By Ann Owens (University of Southern California) and Jennifer Candipan (University of Southern California)
When neighborhoods’ socioeconomic status (SES) improves, does their racial/ethnic composition change? Is socioeconomic change also a process of racial/ethnic transition from minority to white? Often when minority neighborhoods are experiencing socioeconomic increases, residents and anti-gentrification activists perceive such a threat—that higher-income, mainly white newcomers will “invade” the neighborhood, potentially displacing residents and altering the neighborhood’s racial/ethnic makeup. For example, as the socioeconomic status of historically black neighborhoods like Harlem in New York or Shaw in Washington, DC, has increased, residents and scholars alike have monitored whether white gentrifiers moved in and altered the neighborhoods’ composition (Freeman 2006; Hyra 2017).
But to what degree does this occur, nationally? We use Census and American Community Survey data on all metropolitan neighborhoods in the U.S. from 1990 to 2010 to document how frequently neighborhood socioeconomic ascent corresponds to drastic racial/ethnic change. We define neighborhood socioeconomic ascent as substantial increases in housing costs and residents’ incomes, education, and occupational levels, relative to other neighborhoods in the same metropolitan area. In the average ascending neighborhood, median income doubled, share of residents with a BA increased by 14 points, share of residents with a white-collar job increased 15 points, and housing costs doubled over this 20-year period. About 25% of metropolitan neighborhoods experienced ascent during our study period.
We classify neighborhoods as predominantly (>75%) white, predominantly (>50%) black, predominantly (>50%) Hispanic, predominantly (>30%) Asian, and mixed race (no predominant racial/ethnic group) in 1990 and 2010 (white, black, and Asian refers to non-Hispanic). We find that transitioning from predominantly non-white to predominantly white during this twenty year period is rare, but it is a process that ascent accelerates. Among neighborhoods that did not ascend, 90% of predominantly black and Asian neighborhoods and 97% of predominantly Hispanic neighborhoods remained so. Among neighborhoods that did ascend, only 71% of black, 82% of Asian, and 81% of Hispanic neighborhoods remained so. On average, the white population share grew in initially majority-black and –Hispanic neighborhoods that ascended, during a period when the overall white population share in the U.S. was declining. If we examine transition to more than 50%, rather than 75%, white, eleven percent of initially majority-black neighborhoods and 6% of initially majority-Hispanic ascendant neighborhoods became majority-white, compared to less than 1% among non-ascendant neighborhoods.
Most of the ascendant predominantly black or Hispanic neighborhoods that changed racial classification became mixed race. This may suggest that ascent can lead to increased racial diversity. However, initially mixed race neighborhoods that ascended during this time were nine times more likely to become predominantly (>75%) white than initially mixed race neighborhoods that did not ascend. Therefore, ascent may contribute to a slow transition from predominantly-minority to mixed race to, ultimately, predominantly white, rather than creating permanently mixed race neighborhoods.
Among white neighborhoods, ascent is a process that shores up the white population—during a period when the white population share in the U.S. declined from 76 to 64%, predominantly white neighborhoods retained a higher white population share, and three-quarters of >75% white ascendant neighborhoods remained >75% white, compared to 58% of non-ascendant neighborhoods.
These patterns are illustrated in this map of part of the Los Angeles metropolitan area. Census tracts with triangles are those that ascended from 1990 to 2010. Tracts are shaded according to the change in their white population share during this time. White population share declined in tracts shaded green or yellow, while white population share increased in tracts shaded pink or purple. The map illustrates that ascent and white population change went hand-in-hand—most ascent occurred in places where white population share increased or decreased more modestly (white population share in the Los Angeles metropolitan area declined 13% during this time, so the lightest yellow-shaded tracts represent areas where the white population declined, but less than average).
We then examined the socioeconomic characteristics of newcomers and long-time residents by race/ethnicity. Are white newcomers with higher socioeconomic status (SES) a key driver of ascent in predominantly non-white neighborhoods? Are non-white higher-SES households also moving in? What about long-time residents’ SES— is it improving?
We find that in predominantly black and Hispanic neighborhoods, ascent occurred as high-SES whites moved in. While black and Hispanic in-movers had modestly higher SES than that of long-term residents, white and Asian/other race in-movers’ SES was one-half to a full standard deviation higher. The racial/ethnic gap in SES in ascendant neighborhoods grows for each successive cohort of in-movers. Ascent in majority-white neighborhoods is also driven by the in-migration of high-SES whites, with little racial/ethnic composition change accompanying SES increases. Increasing SES among long-time residents may contribute to ascent in majority-black neighborhoods, but the highest-SES residents in ascendant majority-black neighborhoods are newly arrived non-black residents.
Together, our results paint a complex portrait of the racial/ethnic demography of neighborhood ascent. On one hand, the modal and average majority-minority neighborhood remained majority-minority as socioeconomic ascent was occurring. On the other hand, ascent is associated with greater gains in white population and declines in minority population than in non-ascendant neighborhoods, and a key driver of ascent in minority neighborhoods is the in-migration of white higher-SES households.
While neighborhood socioeconomic ascent is a process of neighborhood change, it may perpetuate the pre-existing neighborhood racial/ethnic hierarchy. Given historical economic, social, and political inequalities, white neighborhoods are advantaged compared to minority neighborhoods, and when white neighborhoods ascend, they remain white. Ascent provides a pathway for white neighborhoods to increase or cement their advantages. In contrast, when minority neighborhoods ascend, their racial/ethnic composition changes, and the proportion white increases. Ascent may therefore not provide permanent advantages for minority neighborhoods because racial/ethnic change accompanies socioeconomic change—the newly ascendant neighborhood may no longer be majority-minority. Ascent may be a process that reinforces, rather than disrupts, existing racial/ethnic hierarchy among neighborhoods.
Ann Owens, PhD, is associate professor of sociology at the University of Southern California. She earned a PhD in Sociology & Social Policy from Harvard University. Her research investigates the causes and consequences of social inequality, with a focus on neighborhoods, housing, education, and social policy.
Jennifer Candipan, MA, is a doctoral candidate in sociology at the University of Southern California. Her research is broadly interested in social and spatial dimensions of inequalities. Recent work examines links between neighborhood and school composition over time, with a focus on neighborhoods undergoing demographic and socioeconomic change, and how neighborhood and school contexts shape children’s residential mobility and educational outcomes.