as disinvestment, gentrification, displacement, and segregation. Knowledge of where these processes are occurring, both with respect to types of housing units and their locations, can help guide action and enable the evaluation of progress on these fronts.
Qualitative measures related to residential mobility
The decision to move and the choice of where to move is the end result of a unique combination of household, housing unit, and neighborhood factors that are difficult to capture with quantitative data. Data gathered through in-depth interviews or focus groups with both movers and stayers shed light on how individuals take these factors into account when deciding where to live. Studies conducted with families that were offered the opportunity to move from extremely poor to low poverty neighborhoods as part of the Moving to Opportunity (MTO) experiment illustrate the value of qualitative data. Even though on average these families moved to lower-poverty areas, many MTO households were not able to move to significantly better, or more racially diverse, places. In fact, their decisions were influenced by a number of considerations that would not have been easy to quantify, such as connections with family and friends, perceptions of whether the neighborhood and school were a comfortable fit for their children, and other qualities of the physical and social environment. In addition, movers often lacked information about housing options and neighborhood and school quality that could have been helpful in seeking the best locations for themselves and their children. Combining qualitative and quantitative data allows communities to better understand the factors underlying residential mobility and to develop strategies that enable residents to make successful residential decisions.
Conclusions and Implications
Community organizations with access to a variety of measures that can shed light on residential mobility processes will be best positioned to make strategic decisions about their work with people and places and accurately evaluate the effect of their efforts. Communities would be wise to look beyond census measures of residential mobility and demographic profiles to fully evaluate neighborhood change. Longitudinal surveys of households and housing units, although costly, can provide additional insight into characteristics of households that move, whether or not they improve their circumstances, and how these individual decisions shape the trajectory of neighborhoods. The Making Connections survey is a good example of a useful community data collection tool. Integrated Data Systems are also promising adjuncts to census surveys. Networks such as Actionable Intelligence for Social Policy share the lessons of what it takes to launch and sustain these systems and can help nonprofits advocate for these systems in their communities. Finally, qualitative data gathered from in-depth interviews with residents moving in, moving out, or staying in place can provide additional insight into how residential choices are made. The field can do a better job of sharing procedures and protocols to help nonprofits conduct their own interviews or partner with other research organizations to do so.
Even with these data sources, measuring and understanding residential instability is a challenging task, particularly for nonprofits without in-house research capacity. Establishing partnerships with local researchers may be an effective method for more completely understanding this important dynamic. Those conducting community initiatives should do their best to map out the best strategies for data collection given the resources and type of intervention, but the picture will be incomplete in many cases. Community stakeholders can also apply their on-the-ground knowledge and lessons from other studies in the field about how mobility affects low-income households and neighborhoods. Using this mosaic of sources, we can be smarter in designing programs that consider patterns of mobility in our neighborhoods, identifying how positive and negative mobility affects our progress along the way and in tracking and interpreting neighborhood change.
 C. Coulton, B. Theodos, and M.A. Turner, “Residential Mobility and Neighborhood Change: Real Neighborhoods under the Microscope,” Cityscape: A Journal of Policy Development and Research 14 (3) (2012): 55−89. http://www.huduser.org/portal/periodicals/cityscpe/vol14num3/Cityscape_Nov2012_res_mobility_neigh.pdf.
 T. Jelleyman and N. Spencer, “Residential Mobility in Childhood and Health Outcomes: A Systematic Review,” Journal of Epidemiology and Community Health 62 (7) (2008): 584−592.
 Coulton et al., 2012.
 J.D. Morenoff and R.J. Sampson, “Violent Crime and the Spatial Dynamics of Neighborhood Transition: Chicago, 1970–1990,” Social forces 76 (1) (1997): 31−64. Also see R.J. Sampson, S.W. Raudenbush, and F. Earls, “Neighborhoods and Violent Crime: A Multilevel Study of Collective efficacy,” Science 277 (5328) (1997): 918−924.
 R.J. Sampson and P. Sharkey, “Neighborhood Selection and the Social Reproduction of Concentrated Racial Inequality,” Demography 45 (1) (2008): 1−29.
 The Annie E. Casey Foundation’s Making Connections initiative was a decade-long effort focused on target neighborhoods in 10 cities: Denver, Des Moines, Hartford, Indianapolis, Louisville, Milwaukee, Oakland, Providence, San Antonio, and White Center (outside Seattle). The target neighborhoods offer a unique and valuable window to the dynamics of low-income, mostly minority neighborhoods nationwide.
 K. Bachtell, N. English, and C. Haggerty, “Tracking Mobility at the Household Level,” Cityscape: A Journal of Policy Development and Research 14 (3) (2012): 91−114. http://www.huduser.org/portal/periodicals/cityscpe/vol14num3/Cityscape_Nov2012_tracking_mob.pdf.
 The Actionable Intelligence for Social Policy program at the University of Pennsylvania provides support for this work and information on how to develop such systems (see http://www.aisp.upenn.edu/).
 I. Gould Ellen and K. M. O’Regan, “How Low Income Neighborhoods Change: Entry, Exit, and Enhancement,” Regional Science and Urban Economics 41 (2) (2011): 89−97.
 B. Theodos, C. Coulton, and R. Pitingolo, “Neighborhood Stability and Neighborhood Change: A Study of Housing Unit Turnover in Low Income Neighborhoods.” Paper presented at Federal Reserve System Community Development Research Conference (Washington, DC, April 11, 2013). http://www.frbatlanta.org/documents/news/conferences/13resilience_rebuilding_theodos.pdf.
 X. de Souza Briggs et al., “Why Did the Moving to Opportunity Experiment Not Get Young People into Better Schools?” Housing Policy Debate 19 (1) (2008): 53−91.
 S. DeLuca and E. Dayton, “Switching Social Contexts: The Effects of Housing Mobility and School Choice Programs on Youth Outcomes,” Annual Review of Sociology 35 (1) (2009): 457–491.