Who Moved and Where Did They Go? An analysis of residential moving patterns in King County, WA between 2002–2017

Download the full report (pdf, 2.12 mb)

Executive Summary

• From 2002-2017, moderate and middle socioeconomic status (SES) King County residents exhibited greater rates of moving compared to the lowest and highest SES residents. Geographically, movement rates were highest in East King County and North Seattle. High-SES residents were the least likely to move during all years.
• The percentage of King County residents in the high-SES group increased by 8.9 percentage points between 2002 and 2017, while the percentages of middle-, moderate- and lower-SES residents decreased (by 3.3, 3.2, and 2.4 percentage points, respectively).
• Neighborhoods in Seattle and East King County have seen the biggest percentage increases in high-SES residents from 2002 to 2017.
• Low-SES residents in all regions were the most likely to move out of the Puget Sound area. High-SES residents were most likely to move within their neighborhood.

Introduction

Context

A healthy and sustainable regional economy allows for residents across the income spectrum to live in good-quality neighborhoods that enable economic opportunity. Counties with less concentrated poverty, less income inequality, better schools, and lower crime rates tend to enable better outcomes for children in low-income families; these phenomena are associated with residential stability (Chetty and Hendren 2018).

Residential stability, or the ability of individuals and families to stay in one neighborhood long-term if they choose to do so, has implications not just for the wellbeing of the individual, but also the community. Residential stability can reinforce family, educational, and neighborhood stability (Evans 2004). High levels of residential instability have been linked to social and health disparities (Jelleyman and Spencer 2008; Sharkey and Sampson 2010), including lack of access to healthcare (Kirby and Kaneda 2006), higher crime rates (Sampson, Raudenbush, and Earls 1997), and poor mental health (Ross, Reynolds, and Geis 2000). Low-income Americans disproportionately experience residential instability. Between 2005 and 2010, half of all United States (US) households below the poverty line moved at least once (Ihrke and Faber 2012; Phinney 2013) and low-income children are almost twice as likely to experience acute residential instability than their wealthier counterparts, moving more than six times before adulthood (Wood et al. 1993).

Residential instability in low-income communities is partly driven by the growing lack of affordable housing across the US. Many low-income residents of the Seattle metro area, including King County, have struggled with housing affordability brought on by rising housing costs, insufficient wages, and other structural factors. These constraints have likely been further exacerbated by the COVID-19 pandemic (PHSKC, 2020). Pre-pandemic roughly a third of King County households paid more than 30% of their household income for housing.i

Community accounts have indicated growing displacement and neighborhood change may be disproportionately affecting low-income residents in the region, but quantitative analyses of moving trends are lacking. Understanding moves prior to the COVID-19 pandemic and through the course of the last recession enables greater understanding of areas and populations vulnerable to additional financial stressors in the face of the recession caused by the COVID-19 pandemic. Rising housing costs, lack of a living wage, and frequent moves contribute to low-income residents being stuck in a cycle of poverty; exposure to poor neighborhoods, regardless of individual and family background, can diminish economic outcomes and social development (Ellen and Turner 1997; Brooks-Gunn, Duncan and Aber 1997; Sampson, Morenoff, and Gannon-Rowley 2002).

Implications of Frequent Moves

Those who are forced to move may face difficulty affording basic needs and dealing with the educational challenges that arise when children switch schools. Movers may also encounter longer and costlier commutes, as well as disruptions to social networks and access to cultural resources. In this way, residential instability impacts not just social relationships (Sampson and Groves 1989; Sampson et al. 1997), but access to opportunities based on a social/communal network

Due to well-documented historical policies and practices that discriminated on the basis of race and ethnicity in the US, socioeconomic status and race/ethnicity are deeply linked, meaning that people of color are more likely to belong to a low-income household and to experience residential instability and the associated harms (Mattiuzzi 2022). This reinforces patterns of segregation in urban areas (Sampson 2008) and has negative implications for future economic prosperity in communities of color. Residential segregation also impacts mental and physical health through the limitation of access to care, healthy food options, and social capital, among other things (Crowder and Krysan 2017.

As more people are ‘priced out’ of their neighborhoods, fewer affordable options may push low- and moderate-SES residents to neighborhoods with fewer opportunities and resources and/or to outer areas of the region or from the Pacific Northwest altogether. Promoting residential stability by building affordable, sustainable, and vibrant communities has wide-ranging potential for the success and wellbeing of King County residents.

Report Roadmap and Research Questions

This report describes residential moves by King County residents within the three-county Seattle metropolitan region in the Puget Sound from 2002 through 2017. Data from large individual-level data sets were analyzed for trends over 15 years to see how people in King County, Washington moved within the region, and compare trends by SES level and neighborhood/geography. Short moves and long-distance moves were examined within King County and to neighboring Snohomish County to the north and Pierce County to the south. Analyses presented in this report offer insight into four questions about residential moves by King County residents:

• Who is moving in King County? We sought to assess whether SES was associated with frequency of moving, whether people at different SES levels move more often during economic boom or bust years, and whether different parts of King County are becoming more economically segregated as a result of these moves.
• How have these moves changed the profile of who lives in King County? Is King County becoming a region of greater SES inequality where fewer middle-SES people live? Does economic segregation vary by the four subregions within King County—the City of Seattle, North King County, the Eastside, and South King County?
• Where are people moving to? How far away do people move? When people move, do they stay in King County, move to a neighboring county, or leave the Seattle metro area? Are these patterns the same for different SES groups?
• When people move within King County, how does the neighborhood they left compare to their new neighborhood on socioeconomic measures? Does this differ by SES?

The analyses add to findings from analyses in other major cities showing that housing unaffordability has created residential instability, that low- and high-SES residents move less than moderate- and middle-SES residents, and low-SES residents are more likely to move out of the neighborhood (in this case, out of the county) than other groups (Hwang and Shrimali 2021; Ding et al. 2016).

Data Sources

This report uses data from the Federal Reserve Bank of New York Consumer Credit Panel (CCP) and from the American Community Survey (ACS).

Federal Reserve Bank of New York Consumer Credit Panel/Equifax Data: This analysis uses quarterly information on a 5% sample of adult consumers from January 2002 to December 2017, with census block group-level information on where individuals live, as well as their age, loans, mortgages, financial issues (e.g., delinquencies, bankruptcy, foreclosure), and Equifax Risk Scores (credit scores that indicate financial stability). These data are used to analyze individuals’ financial health and moving patterns over time. The data is comprised of adult consumers with at least one credit account or collection/public record (such as bankruptcy or foreclosure), as well as those with closed or authorized user accounts. Nearly half of adults who do not have credit scores are represented in the data because they do have credit history, however those without any credit score or credit history are not included in the analysis, so low-SES residents are likely underrepresented.

The following analyses include residents ages 25 to 84 years old. Data from 2004 and are not included because geographic data were inconsistent due to changes in the geocoding procedures by the data vendor. Data points for 2004 are shown as an average of 2003 and 2005. Residents younger than 25 are underrepresented in the data and can have inaccurate address reporting due to moving for reasons related to higher education during this period. Residents older than 84 years old are overrepresented in the data, likely due to a lag in registered deaths in the data.

U.S. Census and American Community Survey (ACS): These publicly available datasets provide information for several variables to characterize neighborhoods people are moving to and from, including health (life expectancy), socioeconomic (poverty), and housing (home value) indicators.ii

Key Definitions and Measures

Socioeconomic Status (SES) Levels: Individual-level measures of SES are defined using Equifax Risk Scores, a credit score that ostensibly reflects the likelihood that an individual will pay their debts without defaulting. These scores are a proxy of financial stability and reflect a distinct dimension of SES from other measures, such as income or wealth, and are particularly relevant to the housing market, where landlords often use credit scores to screen tenants and lenders use credit scores to distribute mortgage products and make lending decisions. We define the SES categories in the following way by their Equifax Risk Scores, which range from 280 to 850iii:

• Low-SES: < 580 or no Score (too few accounts or new credit)
• Moderate-SES: 580-649
• Middle-SES: 650-749
• High-SES: 750 or higher

Housing Periods: The results are separated by four economic housing periods based on market trends from the Standard & Poor Case-Schiller Home Price Indices for WA-Seattleiv (years represent the initial year of each annual sample of the CCP data): 

• Boom: 2002-2006
• Bust: 2007-2011
• Recovery: 2012-2015
• Post-Recovery: 2016-2017

Health Reporting Areas: Results are reported by cities and neighborhoods in King County using Public Health – Seattle & King County-defined “Health Reporting Areas.” These 48 areas were designed in collaboration between Public Health – Seattle & King County, local jurisdictions, and community groups in 2011 to facilitate reporting of health metrics and other related data. Where possible, Health Reporting Areas match local definitions of neighborhoods within large cities, political boundaries of smaller cities, and locally understood places within unincorporated areas of King County. For confidentiality and data reliability, some smaller reporting areas were combined.v A map of Health Reporting Areas is in Attachment B.

Regions: Results are also reported by larger geographic areas—the four subregions within King County. These are North King County, South King County, the City of Seattle, and the Eastside of the county (also called East King County below). These regions are defined either by combined postal ZIP codes or by city-based Health Reporting Areas. The North Region includes the areas of Bothell, Cottage Lake, Kenmore, Lake Forest Park, Shoreline, and Woodinville. South King County region contains Auburn, Burien, Covington, Des Moines, Enumclaw, Federal Way, Kent, Maple Valley, Normandy Park, Renton, Tukwila, SeaTac, White Center/Boulevard Park, and Vashon Island. The Eastside Region includes Bellevue, Carnation, Duvall, Issaquah, Kirkland, Medina, Mercer Island, Newcastle, North Bend, Redmond, Sammamish, and Skykomish.vi

The views expressed in this report are those of the authors and do not necessarily reflect the views of the Federal Reserve Bank of San Francisco or the Federal Reserve System.

Acknowledgments

This report was undertaken as a partnership between the Federal Reserve Bank of San Francisco Community Development staff led by Bina Shrimali, the Changing Cities Research Lab at Stanford University led by Professor of Sociology Jackelyn Hwang, and Public Health – Seattle & King County staff in the Assessment, Policy Development, and Evaluation section. The authors would like to thank Vineet Gupta who provided analyses for this work and Maxine Wright for subject matter input and writing for this report.

End Notes

i. Retrieved from Public Health – Seattle & King County, Community Health Indicators.

ii. Census Bureau, “Glossary,” Block Group, May 10, 2021.

iii. These cutoffs are based on common credit underwriting thresholds for mortgage products. Separate analysis of the distribution of residents in the San Francisco Bay Area by these SES categories are similar to the distribution of adult residents in the following income categories, respectively: < 50% of the US median household income; between 50%-100% US median household income; between 100-200% of the US median household income; and 200% of the median household income. These categories are not directly comparable to the U.S. Department of Housing and Urban Development (HUD) Area Median Income (AMI) categorizations, which are based on metropolitan area, family size, and income. The Changing Cities Research Lab’s analysis of population distributions using data from the Comprehensive Housing Affordability Strategy (CHAS) for the City of Oakland show that the SES categories are similar to the following HUD AMI categories, respectively: <30% AMI (“extremely low”, as labeled by the States of California and Washington), between 30% and 50% AMI (“very low”), between 50% and 100% AMI (“low” and “moderate”), and above 100% AMI (“high”). Seattle and Oakland have similar AMI profiles: Oakland’s 2019 30% AMI was $26,040, while Seattle’s was $24,300.

iv. Federal Reserve Bank of St. Louis, FRED S&P/Case-Shiller WA-Seattle Home Price Index.

v. Public Health – Seattle & King County, “Health Reporting Areas,” May 10, 2021.

vi. Public Health – Seattle & King County, “Region,” May 10, 2021.