Why Don't American Cities Burn?. Michael B. Katz

than native-born residents, stimulating growth in sectors from food manufacturing to health care, creating loads of new jobs and transforming once-sleepy neighborhoods into thriving commercial centers.”²⁷ Immigration also fueled growth and economic revitalization in small cities such as Chelsea and Lawrence, Massachusetts. Two officials of the Federal Reserve Bank of Boston reported that between 1990 and 2000, immigrants accounted for almost half of New England’s population growth, and more in some states, such as Connecticut, where it was responsible for 76 percent of growth. “Overall, the region’s population grew only 5.4 percent over the decade, but without foreign immigration, it would have been virtually stagnant.”²⁸ The New Jersey Urban Revitalization Research Project reported, “Immigration is reshaping many of New Jersey’s older communities, and accounts for the greater part of the population growth of most cities experienced during the 1990s.”²⁹ In metropolitan Philadelphia, immigration was the source of 75 percent of labor force growth between 2000 and 2006.³⁰

      Immigration, Mike Davis observes in Magical Urbanism, redefined urban space. “As emergent Latino pluralities and majorities outgrow the classic barrio,” he writes, “they are remaking urban space in novel ways that cannot be assimilated to the earlier experiences of either African Americans or European immigrants.” These Latino metropolises differ from one another in their “geometries,” which Davis classifies with a provisional typology whose newest and unprecedented category, “city-within-a-city,” represented by late twentieth-century Los Angeles, results from the intersection of immigration with the location of low-wage jobs.³¹

      Immigration, suburbanization, and racial segregation transformed urban space. Suburban growth, which had begun much earlier, exploded in the years after World War II, with suburbs growing ten times faster than cities in the 1950s. Population, retailing, services, and industry all suburbanized. Suburbs remained predominantly white until late in the twentieth century, when African American suburbanization became an important trend, although even in the suburbs African Americans often clustered in segregated neighborhoods or dominated some suburban towns.³²

      The image evoked by the term “suburb” was never accurate. Constructed at various points in history, from the transportation revolution of the nineteenth century to the communications revolution of the late twentieth century, and reconstructed repeatedly by demographic, economic, social, and political change, places labelled “suburb” have always, in fact, varied. Long before World War II, suburbs were industrial as well as residential; they housed working-class as well as middle-class families; and they were home to many African Americans. In the post–World War II era, the massive building of new suburbs like Levittown, highway construction, cheap mortgages, and especially the GI Bill reinforced the popular meaning of “suburb” as a bedroom community populated mainly by families with children. By the last decades of the twentieth century, whatever uniformity had existed among suburbs shattered. A variety of suburban forms dotted metropolitan landscapes as social scientists and regional policy advocates scrambled to create new typologies that would capture the components of the new geography that had rendered the binary of city/suburb obsolete.³³

      Both gentrification and dramatic shifts in the balance among family types resulted in new domestic landscapes, further collapsing differences between city and suburb. Gentrification played modest counterpoint to urban renewal. Gentrification refers to rehabilitating working-class housing for use by a wealthier class. Movement into gentrified neighborhoods was not great enough to reverse overall population decline outside of select neighborhoods, but it did transform visible components of cityscapes as it attracted young white professionals with above-average incomes and empty nesters who demanded new services and amenities.

      Young professionals and affluent empty nesters repopulating center cities signified transformations of family and life course that undermined old assumptions about urbanism by undermining distinctions between cities and suburbs through the creation of new domestic landscapes. Consider the revolutionary rebalancing of family types between 1900 and 2000. In both years, most people lived in one of four combinations of family and house hold type: married couples with children; female-headed house holds with children; empty-nest couples; and non-family house holds (unmarried young people living together). Over the course of the twentieth century, the relative proportions living in each house hold type changed dramatically. In 1900, married couples with children comprised 55 percent of all house holds, single-mother families 28 percent, empty-nest house holds 6 percent, and nonfamily house holds 10 percent, with a small remainder in different arrangements. By 2000, the proportions had changed: married couple house holds comprised 25 percent of all house holds, single-mother families 30 percent, empty- nest house holds 16 percent, and nonfamily house holds 25 percent.³⁴ (The relatively small increase in single- mother families masks an enormous change. Earlier in the century they mainly consisted of widows; late in the century they were mostly never married, separated, or divorced.)

      This new balance among house hold types had accelerated with astonishing speed after 1970. One of its results was a new domestic landscape that changed the meaning of “suburbs.” By 1970, more Americans lived in suburbs than in cities or rural areas. In these early years—captured brilliantly by Herbert Gans in The Levittowners—the suburbs’ primary function was to provide housing for families with children.³⁵ During the last three decades of the twentieth century, suburban demography and function changed, with the result that cities and suburbs grew more alike.³⁶ Between 1970 and 2000, the proportion of suburban census tracts where married couples with children comprised more than half of all house holds plummeted from 59 percent to 12 percent and in central cities from 12 percent to 3 percent. By 2000, the great majority of suburbanites—including those in the Sun Belt as well as the Rust Belt—lived where married couples with children made up a small share of all families. Single mothers replaced many of these traditional families in both suburbs and cities. Between 1970 and 2000, the share of the suburban population living in census tracts where single-mother families made up at least 25 percent of all house holds leaped an amazing 440 percent—from 5 to 27 percent. In cities, it rose 84 percent—from 32 to 59 percent. As suburban populations aged, empty-nest house holds became more common. In suburbs, the share of the population aged sixty-five or older rose from 11 to 16 percent, a 45 percent increase—while it remained virtually the same, 18 percent compared to 17 percent, in central cities. The share of the suburban population living in census tracts where empty-nest house holds comprised more than 45 percent of all house holds shot up from 14 percent to 25 percent, while in central cities it dropped from 30 percent to 21 percent. In central cities, immigration combined with the increase in nonfamily and single- mother house holds to dampen the influence of population aging. Nonfamily households—young, unmarried people between eighteen and thirty-five living alone or without relatives—replaced traditional families in both cities and suburbs. Between 1970 and 2000, the share of the population living in census tracts where nonfamily house holds comprised at least 30 percent of all house holds rocketed from 8 to 35 percent in suburbs and from 28 to 57 percent in cities.

      Figures 1.1 through 1.8 illustrate how these trends remapped domestic space. These maps show the change in the distribution of married-couple-with- children house holds and nonfamily house holds in metropolitan Atlanta and Boston between 1970 and 2000. Despite the differences between these Sun Belt and Rust Belt regions, trends were amazingly similar, showing the near disappearance of suburbs dominated by traditional families and the prominence everywhere of unrelated individuals living together.

      A new domestic landscape emerged from these remapped house hold types. The concentration of young adults and empty nesters redefined urban economic zones. “Gentrification,” in fact, is shorthand for the impact of changing family and house hold forms on urban space. Increased numbers of single mothers living in poverty shaped new districts of concentrated poverty in central cities and fueled a rise in suburban poverty, especially in suburbs that bordered on cities. At the same time, by bringing young, working-class families back to several cities, immigration slowed the disappearance of traditional families and moderated the gulf separating gentrified neighborhoods from vast areas of concentrated poverty. Waves of immigration and industrial change had repeatedly rearranged the social geographies of cities. But the new domestic landscape demanded nothing less than a redefinition of suburban character and purpose. As the distinctions between city