Over the next generation or two, America’s older, largely white population will increasingly be replaced by today’s disproportionately poor minority children. All future growth will come from populations other than non-Hispanic whites as America moves toward a majority-minority society by 2043. This so-called Third Demographic Transition raises important implications about changing racial boundaries in the United States, that is, about the physical, economic, and sociocultural barriers that separate different racial and ethnic groups. America’s racial transformation may place upward demographic pressure on future poverty and inequality as today’s disproportionately poor and minority children grow into adult roles. Racial boundaries will be reshaped by the changing meaning of race and ethnicity, shifting patterns of racial segregation in neighborhoods and the workplace, newly integrating (or not) friendship networks, and changing rates of interracial marriage and childbearing. The empirical literature provides complicated lessons and offers few guarantees that growing racial diversity will lead to a corresponding breakdown in racial boundaries—that whites and minorities will increasingly share the same physical and social spaces or interact as coequals. How America’s older population of elected officials and taxpayers responds today to America’s increasingly diverse population will provide a window to the future, when today’s children successfully transition (or not) into productive adult roles. Racial and ethnic inclusion will be reshaped by changing ethnoracial inequality, which highlights the need to invest in children—now.
The United States may be the most demographically diverse country among the world’s Western democracies. America’s population diversity is revealed along many different dimensions that sometimes divide us, including age, class, religion, nativity (and language), politics, sexual orientation, geography, and, of course, racial and ethnic background, which animates the social, economic, and political implications of virtually every aspect of diversity. Growing population diversity—and the myriad ways it is expressed every day in America—will be driven by ongoing demographic processes and shifting ethnoracial boundaries. To be sure, America is a land of racial and ethnic immigrants with diverse sociocultural and economic backgrounds (Hirschman 2005; Tienda 2002). The nation has moved well beyond the “melting pot” metaphor. We have instead embraced a new multiculturalism, while also acknowledging the possibility of segmented assimilation, transnationalism, and downward mobility (among some groups) into a minority underclass. Assimilation into a white majority is neither inevitable nor always desirable (Alba and Nee 2003; Portes and Zhou 1993).
A concern today is that racial and ethnic diversity—which is often celebrated in anticipation of achieving a new postracial society—may instead be a source of growing political conflict, cultural disunity, and loss of community or cohesion. The politics of racial diversity may grow in importance rather than dissipate over the foreseeable future as the United States moves toward becoming a majority-minority society as early as 2043 (U.S. Census Bureau 2012a).
Is America fragmenting?1 This is a contentious question that politicians and pundits often raise today (e.g., Buchanan 2011; Huffington 2011; Murray 2012), but also one that has been common historically in America’s social and political discourse, especially during periods of social and political upheaval and rapid immigration (see Fischer and Mattson 2009; Hollinger 2008; Portes and Vickstrom 2011). Now is one of those times.
Indeed, current and future demographic shifts pose extraordinary challenges for America that cannot be ignored or easily dismissed. Growing racial diversity may provide opportunities for a more integrated future (Alba and Nee 2003; Hochschild et al. 2012; Lee and Bean 2010), but this will not come easily (Bobo 2011; Hero 2007). The failure to effectively address questions of persistent racial and ethnic economic inequality, immigration (especially undocumented arrivals) and immigrant incorporation, and racial and ethnic geographic balkanization may reveal new and unanticipated patterns of cultural and economic fragmentation. As the future unfolds, how will the American people—individually and collectively—respond to the nation’s rapidly changing racial and ethnic mix? Will racial boundaries be redrawn and persistent racial and ethnic inequality decline over the foreseeable future? Will America become “one nation” with a common national and cultural identity? Will it become a postracial society, where race no longer matters? Or will America become something else?
This article highlights the demographic parameters of current and future racial and ethnic change in America. I begin by presenting evidence of a “Third Demographic Transition” marked by unprecedented changes in America’s racial and ethnic makeup over the next 40 years.2 Racial diversity will be driven by new immigration, but it also will be shaped by minority fertility and white natural decrease, which I argue will give demographic impetus to higher poverty rates and more inequality in the future. Diversity begins with children—from the “bottom up.” Over the next generation or two, an older, largely white and affluent population will be increasingly replaced by today’s disproportionately poor minority children, who will reshape America’s future and its place in a globalizing economy.
The social and political implications of growing racial and ethnic diversity will ultimately depend on changing race relations and the breakdown of racial boundaries that have been resilient to change. Racial divisions have been shaped historically by limited opportunities for interracial interaction in neighborhoods, schools, and the workplace and by in-group solidarity (as revealed in racially homophilious relationships), which are often reinforced by racism and anti-immigrant hostility (Massey and Sánchez 2010). Indeed, as I illustrate here, growing racial diversity could “brighten” rather than redraw current racial boundaries or create a new racial order (e.g., Frank et al. 2010; Hochschild et al. 2012). Growing diversity over the foreseeable future is no guarantee that people of different cultural backgrounds will share the same physical and social spaces. New opportunities for intergroup contact, if measured by declining racial segregation, may also be offset by growing in-group racial preferences in social networks, friendship cliques, and interracial marriage.
In the end, opportunities and preferences for interracial contact and affiliation will depend heavily on the prospect of upward socioeconomic mobility among today’s minority children—the next generation (Alba et al. 2011). Here, I highlight several emerging threats to minority socioeconomic mobility and to harmonious race relations in the future. How America responds now to the new challenges of racial and ethnic diversity will determine whether it becomes a more open and inclusive society in the future—one that provides equal opportunities and justice for all. This is a demographic “pipeline” issue that will not go away anytime soon and that will reshape racial politics over the foreseeable future.
America’s Racial Future: A Third Demographic Transition?
The Third Demographic Transition, as coined by David Coleman (2006), refers to the societal transformation from a low-fertility, native-born majority population to a high-fertility, racial and ethnic immigrant population. This population transition is general (at least across today’s developed world, but especially in Europe) and it is irreversible (Coleman 2009).3 To claim that the United States is in the throes of a Third Demographic Transition, we must start with some basic demographic facts, if only to reinforce casual observations or impressions about growing racial diversity in America.
The 2010 U.S. decennial census revealed that the percentage of non-Hispanic whites decreased from 69.1 % to 63.7 % between 2000 and 2010 (Humes et al. 2011).4 More significantly, newly released U.S. Census Bureau projections (based on the 2010 census) indicate that the non-Hispanic white population is expected to decline from roughly 197 million in 2010 to slightly more than 186 million in 2050, as white natural decrease takes a demographic grip on America’s future (U.S. Census Bureau 2012a). America’s minority populations, in contrast, will nearly double in size over the next 40 years, increasing from 112 million to 213 million persons if recent patterns of immigration, fertility, and mortality continue as they have (U.S. Census Bureau 2012b). All of the projected U.S. population increase will come from groups other than non-Hispanic whites. The United States is moving inexorably toward becoming a majority-minority society. Indeed, only 46.6 % of the U.S. population will be classified as white in 2050 if current demographic trends continue. The racial crossover occurs in 2043 (see Fig. 1).
Projected increases in the racial and ethnic minority population may give demographic impetus to more poverty and inequality in the future. This can be effectively demonstrated with a simple accounting exercise. Applying age-race-sex–specific poverty rates (from the 2010 American Community Survey) to projected age-race-sex populations in 2050 yields projected poverty rates in 2050 (see Murdock et al. 2003 for similar approach). As shown in Table 1, this exercise implies a 2050 poverty rate of 16.6 %, which compares unfavorably with the published poverty rate of 15.1 % in 2010 (DeNavas-Walt et al. 2011).
Of course, projected poverty rates in 2050 reflect not only the influences of racial composition but also changes in the age-sex composition (e.g., an older population will have lower overall rates of poverty). Yet, as shown in Table 1, projected poverty rates in 2050 exceed the poverty rate in 2010 for all groups—children, prime-age adults, and seniors—a fact that mostly reflects shifts in the racial and ethnic composition only. These projected increases, although seemingly small (i.e., representing roughly a 10 % increase in poverty), nevertheless translate into almost 5 million more poor people in 2050 than if poverty rates stayed at 2010 levels. And they also tell us that over 70 % of America’s poor in 2050 will be minorities—a fact that has its own political consequences, if measured in public support for the poor (Gilens 2000).
Of course, future age-race-sex–specific poverty rates may not stay at 2010 levels; they may increase or decrease unpredictably over time for different segments of the U.S. population.5 Yet, looking back 35 years rather than forward, the child poverty rates in 1976 were 40.6 % and 30.2 % for blacks and Hispanics, respectively. These figures are little different from 2010 poverty estimates (i.e., 39.0 % for blacks and 34.9 % for Hispanics; U.S. Census Bureau 2012c). Of course, race categories are highly fluid (Perez and Hirschman 2009; Waters 1990). Racial self-definitions are often situational and can change over time and across generations (Duncan and Trejo 2007), which potentially has uneven implications for poverty measurement across ethnoracial groups. Immigration and immigration policy also may change, perhaps unexpectedly, and America’s economic and job outlook is impossible to forecast accurately over the long term. Thus, my point here is largely a heuristic one—that a disproportionately white, elderly, and nonpoor population will be replaced by a disproportionately nonwhite, non-elderly, and poor population. Projected changes in America’s racial composition may mechanically drive up U.S. poverty rates unless age-race-sex–specific poverty rates decline or racial and ethnic differences in poverty are reduced or eliminated. This demographic and economic forecast can be avoided, but it will require the political will to act now to support America’s children and minority families.
Population Processes and Ethnoracial Diversity
Immigration and Diversity
The conventional view today is that immigration is driving America’s growing ethnoracial diversity. This is understandable in light of the massive new immigration into the United States since 1980. Net international immigration into the United States accounted for 7.5 million new residents or 31.3 % of the nation’s total population growth of 23.9 million people between 2000 and 2010 (U.S. Census Bureau 2012d). The United States has more foreign-born residents (about 39 million) than any other country in the world; more than 4,000 new immigrants (legal and unauthorized) arrive in the United States each day (Martin and Midgley 2010). The overwhelming share of new immigrants and the foreign-born population are racial and ethnic minorities, mostly from Asia, Mexico, and other parts of Latin America. Moreover, the secondary migration of racial and ethnic minorities has hastened the pace of racial and ethnic change in cities, suburbs, and small towns across America (Kritz et al. 2011; Lee et al. 2012; Lichter and Johnson 2009; Massey 2008). The recent spatial diffusion of Hispanics to “new destinations” and Asians to (mostly) metropolitan “ethnoburbs” has reinforced the view that new immigration has altered America’s fundamental character. Immigration has become a national political issue rather than just a state or local one.
The big picture of widespread and growing diversity, however, masks a more fundamental demographic reality: changes in America’s racial mix have been observed first and disproportionately among the nation’s children and youth (Johnson and Lichter 2008, 2010). A large but often unappreciated second-order demographic effect of recent immigration is the changing racial mix of newborns and children. The children of America’s racial minority and immigrant populations—the new second generation—will be the lifeblood of America’s economic and political future. They are in the vanguard of the Third Demographic Transition that will remake America.
Majority-Minority Children: Diversity From the Bottom Up
For America’s children, the future is now. Recently released estimates from the U.S. Census Bureau showed that minorities—that is, populations other than non-Hispanic whites—accounted for the majority (50.4 %) of the U.S. population under age 1 in 2011 (U.S. Census Bureau 2012e).6 In 2010, fertility among non-Hispanic whites (1.79) and Asians (1.69) was well below replacement; rates among non-Hispanic blacks (1.97) were much closer to replacement levels (Martin et al. 2012). For Hispanics, however, the total fertility rate was well above replacement (2.35), especially among Central and South Americans (2.98). If current period rates continue indefinitely, the U.S. white population will experience natural decrease—deaths exceeding births—along with absolute population decline in the absence of more white immigration. White natural decrease will exacerbate the downward share of the current white majority, especially as population aging increases rapidly over the next decade, when the numerically large baby boom cohort moves into retirement and dies off.7 America’s children have become much more diverse than the elderly population.8
The potential long-term demographic implications of today’s diversity at the bottom of the U.S. age distribution is revealed in Fig. 2, which gives the age-sex pyramids of projected majority and minority populations in 2050 (see U.S. Census Bureau 2012b for a description of the data). These pyramids show rather extraordinary minority population surpluses at younger ages, especially among children, along with a majority white population among those aged 55 or older. Even if the pipeline of immigration slows to a trickle, ethnoracial diversity will continue to grow through the large second-order effects of past and current immigration: high fertility among Hispanics, below-replacement fertility among white women (who are declining in population size), and population aging, especially as the mostly white baby boom generation dies off.
In his seminal presidential address to the Population Association of America nearly 30 years ago, Samuel Preston (1984) documented the different trajectories of economic well-being among America’s younger and older dependents over the 1970–1982 period. At the time, the elderly had benefited enormously from governmental responses to high rates of poverty through Social Security and Medicare. America’s children were not so fortunate. Only 38 % of U.S. voters at the time were living with a child and could represent children’s interests through the ballot box (Preston 1984:452). Today, this percentage is even lower, and the racial profile of the U.S. voting-age population is much different for the electorate (i.e., the voting-eligible population) than for the population that actually votes. Perhaps not surprisingly, the poverty gap between young and old today is now larger than ever (Fig. 3). In 2010, the age poverty gap was 13.0 percentage points (22.0 % among the elderly over age 65 vs. 9.0 % among children under age 18) compared with 7.3 percentage points (21.9 % vs. 14.6 %) in 1982, at the time of Preston’s analysis.
Two questions concluded Preston’s (1984) address: (1) do we care collectively about our children, and (2) how should we proceed as a society if we do? Both questions have large racial dimensions today. The concerns are obvious: will a shifting age-race profile of America’s population lead to a new generational divide between a slow-growing, older, non-Hispanic white population and fast-growing, younger, minority populations? Do older white people care about “other” people’s children? The answer—and a glimpse of the future—is contained in a recent study by Figlio and Fletcher (2012), which showed that population aging in America’s suburbs is causally linked to cutbacks in local school funding. Cutbacks were especially large if there was a racial mismatch between the elderly and the school-aged population (see also Poterba 1997).
What does the bottom-up nature of America’s racial and ethnic transformation mean for America’s investments in the future? The answer depends on who is bearing and rearing today’s children, which can be found in data from the 2006 to 2010 microdata files of the American Community Survey (ACS). The ACS asks this question of women aged 15–50: “Has this person given birth to any children in the last 12 months?”9 Past-year childbearing can be linked directly to households, including the social and demographic characteristics of the coresidential mothers. As shown in Fig. 4, newborns today are overrepresented among poor families, which have the least educated mothers who are often unmarried (Sanders and Lichter 2012).
Indeed, these data show that 23 % of America’s babies were born to poor women. Although not shown here, the share is much higher (57 %) among babies of racial or ethnic minority mothers. Nearly one-third were born to single mothers (measured at the time of the interview), and almost 40 % of newborns had mothers with low education (i.e., a high school diploma or less), while one-fifth were born to foreign-born mothers. Overall, nearly 40 % of America’s babies today had mothers with two or more disadvantages, and over 60 % had at least one disadvantage. Among the nonwhite minority population, almost 80 % had at least one disadvantage, while 55 % had multiple disadvantages.
Ethnoracial diversity in the future clearly is rooted in current fertility. But perhaps more sobering, today’s fertility patterns are driving current and future poverty and economic inequality. In life’s race, America’s minority children today begin well behind the starting line. Now is not the time for cutbacks in investments in America’s children.
Racial and Ethnic Boundaries: Now and in the Future
Racial Diversity and Demodystopia
Projected changes in racial and ethnic composition, especially among the young, will reshape America’s future. Exactly how is unclear, but it will almost certainly depend on whether racial relations improve as today’s children grow into adulthood. In the extreme, some pundits (e.g., Buchanan 2011), politicians (who exploit white fears), and even some scholars (e.g., Huntington 2004) are concerned that the United States may become a “third-world nation” or that it is committing “race suicide” unless the influx of racially or culturally different groups is curtailed significantly or stopped altogether. According to this view, U.S. society may be rocked in the future by racial vitriol and cultural disunity.
In demography, there is a tendency to make alarmist claims of impending “demodystopia”—demographic forecasts of future “demographic hell” or imminent catastrophe (Domingo 2008). This has been evident in past debates about overpopulation and the “limits to growth” and more recently when drawing out the implications of lowest-low fertility in Europe and East Asia (for discussion, see Lam 2011; Morgan 2003). But demographers have generally steered clear of discussions of demodystopia tied to growing racial and cultural diversity; instead, many (but not all) give a rather hopeful vision of future minority integration and accommodation.10 For example, cohort succession presumably will mean that less racially receptive segments of the U.S. population (i.e., older whites) will be replaced with more racially tolerant populations (including young minorities).11 Jennifer Lee and Frank Bean (2010) coined the term “diversity paradox”; they argued that growing diversity and sociocultural changes—ushered in by new immigration—are now breaking down ethnoracial boundaries and buffering persistent black-white antagonisms (see also Hochschild et al. 2012). Shifting racial boundaries also are likely to occur first among America’s children and youth. In fact, polling data confirm that young people today are much less race-sensitive than America’s older populations.12 Ongoing demographic processes of immigration and cohort replacement may thus diminish the relevance of racial boundaries in the future.
Richard Alba (2009:162) also argued that the next several decades offer an “extraordinary opportunity for minority mobility.” America may be entering a period of non-zero-sum mobility, where the upward mobility of minority populations does not come at the expense of the majority white population. The impending retirement of America’s baby boomers may create a vacancy chain of job opportunities that benefits historically disadvantaged groups. Economic threats to “white privilege” will be less pronounced than in the past, and the benefits will redound to America’s minorities, especially today’s disproportionately nonwhite children. The key, according to Alba (2009), will be to effectively convert their upward socioeconomic mobility into greater proximity to whites in various arenas of social interaction (e.g., neighborhoods).
Such hopeful forecasts beg a simple question: what does the recent empirical social science literature tell us about the association between rising racial diversity—in communities, neighborhoods, workplaces, and organizations—and race relations? Do race relations improve or not? By racial relations, I mean the behavioral or institutional mechanisms that create, reinforce, or eliminate boundaries that divide us. Many scholars draw hope from the historical record—from the erosion of large sociocultural and economic differences among America’s white ethnics (of different national origins) over the course of the twentieth century. The optimistic view is that cultural, political, and economic fracturing along racial and ethnic lines may similarly decline in the future, even as racial diversity increases. After all, America recently elected its first African American president in 2008, who was then reelected in 2012 with the help of a diverse coalition of racial and ethnic minorities (including an emergent block of Hispanic voters). Clearly, along many economic and political dimensions, America has made demonstrable racial progress since the Civil Rights period of the 1960s (Bobo and Charles 2009; Farley 1988).
Yet, social boundaries of all kinds separate one group from another, insiders from outsiders, and us from them (Tilly 2004; Wimmer 2008). Boundaries, as an organizing concept (Lamont and Molnár 2002; Sanders 2002), have been especially useful for studying racial and ethnic divisions (Alba 2009; Massey and Sánchez 2010). In the future, as in the past, racial boundaries presumably can shift, be crossed, or blur (Alba and Nee 2003). Boundary shifts occur when entire groups are redefined racially. For example, a century ago, the Irish, Italians, and Jews were redefined as “white” with generational succession and assimilation (Alba 1985; Hout and Goldstein 1994). Some scholars have argued that Hispanics today are not unlike Italians from the early 1900s (Perlman 2005).13 Racial boundaries also can be crossed, which usually means that racial and ethnic minorities, by virtue of improvements in socioeconomic status (e.g., increasing education or income), are able to operate on either side of the racial divide. Upwardly mobile minorities, students in racially integrated schools, and interracial couples serve as associational brokers that bridge majority and minority populations and help break down racial boundaries and promote mutual understanding. Finally, boundary blurring means that racial boundaries have become less distinct or “bright” (Alba 2005); racial identification can also be dimmed or lost from generation to generation (Duncan and Trejo 2011). Mixed-race individuals, for example, often occupy a unique place in the racial hierarchy. They serve as racial ambassadors who stitch together different racial and ethnic segments of the populations, although they also may be marginalized by both majority and minority populations (for a critique, see Spencer 2011).
Better understanding America’s racial future requires giving attention to two key questions: (1) how have racial boundaries changed—shifted, crossed, or blurred—over the past decade or two, and (2) how are boundaries likely to reveal themselves in the future, especially as today’s children lead us toward a majority-minority society? From a demographic perspective, two indicators of changing racial boundaries seem especially relevant: racial segregation and racial homophily, which represent opportunities and preferences, respectively, for intergroup contact.
Opportunity: Changing Residential Segregation
Spatial boundaries mark arenas for social interaction. A century ago, Robert Park (1915:608) asked whether Chicago’s burgeoning ethnic immigrant populations would be revealed spatially in “a mosaic of little worlds that touch but do not interpenetrate.” In the future, projected increases in America’s racial minority populations will be played out differently across geographic space, which will shape opportunities for interracial interactions of all sorts—in neighborhoods, schools, community organizations, and work settings. Spatial proximity or propinquity is a necessary condition for intergroup contact and mutual understanding among racial groups (Putnam 2007).14
Some optimism about the future is usually drawn from conventional demographic analyses showing long-term declines in neighborhood racial and ethnic segregation, especially among blacks. A recent Manhattan Institute report (of black/nonblack segregation) by economists Edward Glaeser and Jacob Vigdor (2012) is boldly titled The End of the Segregated Century. Others researchers are less sanguine. In a report titled The Persistence of Segregation in the Metropolis: New Findings From the 2010 Census, sociologists John Logan and Brian Stults (2011) showed that black-white neighborhood segregation (measured with the index of dissimilarity) declined from an average of 79 in 1970 to 59 in 2010 in the 50 largest U.S. metropolitan areas. For 2010, this means that 59 % of blacks must move to other neighborhoods (i.e., disproportionately white census tracts) to achieve equality in the percentage distribution of blacks and whites across all neighborhoods. These recent downward trends in residential segregation presumably reflect the enforcement of fair housing legislation, changing racial attitudes, rapid suburbanization of minority populations, and the demolition of public housing projects that warehoused poor minorities in the past (for review, see Charles 2003).
Yet, despite these declines, black-white segregation remains very high. The average black person lives in a neighborhood that is roughly 35 % white. And there has been little or no change or perhaps even small increases in segregation among the nation’s fastest growing populations of Asians and Hispanics. With America’s growing diversity, there seems to be a pause in previously declining rates of neighborhood segregation. My general concern is straightforward: as today’s children transition into adulthood over the next decade or two, conventional demographic approaches to segregation (i.e., those based on analyses of census tracts in big cities alone) may misrepresent the extent of racial residential segregation and its many causes.
Specifically, racial residential segregation in the future may be increasingly characterized by (1) emerging patterns of macro-segregation at higher levels of spatial aggregation (such as places) than the neighborhood; (2) growing segregation between and within different racial and ethnic groups; and (3) changing “causes” of segregation that may increasingly reflect race-based residential preferences rather than income inequality or housing discrimination. I consider each point in turn below.
First, racial segregation may be occurring at different spatial scales than in the past (Fischer et al. 2004; Lichter et al. 2007b; Massey and Fischer 2003; Parisi et al. 2011). Over the recent period of declining neighborhood segregation, most big cities shifted rapidly from being majority-white to becoming majority-minority populations. In fact, most of the 100 largest U.S. primary cities (of metropolitan areas) now have majority-minority populations, increasing from 43 in 2000 to 58 in 2010 (Frey 2011). For example, Detroit today is more than 80 % black, up from 62 % in 1980. Black-white neighborhood segregation, measured by the index of dissimilarity, nonetheless declined from 83.0 to 79.6 between 1980 and 2010 (Logan and Stults 2011). Blacks may increasingly be living in the same neighborhoods as whites (Iceland et al. 2013), but in Detroit, as in many other cities, the population is becoming increasingly composed of racial and ethnic minorities.
Suburban and rural communities with fast-growing Hispanic populations also are being rapidly transformed into majority-minority populations while also experiencing residential segregation that rivals the high level of black-white segregation in cities (Hall 2013; Lichter et al. 2010). As shown in Table 2, U.S. decennial census data reveal remarkable increases—more than a doubling—in the number of majority-minority rural communities over the past two decades, from 757 in 1990 to 1,760 in 2010 (Lichter 2012a). Processes of racial segregation have taken an unexpected turn. The singular focus on declining big-city neighborhood racial segregation as an indicator of changing racial boundaries is incomplete at best and misleading at worst (Lichter et al. 2007b). In the future, big-city neighborhood segregation may persist at high levels but may represent a declining share of overall racial residential differences in America (Parisi et al. 2011). Macro-segregation may be on the rise, especially as a new political economy of place more effectively excludes undesirable groups (e.g., the poor) (Lichter et al. 2012; Massey et al. 2009).
A second concern is that conventional approaches, based on pairwise comparisons of the segregation index, also potentially distort evidence of racial integration in our multiracial future. This point is not new but is often forgotten or ignored. Asians, Hispanics, and blacks are separated not only from whites but from each other (Fong and Shibuya 2005; Jimenéz 2010; Parisi et al. 2011); different national-origin groups also live apart from each other, even those who share the same panethnicity (Kim and White 2010). And racial minorities represent diverse populations with their own divisions by ethnicity, national origin, nativity, and social class (Hirschman et al. 2000; Landale and Oropesa 2007).
To better address issues of neighborhood diversity, John Logan and Wenquan Zhang (2011) identified global neighborhoods—census tracts with significant percentages of whites, blacks, Hispanics, and Asians. Global neighborhoods increased in number from 2,998 to 5,085 between 1980 and 2010 while also accounting for a growing percentage of the metropolitan population. But the number of neighborhoods with significant numbers of whites, Hispanics, and Asians declined absolutely from 3,802 in 1980 to 2,027 in 2010, while their percentage of the entire metropolitan population declined from 28.6 % to 15.2 % (Logan and Zhang 2011). More significantly, the number of all-minority tracts also nearly doubled, increasing from 1,763 in 1980 to 3,238 in 2010. Logan and Zhang (2011) found little evidence of whites moving into all-minority neighborhoods. They concluded that “there is no route to a fully integrated metropolis” because “white flight continues to create new all-minority neighborhoods” (p. 12).
Most racial and ethnic groups today remain highly segregated from each other. A comparatively small number of neighborhoods, cities, or counties are composed of significant shares of more than one race, even though there are few neighborhoods and places today with all-white populations (Johnson and Lichter 2010; Logan and Stults 2011). Moreover, multiracial neighborhoods are often unstable or “in transition” to becoming racially homogenous neighborhoods (Friedman 2008). And even if different minority populations increasingly live in the same neighborhoods (Farrell and Lee 2011), segregation within census tracts (micro-segregation) often remains large across blocks or block groups (Friedman 2012; Lee et al. 2008). Even when occupancy changes at the housing-unit level, the race of the occupant often remains unchanged (Hipp 2012). If racial integration is a public policy goal (and I understand that this point is debatable), recent empirical evidence is not always reassuring. Past and current research on blacks has appropriately focused on neighborhood “tipping points” that spurred white flight; growing diversity led to a white demographic response. Now, Crowder et al. (2011) have shown a new kind of white flight from immigrants who have moved into America’s suburbs over the past decade or two. What is the case for something different in the future as America diversifies racially?
Under the circumstances, it is hard to imagine a demographic, economic, or political scenario that will lead to the return of middle-class whites to “all-minority” neighborhoods in Detroit, Los Angeles, Miami, New York City (e.g., Queens and The Bronx), or other racially diverse cities or suburbs in sufficient numbers to end the ongoing replacement of white populations. The implausibility of such a demographic scenario arguably is heightened by impending white natural decrease and declines in the absolute numbers of geographically mobile white young adults who will be demographically available to buy homes and who are perhaps most receptive to racial residential integration.
Finally, racial segregation may be increasingly shaped by residential preferences and choices (see Bruch and Mare 2006; Clark 2009; Clark and Fossett 2008; Fossett 2006) rather than by income or discrimination only. Income or education disparities currently account for a surprisingly small share of the large racial differences in residence patterns (Iceland 2009), an empirical fact that has shifted the emphasis to past or current racial discrimination in housing markets or to racially insensitive federal housing policy (Massey 1996; Massey and Denton 1993).15 But, even here, audit studies now indicate that housing discrimination is less blatant (but perhaps less easily detected) than in the past (Roscigno et al. 2009; Ross and Turner 2005). The legacy of past racial housing discrimination and sometimes misguided government housing programs nevertheless persists today in the form of continuing high rates of racial residential segregation (Iceland 2009). In the past, blacks of all income levels were steered into black neighborhoods, where they either stayed or circulated among similar neighborhoods. What will happen when pre-fair-housing black cohorts retire, die, and sell their homes? Who will buy them? If racial income inequality declines and housing discrimination matters less than the past, racial preferences—choice with income constraints—will increasingly dominate patterns of racial residential location among tomorrow’s homebuyers (i.e., today’s children).
On this question, the current literature on racial preferences again provides cautionary lessons about nascent racial residential integration. For starters, whites often prefer white neighborhoods and avoid minority-dominated neighborhoods; they are influenced by racial stereotypes and other race-based concerns (Krysan et al. 2009). White residential preferences continue to be negatively shaped by rising shares of minorities, both blacks and immigrants, living in the neighborhood (Harris 2001; Krysan 2002). More generally, white Americans hold more negative racial attitudes (the “threat hypothesis”) if they live in neighborhoods or communities with larger minority shares (for discussion, see Dixon 2006; Taylor and Mateyka 2011). Like whites, blacks and Hispanics also prefer neighbors with the same ethnoracial backgrounds, although there is some evidence that higher-income blacks increasingly prefer more racially integrated suburban communities (Clark 2007), even as many suburbs have experienced unprecedented increases in poverty (Cooke 2010). Under the circumstances, growing racial diversity nationally is no guarantee that racial and ethnic boundaries will be redrawn in communities and neighborhoods (or schools). In fact, black young adults who escape the segregated neighborhoods of their childhoods often revert to racially segregated neighborhoods as they grow older (Sharkey 2012).
To sum up, the pace of decline in segregation, at least as traditionally defined, has slowed over the past decade or two. At the same time, we have seen little if any decline in majority-minority segregation among Hispanics or Asians, while other dimensions of segregation (e.g., macro-segregation, multiracial segregation, and residential preferences) have become more salient. Significantly, America’s children, who are increasingly minority children, often live in neighborhoods and communities that are more segregated and poorer than the general or adult population (Iceland et al. 2010; Johnson and Lichter 2010). The implications of residential segregation among children and youth will surely be played out in patterns of social interaction (e.g., in schools and friendship networks) and in interracial dating, cohabiting, and marriage (Joyner and Kao 2005; McClintock 2010; Mouw and Entwisle 2006).
Preferences: Changing Patterns of Homophily
Homophily is the tendency for individuals to affiliate socially with similar others (McPherson et al. 2001). Homophily reflects selection into associations (i.e., similar people are attracted to each other), socialization (i.e., peer influences on malleable characteristics, such attitudes), and attrition (i.e., dissimilar people disengage from each other over time). In the case of race, as with other characteristics, homophily reinforces racial identity and group boundaries (Wimmer and Lewis 2010). Cross-race social contacts in friendship networks, work settings, and marriage have the potential to reduce racial prejudice (i.e., the contact hypothesis; see Allport 1954), promote greater access of minorities to majority resources and social capital, and function as “bridging ties” that promote greater cohesion between racial groups (Rude and Herda 2010).
McPherson et al. (2001) claimed that “[r]ace and ethnicity are clearly the biggest divide in social networks today in the United States” (p. 420). More importantly, they argued that homophily ultimately is influenced most by structural opportunities for interracial contact, which they label as baseline homophily. Racial homophily is positively associated with the relative population size of groups, which, perhaps paradoxically, implies that increases in the relative size of minority populations in the future may promote more (baseline) homophily among racial minorities rather than less (see Blau 1977; Laumann 1973). Of course, relative group size is played out differently across geographic and organizational space. Opportunities for interaction do not ensure affiliation; in-group preferences for interaction also matter.
Baseline homophily is often reinforced by so-called inbreeding homophily (McPherson et al. 2001), which occurs when demographic groups, including racial groups, share the same social and economic characteristics (e.g., values, class, education, income) that are themselves a basis for social homophily. Inbreeding homophily reduces the likelihood of cross-cutting social circles between (racial) groups; the cross-cutting social circles can promote cohesion and the breakdown of intergroup barriers. To illustrate, whites and blacks with similar educational levels are more likely to interact with one another than are those with dissimilar schooling. The implication is straightforward: as long as racial stratification or other salient characteristics continue to sort whites and racial minorities into different social groups, racial homophily will persist. Racial homophily also is amplified by “balancing mechanisms”; friendships are reciprocated, and friends of friends befriend one another (Mouw and Entwisle 2006; Wimmer and Lewis 2010).
One view is that inbreeding racial homophily will decline in the future with declining social, economic, and political inequality and more acculturation (i.e., groups share the same cultural values). Cross-cutting social circles among racial groups presumably will proliferate while engendering greater social cohesion and mutual understanding (Blau 1977; Blau et al. 1984). Many empirical studies, however, have shown that racially different friends and coworkers may form weaker rather than stronger social bonds, and generate less trust and lower attachment to the group or network. For example, Stainback and Irvin (2012) found that racial minorities were less attached to organizations with fewer coethnics; they also perceived more discrimination against them, which mediated the relationship between minority group size and organizational attachment. Other studies show that racial similarity in friendship dyads are linked to stability. Mixed-race friendships are more likely than same-race friendships to break up, a process that reinforces racial homophily (Rude and Herda 2010).
It is difficult to forecast future patterns of racial homophily. A well-documented literature on cross-race friendship dyads and networks among adolescents and the school-aged population (Goodreau et al. 2009; Wimmer and Lewis 2010) provides little hard evidence on temporal or life cycle changes in racial homophily. Racial differences in homophily are nevertheless large. For example, McPherson et al. (2001) showed that whites have more racially homogeneous networks than do other racial and ethnic groups. Smith (1999) similarly reported that only 6 % of whites and 15 % of blacks had a close friend of another race. More recently, Mouw and Entwisle (2006) found, using Add Health data, that 85.1 % of the friends of whites were white. For blacks, 71.4 % of their self-nominated friends were black. Smaller shares of same-race friends were reported by Asians (42.3 %) and Hispanics (33.5 %). The low rates among Asians and Hispanics seemingly had less to do with preferences than with limited opportunities for same-race interaction within schools and neighborhoods (i.e., baseline homophily).
Kao and Joyner (2006:972), in fact, provided evidence of “an overwhelming preference for same-ethnic peers over same-race (different-ethnic) and different-race peers.” They also demonstrated that interracial friendships increase in number when opportunities for same-race friendships at school decline (Joyner and Kao 2000); or, stated differently, more opportunities for same-race friends led to more same-race friends (see Moody 2001). This is consistent with Mouw and Entwisle’s (2006) finding that the probability of interracial dyads being friends was lowest in the most diverse schools. In one racially diverse school included in their study, the overwhelming share of the friends of blacks (75 %), Asians (86 %), and Hispanics (84 %) were of the same race. The white students, who were in the minority, were far more likely to have friends of another race. Mouw and Entwisle (2006) found that interracial friendship was heavily shaped by racial segregation or proximity; overall, about one-third of friendship segregation was attributable to neighborhood segregation. Recent evidence of resegregating schools (and their children) does not bode well for achieving greater social interaction (and perhaps mutual understanding) among young people of different races and ethnicities (Reardon et al. 2012; Zhang 2011).
With greater opportunity, minorities seemingly choose friends disproportionately from their own racial group. Racially based social segregation—measured by friendship patterns—may thus be reinforced by high (and perhaps increasing) levels of spatial segregation (as discussed earlier), which is played out locally in schools and work settings. America’s growing minority population is likely to reinforce racial and ethnic boundaries quite mechanically. Moreover, there is little evidence that the share of racially homophilious friendships declines from generation to generation among Hispanics and Asians (Quillian and Campbell 2003). This raises questions about whether intergenerational succession will lead inexorably to a breakdown in racial boundaries in the future.
In the population sciences, racial homophily is most often addressed in studies of interracial sexual intimacy, dating and cohabitation, and marriage (Blackwell and Lichter 2004; Gullickson 2006; Sassler and Joyner 2011). Interracial marriage is typically viewed as the final step in the assimilation process (Gordon 1964) and, as such, is used as an indirect indicator of improving racial relations. A new Pew Report (Taylor et al. 2010), for example, showed that 6.7 % of all newly contracted marriages in 1980 involved interracial couples, increasing to 15.1 % in 2010. However, a much lower percentage (8.4 %) of all married couples were interracial in 2010, a fact that reflects low rates of interracial marriage among older couples and the selection effects of higher divorce rates among racially mixed marriages (Zhang and Van Hook 2009). From a strictly demographic standpoint, the upward trend in interracial marriages is a result of growing racial diversity, especially among racial and ethnic groups that historically have had the highest rates of out-marriage with whites (i.e., Hispanics and Asians).16
The continuing growth of interracial marriage among America’s minority populations, however, is not self-evident for three reasons. First, as the size of America’s minority population increases in the future, rates of racial marital endogamy among the nation’s minorities may increase rather than decrease (Harris and Ono 2005; Lichter et al. 2007a). As with mixed-race friendship patterns, out-marriage is a negative function of relative group size (Qian and Lichter 2011); small minority populations are most likely (quite mechanically) to out-marry because the pool of potential partners is dominated by whites and other larger minority populations. My work with Zhenchao Qian has documented a recent slowdown or even a “retreat from intermarriage” among America’s largest immigrant groups: Asians and Hispanics (e.g., Lichter et al. 2011; Qian and Lichter 2007, 2011). The rapid growth of America’s immigrant minority population, especially from Asia and Latin America, has expanded the marriage pool among U.S.-born minority coethnics while potentially reinforcing distinctive cultural traditions and ancestral identity (Jimenéz 2010; Qian and Lichter 2011). At the same time, black exceptionalism continues to be revealed in low rates of intermarriage with whites; this is especially true for black women.
Second, patterns of intermarriage with whites and other minorities vary widely between and within major racial and ethnic groups. A recent study by Qian et al. (2012), for example, showed that patterns of intergroup and intragroup marriage vary widely among America’s Puerto Ricans, Mexicans, Chinese, and Filipinos. Because of shared religion and language (i.e., inbreeding homophily), Puerto Ricans and Mexicans are much more likely to marry other U.S. Hispanics than Chinese and Filipinos are to marry other Asians. Other studies have shown that different black populations (e.g., African-born blacks and West Indians) have exceptionally low rates of marriage with each other (Batson et al. 2006). Generational patterns in marriage and marital endogamy also are changing. For example, native-born Asians are five times more likely today to marry a foreign-born Asian than in 1980 (Qian and Lichter 2011). Yet, compared with the foreign-stock population, later generations (3+) are more likely than in the past to out-marry to whites (Lichter et al. 2011), a pattern that provides evidence of marital assimilation (Choi et al. 2011). On the other hand, second-generation Hispanics are more likely to marry first-generation Hispanics and less likely to marry third-generation Hispanics or whites than in the past (Lichter et al. 2011). This suggests a possible “return” to one’s ancestral identity among some immigrant populations. One interpretation is that continuing immigration may promote in-group solidarity at the expense of promoting integration and cohesion more broadly across society. An alternative interpretation is that the new generation, through marriage, is offering a “hand up” in the incorporation process (see Stevens et al. 2012).
Third, although most demographers studying interracial marriage have emphasized the implications of population structure (Kalmijn and van Tubergen 2010; Rosenfeld 2008), especially changing marriage-market conditions, other studies have emphasized racial preferences, mostly as revealed through online dating services. For example, recent studies of Internet daters reveal that white men were more likely to exclude black women than other minority women as possible dating partners (Feliciano et al. 2009; Yancey 2009). On some university campuses, black students are often socially isolated and excluded from interracial dating (McClintock 2010). The happenstance of local marriage-market conditions (in neighborhoods, school, or workplace) may be increasingly replaced by a new rationalization of mate-selection processes through Internet dating services that reinforce racial endogamy by optimizing the matching of preferences and behavior. The Internet may be the new intermediary in the marriage market for spouses and partners (Rosenfeld and Thomas 2012).
Like segregation, racial homophily of all kinds is deeply entrenched in American society and is slow to change, for both benign and less-than-benign reasons. Growing minority population size in the future may reinforce (or brighten) racial and ethnic boundaries and promote in-group solidarity among today’s young minority populations. This can happen quite mechanically by changing relative group size, even as racial preferences change (Qian and Lichter 2011). In-group solidarity also may be reinforced by external threats, such as anti-immigrant sentiment (Massey and Sánchez 2010), and by cultural replenishment from new immigration (Jimenéz 2010). Reductions in racial homophily in friendship networks and marriage may ultimately require additional large reductions in racial inequality on those characteristics that allow more interaction as social equals (e.g., education, income, and occupation, the so-called inbreeding component of homophily). However, as I describe below, future declines in racial inequality are threatened, perhaps as never before.
Toward a More Inclusive Society
Threats to Children’s Success
As America diversifies, the growth of new opportunities for harmonious race relations (revealed in less segregation or more intermarriage over the past half-century) and the spread of race-neutral preferences for intergroup interaction are not assured. Racial diversity unfolds most rapidly among America’s children and, more troubling, among those at the bottom of the income distribution, where competition and minority group threats are most intense. Cohort succession and the impending retirement of the baby boom generation may produce a demographic dividend for America’s minority children and young people (Alba 2009; Lee and Bean 2010). But it remains far from certain that today’s minority populations—especially blacks and dark-skinned Hispanics—will be especially well positioned in the current political and economic environment to take full advantage of new opportunities without significant policy interventions that elevate America’s values for inclusion (Alba and Waters 2012; Telles and Ortiz 2008).
Minority children face several threats that distinguish the present from the past and that highlight the policy imperative to better prepare children now for productive adult roles in the future.17
Education Achievement Gap
The education achievement gap between racial minorities and whites is increasing, even with absolute gains in educational achievement among blacks and Hispanics. Racial differences in average reading and math scores in elementary, middle, and high school are large, and they affect admissions into colleges and universities, including elite ones (for discussion, see Alon and Tienda 2007; Espenshade and Radford 2009). Calculations from the 2010 March Current Population Survey, for example, show that only 61.9 % of Hispanics aged 25–29 had completed high school, compared with 86.7 % for the population (data not shown). Among whites, this percentage was 91.6 %. College graduation was especially low among blacks (19.4 %) and Hispanics (13.2 %). Percentage-point increases since 1989 were smaller among blacks (7.7) and Hispanics (3.3) than among whites (10.1) or Asians (11.3), which means that the absolute educational gaps increased over the period.
Blacks, Hispanics, and Native Americans today attend the worst-performing schools (Logan 2011). At the same time, per capita funding of public schools (K–12) has declined, disparities in public school funding have grown, and schools have begun to resegregate. Public education at all levels is under financial assault at a time when America arguably needs to invest in the future as never before. Current racial and ethnic differences in educational attainment (and the quality of education) portend continuing inequality in the future, especially as today’s historically disadvantaged minority children assume adult roles.
Changing family structure imposes large costs on children and society (Hoffman 2006; McLanahan and Percheski 2008). Sara McLanahan’s (2004) influential PAA presidential address raised the specter of “diverging destinies” among America’s children. Diverging destinies are rooted in family disruption and nonmarital fertility, which are observed unevenly across racial and ethnic groups. Roughly 70 % of African American children today are born to single or unmarried cohabiting mothers. At the same time, America’s fastest growing population, Hispanics, may be the new fragile families (Hummer and Hamilton 2010; Lichter 2012b). One-half of all Hispanic children are born outside of marriage, which affects their developmental trajectories and their prospects for productive adult lives.
America’s racial minority and immigrant children are located disproportionately on one track, and white middle-class children (growing up in two-parent families) are on the other (McLanahan 2004; Thomas 2011). These family patterns are often reproduced from generation to generation (McLanahan and Percheski 2008; Sassler et al. 2009). Current racial and ethnic differences in family structure thus undermine racial equality both now and in the future and potentially thwart positive race relations in a diversifying society.
Since 1990, the number of unauthorized immigrants living in the United States has accelerated, reaching a peak of roughly 12 million in 2010 (Passel and Cohn 2010). However, recent reports now indicate that the rate of net unauthorized immigration is effectively zero (Passel et al. 2012). The Great Recession and the militarization of the U.S.-Mexico border have strangled the flow of new immigration from Mexico, while growing anti-immigrant sentiment has encouraged “self-deportation.” Still, roughly 5 million children today live in households with at least one unauthorized immigrant parent, and about one in five of these children are undocumented themselves (Donato and Armenta 2011).
Among (mostly) Hispanics, the primary barrier to cultural and economic incorporation and upward intra- and intergenerational mobility arguably is located in undocumented immigration, which may lead to downward mobility into perhaps a permanent minority underclass (Flippen 2012; Hall et al. 2010). In the current political environment, there is no clear route to integration among today’s unauthorized workers or their children. Menjívar and Abrego (2012) characterized the current situation as “legal violence.”
The prison population in America has exploded over the past two decades (Pettit 2012). Becky Pettit and Bruce Western (2004) showed that 20 % of black men born between 1965 and 1969 had served time in prison; this percentage was even higher among the least-educated black men (nearly 60 %). High rates of incarceration among disadvantaged minorities, especially among African American men, have undermined the social fabric of many minority communities, upset the stability of marriage and family life (including effective parenting and supervision), and impeded success in the labor market (i.e., finding good jobs that provide toeholds in the new economy) (Geller et al. 2012; Waller and Swisher 2006; Wildeman 2010).
Black incarceration has been likened to a new kind of Jim Crow: physical isolation from the mainstream and political disenfranchisement. Alexander and West (2012) argued that mass incarceration is a pernicious new system of social control that has replaced Jim Crow. The intergenerational effects on today’s minority children are potentially large but unknown as they make their way to adulthood.
Income inequality has grown, a trend that Jacob Hacker and Paul Pierson (2010) characterized as the result of a new “winner-take-all politics,” although politics is hardly the only explanation for growing inequality (Jacobs and Soss 2010; McCall and Percheski 2010; Western et al. 2012). The American class structure seemingly has become less open, and the American Dream has faded for some. Intergenerational mobility rates have declined recently (Dunifon 2010; Kearney 2006), and they have declined most rapidly among African Americans. Upward mobility—both absolute and relative—is lower among minorities (especially black women) than among whites. And, compared with mid-century statistics, opportunities for upward mobility have stagnated, a fact that suggests increasing rigidity in America’s class structure (for discussion, see Smeeding et al. 2012).
Today’s children are on the front line of the new mobility regime. Isabelle Sawhill (2012) recently concluded that “[a] growing body of evidence suggests that the United States, far from being the land of opportunity celebrated in our history and our literature, is instead a country where class matters after all, where upward mobility is constrained, especially among those born into the bottom ranks.” Such pronouncements hardly suggest an optimistic forecast for today’s poor minority children or for a more racially integrated and harmonious future.
Residential Segregation by Class
The past decade brought a return to 1980s patterns of concentrated poverty (Lichter et al. 2012; Reardon and Bischoff 2011). From 2000 to 2007, the segregation of family income groups increased in the large majority of metropolitan areas, especially among blacks and Hispanics (Reardon and Bischoff 2011). Poverty is especially concentrated among the nation’s children (Lichter and Johnson 2007). At the same time, we have seen a new geography of exclusion marked by privatized, affluent, and gated (mostly white) communities (Dwyer 2010; Lee and Marlay 2007). The poor and working classes are increasingly cordoned off from the affluent.
Place clearly matters (Dreier et al. 2001; Sampson et al. 2002). In Taxing the Poor, for example, Katherine Newman and Rourke O’Brien (2011) chronicled the evolution of southern state tax codes, which are rooted in class politics and racism and have served mostly white landowners and small-business owners with low property taxes and little or no state income or corporate taxes. Tax inequities are reinforced by legislative inertia and statutory limitations that make change nearly impossible. The only politically viable option to generate new revenues in many southern states is to raise sales taxes, even on necessities. Those at the bottom of the income distribution pay the price. Rightly or not, it is easy to extrapolate patterns in the South historically to the current anti-tax sentiment, especially among older whites (e.g., who dominate the Tea Party), and to the current resistance to invest in public education and (minority) children.
Translating Research Into Public Policy
Each of the aforementioned “threats” deserves an extended treatment (which page limitations prevent); each threat also represents a point of entry for policy intervention and evaluation. For example, the interest in new investments in early childhood education has perhaps never been greater (Heckman 2006). These programs cost money but may help prepare minority and poor children for educational success and a better future (Magnuson and Waldfogel 2005). Results of the 2012 presidential election also have again put immigration reform up for congressional debate. Citizenship provides a pathway to greater economic and political incorporation for minority immigrants. Some legislative version of the DREAM Act will surely help children who came here illegally with their parents achieve their aspirations for a better life. Evidence of growing macro-segregation may mean that programs such as Moving to Opportunity (see Ludwig et al. 2008) may require a broader spatial perspective (e.g., targeting minority families in the rural southern “black belt”) in the colonias along the borderland, and on Indian reservations. And how do we support and strengthen America’s fragile families without creating perverse incentives for single parenthood or divorce? So far, the federal government’s support of abstinence-only education and “Healthy Marriage Initiatives” have met with only limited success (Dion 2005; Kohler et al. 2008; Wood et al. 2012).
More generally, current “threats” point to the real need to invest now in children, especially minority children who are “at risk” and who have sometimes been left behind with growing diversity in bad neighborhoods and schools. It takes political will of the kind that Jane Waldfogel (2010) has described in Great Britain, where the stated goal is to eliminate child poverty altogether. Yet, in America, federal investments in the nation’s children are expected to decline by $35 billion between 2010 and 2015 (Isaacs et al. 2011).
In the end, of course, persistent racial boundaries are not rooted entirely in current class or economic disparities. They are also located in systems of racial ideology that operate independently of class structure and mobility patterns (Burk and Espinoza 2012; Burton et al. 2010). Economic mobility alone is no panacea. This point is made abundantly clear in Edward Telles’ (2004) comparative analysis of Brazil and the United States. Brazil is much less segregated than the United States and has much higher interracial marriage rates (Heaton and Mitchell 2012). Racial boundaries are considerably more fluid. Yet, racial and class inequality is greater than in the United States; in Brazil, the top of the income distribution is overwhelmingly identified as “white.” In America, improvements in minority socioeconomic status may likewise be necessary but insufficient to ensure greater interracial interaction and cultural cohesion.
No one can predict America’s racial future with certainty. Yet, there can be little disagreement that America’s ongoing racial transformation—a Third Demographic Transition—will continue well into the foreseeable future, even if restrictive immigration legislation slows the flow of legal immigrants and effectively ends the influx of new unauthorized workers. As racial and ethnic diversity unfolds from the bottom up—beginning with children—America may be challenged as never before. Racial inclusion, which has been reflected in past declines in racial residential segregation and racial homophily, is hardly guaranteed. Indeed, past progress may be lost.
Here, I have emphasized the implications of America’s movement to a majority-minority society and of changing racial boundaries, mostly from a perspective that emphasizes minority demographic and economic processes (e.g., marriage and fertility, immigration and segregation, and poverty and inequality). But America’s current demographic majority, non-Hispanics whites, also will play a large and active role in the maintenance (or not) of current racial and ethnic boundaries, in the contestation of space, and in racial politics (i.e., in either welcoming or resisting growing racial diversity and fairness). How an older, largely white population of elected officials and taxpayers responds now to America’s increasingly diverse population provides a window to the future, when today’s children successfully transition (or not) into productive adult roles.
To be sure, some optimism about the future can be found in laissez-faire solutions—those that rely heavily on “the market” or on “naturally occurring” demographic processes (e.g., cohort replacement or generational shifts). But these processes may yield fewer salutary benefits than expected. As I have argued here, past and current empirical literature on the effects of growing diversity provide important but complicated lessons for the future. Segregation may be increasing at the macro or community levels while simultaneously becoming more fine-grained at the subtract level. Growing shares of racial and ethnic minorities also create built-in pressure for more racial homophily in friendship and marriage patterns. This does not necessarily mean that America will see a rise in racial animus or hostility. In the future, there may be a new accommodation, such that different groups maintain their racial and ancestral identities (if we embrace multiculturalism) but interact with each other harmoniously—a kind of organic solidarity born of racial heterogeneity rather than a mechanical solidarity based on racial and ethnic homogeneity (for policy discussion, see Kesler and Bloemraad 2010). More troubling for the long term, however, is a 2011 Associated Press poll indicating that racial prejudice against African Americans and Hispanics has increased since 2008 (Junius 2012).
At a minimum, my reading of the empirical evidence provides cautionary lessons about America’s majority-minority future and makes the question of racial relations and boundaries fraught with uncertainty. My observations, in the end, are a call for a new scholarly commitment to research on race and ethnicity in the population sciences and demography. Children today face serious threats to inclusion as they make their way to adulthood, productive adult roles, and good citizenship. As I have argued here, America arguably will need substantial new public and private investments in today’s children (and their families) at a time when a disproportionately white elderly population holds the purse strings, when government is often viewed as a problem rather than solution, when anti-immigrant and anti-minority sentiment is high, and when a workable political consensus about how best to solve America’s problems is hard to reach. Current demographic conditions and projections for the future nonetheless suggest the need for much greater societal commitment to racial inclusion—and to children—than we have seen recently.
Finally, the field of demography is not immune to impending racial change or to these demographic pipeline issues. As demographers, what is our collective role or responsibility in creating new structural opportunities for positive interracial interaction and for breaking down existing racial and ethnic boundaries in our own profession?18 Faculty in most American universities presumably will be replaced over the next decade by a new generation of educators and administrators with a much different racial and ethnic mix and perspective. Demography as a profession will (and arguably must) diversify in terms of its practitioners, its subject matter, and its methodological approaches (Zuberi and Bonilla-Silva 2008). The topics and approaches that have occupied demographers in the past will be redefined in the future by today’s children. It is time to get to work on these issues and to pave the way for a new generation of population scientists.
This article is a revised version of the presidential address presented at the annual meetings of the Population Association of America, May 4, 2012, in San Francisco. The author acknowledges the computing assistance of Scott Sanders and Richard Turner as well as helpful comments from Fenaba Addo, Richard Alba, Linda Burton, William A.V. Clark, Gordon DeJong, Mark Fossett, Samantha Friedman, Matthew Hall, Kenneth Johnson, Kara Joyner, Maria Krysan, Marlene Lee, Dean Lillard, John Logan, Steve Morgan, Kelly Musick, Dowell Myers, Domenico “Mimmo” Parisi, Zhenchao Qian, Marty Farnsworth Richie, Scott Sanders, Sharon Sassler, Stewart Tolnay, Kim Turner, Kim Weeden, Alexander Weinreb, Michael White, and two anonymous reviewers. Of course, the author alone is responsible for the views expressed in this article.
This theme is familiar from previous presidential addresses of the Population Association of America. Samuel Preston (1984) was concerned about the growing economic divide between America’s elderly and children. Sara McLanahan (2004) emphasized the “diverging destinies” of American children and how they are shaped by economic inequality and family diversity. Douglas Massey (1996) highlighted America’s new “age of extremes,” marked by the concentration of poverty and increasing geographic separation of the affluent from the poor.
These issues are not limited to the United States. David Coleman (2006, 2010) refers to the Third Demographic Transition in much of Europe, including Great Britain, which will undergo profound changes in its racial and ethnic composition over the next half-century.
In some countries, diversity is often framed in terms of religion, but the issues are much the same. Take the example of Israel. There, ultra-orthodox Jews and Arabs exhibit extraordinarily high rates of fertility while secular Jews have below-replacement fertility. Group differences in socioeconomic status, education, and welfare dependence have fueled contentious debates about Israel’s continuing sustainability, political conflict, and sovereignty (Rebhun and Malach 2012).
Current figures mask the full extent of racial and ethnic diversity in America. Racial and ethnic classifications, based on self-identification, are both crude and fluid (Perez and Hirschman 2009). The Office of Management and Budget (OMB) defines “whites” as persons “having origins in any of the original peoples of Europe, the Middle East, or North Africa.” It also includes people who listed themselves as Caucasian or self-identified as the Irish, Lebanese, or Egyptian, among others. Some whites enumerated in the 2010 census also claimed other races as part of their racial identity. This is the so-called multiracial population. In 2010, 3.2 % of the white population (alone or in combination with another race) checked another box on the race question or wrote in another race on the census schedule (Hixson et al. 2011). Overall, 2.9 % of the U.S. population listed two or more races in 2010, up from 2.4 % in 2000 (Humes et al. 2011). Such diversity is only the tip of the iceberg; a large but unknown percentage of people who listed only one race are in fact multiracial, especially in the case of African Americans (see Perez and Hirschman 2009). Moreover, growing diversity led the U.S. Census Bureau to translate its 2010 census schedule into 59 different languages for distribution to different population groups across the United States. The Census Bureau also used bilingual (e.g., English/Spanish) forms in some parts of the country.
For example, the generational mix of Hispanics, who have disproportionately high poverty rates, will change dramatically from a population of immigrants to one of native-born Americans, who have much lower rates of poverty (Lichter et al. 2006; Van Hook et al. 2004). As the minority share of the electorate increases, voting power alone may change the political discourse and the nature of public investments.
In 2009, natality statistics from the National Center for Health States revealed 4.1 million births reported to U.S. residents (Martin et al. 2011). Of this total, 2.2 million or 46.4 % of births were to non-Hispanic white women. Of course, these numbers are based on the race and ethnicity of the mother, not the father. NCHS figures undoubtedly underestimate the percentage of minority births. About 5 % of children under age 5 are multiracial.
An early view of America’s future is revealed in the rural heartland (Johnson and Lichter 2012) and California (Myers 2007a), where new immigration and minority fertility have offset population losses associated with widespread white natural decrease.
That racial and ethnic diversity is occurring from the “bottom up” can be demonstrated with the so-called Simpson diversity index. This index gives the probability that two randomly chosen people will be of a different race or ethnicity. My calculations using data on seven ethnoracial groups from the 2010 American Community Survey indicate that diversity among children is much larger (.637) than among working-age adults (.542), especially the elderly (.350). A random draw of two children from America’s population is twice as likely to yield children of different races as of the same race. The opposite pattern exists for America’s elderly.
The ACS is a rolling monthly survey. Births recorded in a particular survey year may have occurred in the year preceding the survey (e.g., mothers interviewed in June 2007 may have borne children between July 2006 and June 2007). ACS births recorded in a particular calendar year therefore do not correspond precisely with the annual data provided through the vital registration system (collected by the National Center for Health Statistics), where births are recorded in the year that they occurred. For a comparison of ACS and NCHS fertility data, see Dye (2010).
This optimistic theme clearly is evident in the titles of several recently published books, including Creating a New Racial Order (Hochschild et al. 2012), The Diversity Paradox: Immigration and the Color Line in 21st Century America (Lee and Bean 2010), Immigrants and Boomers: Forging a New Social Contract for the Future of America (Myers 2007b), Achieving Anew: How New Immigrants Do in American Schools, Jobs, and Neighborhoods (White and Glick 2009), and Blurring the Color Line: The New Chance for a More Integrated America (Alba 2009).
The optimistic inferences we draw from observing the seemingly color-blind associations among our children sometimes ignore ongoing developmental processes. That is, racial identities (and racial attitudes) are not innate but come from social interactions that unfold over time with family members, and with others in schools, neighborhoods, peer groups, and voluntary associations (that are often homogeneous racially) (Raabe and Beelmann 2011).
For example, Taylor and Mateyka (2011), using the General Social Survey, reported that age is associated with more stereotypical beliefs about blacks, less receptivity to sharing social spaces (e.g., living in neighborhoods), negative attributions of poverty (e.g., laziness as opposed to structural conditions), and racial resentment.
Hispanics may be “whitening” along a continuum between whites and blacks at opposite poles, although their placement will be heavily influenced by skin tone (Burton et al. 2010). The racial divide may be changing from white-black to white-nonwhite to black-nonblack, or even to a new tripartite classification in which nonwhite nonblacks occupy a growing middle category (for discussion, see Bonilla-Silva 2002; Lee and Bean 2004). Others claim that Hispanics are being “racialized,” at least for some national-origin groups (e.g., Mexico’s mestizos) and for those with “dark” skin (Telles and Ortiz 2008; Telles and Sue 2009).
Separating opportunity structure from preferences is a difficult task; people—at least some—choose their own opportunities. That is, they choose to live (or not) in racially integrated neighborhoods. This is the endogenous membership problem that makes separating opportunities from preferences an extraordinarily difficult task.
Preferences are not easily separated from structural conditions. Blacks may prefer to live in predominately black neighborhoods precisely because they believe that they are unwelcome in white neighborhoods. The endogenous membership problem—people choose their neighborhoods—makes the study of segregation difficult (Mouw and Entwisle 2006). Not surprisingly, a frequent goal of recent research is to separate opportunity from preference (Wimmer and Lewis 2010).
The Pew study, for example, reported that 63 % of all Americans said they “would be fine” if a family member married outside their race. This high percentage is not being driven by minorities alone. Among whites, the percentage was nearly as high (at 61 %) as the national percentage. Of course, these figures are far in excess of observed rates of intermarriage, which raises questions about the mismatch between attitudes and behavior (Herman and Campbell 2012). Although intermarriage may not be experienced directly, it is often experienced indirectly. Goldstein (1999) showed that roughly 20 % of all Americans have a close relative who is of another race.
Lest we despair, Gunnar Myrdal (1944:51), in An American Dilemma, reminds us that the “children and grandchildren of [previous] unassimilated foreigners are well-adjusted Americans,” despite receiving “their first course in Americanization in the squalid and congested quarters of New York’s East Side and similar surroundings.”
Data from the National Center for Educational Statistics show that 75 % of the full-time college and university faculty (Fall 2009) are white; only 5.4 % and 3.8 %, respectively, are blacks and Hispanics (National Center for Education Statistics 2013:Table 264). Yet, for 2009, data from the Current Population Survey show that 18.4 % of the U.S. population aged 15–24 was black, and 15.5 % was Hispanic. In 2009, only 13.2 % of all doctorates awarded in 2009 went to underrepresented groups (American Indian, black, or Hispanic) (see The Chronicle of Higher Education 2011:B44).