Richard Alba writes in this 2020 book:
The third chapter addresses the question: how do Americans arrive at ideas about ethno – racial change in their society? The notion that whites will become a numerical minority has been around at least since President Bill Clinton, in a 1997 speech, claimed that this would happen in a half – century. But the pronouncements more recently of what I call our “demographic data system” have been critical to Americans’ acceptance of this idea. It is certainly true that, in an era of large – scale immigration, various observers could have arrived at this notion and publicized it. Without the data and interpretations coming from the Census Bureau and other parts of this system, however, the idea would have lacked the imprimatur that gives it legitimacy.
The chapter reviews Census Bureau data and pronouncements about population change and the ways in which they have been taken up by the mass media. A Census Bureau press release introducing the notion of a “majority – minority nation” about a decade ago was especially consequential. The chapter then explores the reactions to the census data from political and cultural commentators, from Pat Buchanan to Ezra Klein. The reactions on the right and the left are, not surprisingly, different: the right issues dire warnings about national decline, while the left exudes a confident sense of inevitability, combined with some degree of celebration of “the end of white America.” The chapter also considers white Americans’ everyday experiences with diversity, especially in their neighborhoods. I summarize the evidence about the sharply rising diversity in white neighborhoods over the last several decades and what we know about whites’ responses to it.
The fourth chapter examines how our demographic data system has produced the majority – minority prediction for the next several decades and also why, despite the critical innovation of multiple – race reporting in the 2000 Census, it has failed to call an equivalent attention to the surge of ethno – racial mixing in families. The chapter introduces the reader to the Census Bureau’s measurement of race and ethnicity and includes a brief tour of its history. The current questions and the construction of data from them are discussed, as are the bureaucratic, political, and legal constraints on census data, exemplified by the role of the Office of Management and Budget in developing standards for ethnic and racial data reporting. The chapter then brings into play the increasing extent of ethno – racial mixing in families, beginning with the steady rise in marriages across the major ethno – racial divisions. This mixing leads naturally to increases in the number of children with mixed backgrounds (whether formed through marriage or not); the great majority of them have one white and one minority parent. I present data from census data sets and birth certificates to demonstrate the rapid growth of mixed parentage among infants and the relative frequency of different ethno – racial combinations among them. The chapter concludes by examining how census data procedures have dealt with this momentous new development. For reasons I develop, those procedures have proven inadequate to give Americans an accurate understanding of ethno – racial mixing in families and its implications for the future. I show, for example, that the group with mixed minority – white parentage is the pivot on which the outcomes of Census Bureau population projections depend; if we change our assumptions about its classification, the projected future looks quite different…Demography is destiny.
— SCAMMON AND WATTENBERG (1970)
A half – century from now, when your own grandchildren are in college, there will be no majority race in America.
— PRESIDENT BILL CLINTON, SPEAKING AT THE 1997 UNIVERSITY OF CALIFORNIA – SAN DIEGO COMMENCEMENTAny American paying attention to the news in the new century could be forgiven for thinking that whites, the dominant racial group in the United States ever since the nation’s founding, were teetering on a demographic precipice and about to topple into a numerical minority, with potentially huge implications for their position of dominance. Power in a society is, of course, not strictly equatable with numbers but rests also on relative economic status and positioning within the political system. In these respects, the dominance of whites is much more secure than is implied by demographic strength alone. Nevertheless, in a democratic society, population sizes are undeniably related to electoral weight and ultimately to power.
The demographic news for several decades has been a steady drumbeat of exploding ethno – racial diversity, of expanding minority populations, combined with the aging and numerical decline of whites in the aggregate. 1 Immigration has been a powerful driver of change. Large – scale immigration to the United States had been shut down by restrictive legislation in the 1920s and then by worldwide economic depression and war in the following two decades. Immigration began to resume after the end of World War II and received a mighty boost from passage in 1965 of the Immigration and Nationality Act (or Hart – Celler Act) under the influence of the civil rights movement. That legislation finally dismantled the regime of regulating immigration by national origins and implicitly by ethnicity and race. 2 The national – origins quotas that restricted immigration from southern and eastern European countries were gone, and so too were the draconian racist limitations on immigration from Asia. The United States had opened up to immigration from anywhere on the globe, subject to a uniform country quota set initially at twenty thousand per year…Given the apparent decline of the white share of the American population, it comes as no surprise that the nation’s demographic data system envisions whites as an eventual numerical minority. I refer to a “demographic data system” rather than simply to the Census Bureau because the bureau is not the sole actor, even if it is the nation’s official custodian of demographic data and the public face of the system. Functioning under bureaucratic, legal, and political constraints, the Census Bureau does not have full freedom of action when it collects and reports data. By law, it is required to follow the guidelines of the federal Office of Management and Budget (OMB) for ethnicity and race data, and those guidelines, as the next chapter relates in detail, have played a critical role in shaping the data as publicly presented. Also part of the system are the demographers who help Americans, through reports, articles, and books intended for a broad audience, to understand the patterns in demographic data and their implications.
This system and the Census Bureau as its public face play a critical role in the wide acceptance of the majority – minority perspective on the nation’s future. Of course, many whites, especially those in communities experiencing economic and population stagnation if not decline and those adhering to white nationalist ideologies, are capable of independently coming to the view that their group is in decline. Without the imprimatur of the demographic data system, however, the anxious imagining that the white group may soon be eclipsed numerically by minorities would amount to little more than opinion. The demographic data and interpretations issued by the system rescue these ideas from the realm of opinion and transmute them into common knowledge.
The Census Bureau regularly prepares population projections that divide the population into its major ethno – racial components: American Indians, Asians, blacks, Hispanics, and whites. This division is required for credible projections because the demographic drivers — fertility, mortality, and migration — vary substantially among these groups. Starting with the projections prepared in 2000, a momentous transition began to appear of whites from a numerical majority to a minority. The 2000 projection foresaw this transition as occurring in 2059. 8
Mass media were slow to pick up on what was initially an unheralded feature of the projections. Perhaps the nomination in the summer of 2008 of the first nonwhite candidate to head a major – party ticket for president sharpened sensitivities to the rapidity of demographic change. At any rate, in the New York Times , the first announcement of the possible future minority status of whites appeared that summer in an article by the demography reporter Sam Roberts, under the headline “In a Generation, Minorities May be the US Majority.” 9 Citing the Census Bureau’s projections issued that year, the article stated that “the census calculates that by 2042, Americans who identify themselves as Hispanic, black, Asian, American Indian, Native Hawaiian and Pacific Islander will together outnumber non – Hispanic whites.” The claim, incidentally, is not accurate, because of ambiguities that plague the projections, but we will get to that later. Roberts cited increasing immigration levels and higher birthrates for the shortening of the time to the transition (from 2059) and quoted an expert at the Population Reference Bureau: “No other country has experienced such rapid racial and ethnic change.” (“In peacetime” probably should have been a qualifier.)
The word was out. Subsequent population projections would receive much more public attention. The Census Bureau redid the projections in 2009, and then the next big release, based on 2010 census data, took place in 2012. The press release highlighted the growth of ethno – racial diversity over the half – century time span of the projections and invoked the concept of a “majority – minority nation,” which was projected to occur, it asserted, by 2043. 10 In its report, the New York Times did not employ this phrase, but did say in its headline that “census officials” foresaw “a plurality nation.” 11 According to the Census Bureau’s acting director, Thomas Mesenbourg, “The next half century marks key points in continuing trends — the US will become a plurality nation, where the non – Hispanic white population remains the single largest group, but no group is in the majority.”
Based on reporting by the Associated Press, numerous newspapers and other media throughout the country picked up on this important development in the demographic future. The Pittsburgh Tribune Review reported the news with the headline “Whites No Longer a Majority in the US by 2043, Census Bureau Projects.” The Pantagraph of Bloomington, Illinois, was even more terse: “Census: Whites in Minority by 2043.” The Bismarck Tribune ran the Associated Press article, with the same headline. The Charleston Daily Mail in West Virginia varied the headline slightly. 12
For the next set of population projections, the Census Bureau published a report in 2014 as part of the regular Current Population Reports series to accompany its bare – bones press release. The report, written by two Census Bureau demographers, made clear in its first paragraph that by midcentury whites would be a numerical minority of the population: “More than half of all Americans are projected to belong to a minority group (any group other than non – Hispanic white alone).” The report went on to declare that, “by 2060, the share of this group [non – Hispanic whites] is projected to be just 44 percent.… According to these projections, the majority – minority crossover will occur in 2044” (emphasis mine). 13
For the most recent projections, issued publicly in 2018, the Census Bureau press release tried to avoid the majority – minority issue. The headline emphasized the aging of the nation. In discussing race and ethnicity, the statement muddied the waters by calling attention to the growth over time of the “White – alone population, regardless of Hispanic origin,” though it also acknowledged the shrinkage of the “non – Hispanic White – alone population.” 14 In any event, the genie was out of the bottle. Others lost little time in identifying the crossover year to a majority – minority nation (when whites would become a minority). William Frey, the Brookings Institution’s Institution’s well – known demographer, posted a report a day after the Census Bureau’s release with the headline “The US Will Become ‘Minority White’ in 2045, Census Projects.” 15 In truth, this deduction wasn’t too difficult, since detailed tables produced by the bureau to present the projections showed that the percentage of non – Hispanic whites would dip below 50 percent in that year.
Projections are of course about the future, but during the Obama administration the Census Bureau began to issue estimates about the existing population that amounted to signposts on the way to the majority – minority country. These focused on the child population, since the demographic dynamic engendering a majority of minorities was emerging from a sharp age disjunction between the heavily white older population, dwindling because of mortality, and the increasingly diverse population of children.
The first releases concerned infants (children who have not reached their first birthday). Starting with the births in 2011, the Census Bureau declared that the majority of infants were members of minorities. Subsequent corrections to the data showed that whites retained a narrow majority among the infants born in that year. However, a majority – minority situation among infants was established for the babies born in 2013, and it has been repeated every year since then. In 2016, according to census estimates, minority births exceeded white births by about sixteen thousand — granted, a very small margin when the total number of infants is almost four million. 16
If the margin was small, the news was big. Here was indisputable evidence of the ineluctable emergence of the future majority – minority society. National Public Radio’s article “Babies of Color Now the Majority, Census Says,” invoked the Census Bureau projections: “Nonwhites are expected to become the majority of the nation’s children by 2020.… This is now the reality among the very youngest Americans: babies.” Responding to the earlier, erroneous Census Bureau report, a Washington Post article had made the same claim in 2012: “The census has forecast that non – Hispanic whites will be outnumbered in the United States by 2042, and social scientists consider that current status among infants a harbinger of the change.” USA Today , with a broader reach, asserted that “minorities are now a majority of births; Census shows how fast the nation is changing.” 17
In short order, the Census Bureau reported that minorities dominated among America’s young children. This announcement came in 2015 for those under the age of five. US News & World Report promptly blared, “It’s Official: The US Is Becoming a Minority – Majority Nation,” opening its article with, “They may not know it, but for kids under the age of 5, the day the United States became a minority – majority nation has already arrived.” 18 Just a few years thereafter, the demographer William Frey pointed out that the latest census population estimates now showed minorities to be the majority of children under the age of ten. Labeling the children born since 2007 as “generation Z – Plus,” he stated that “we are on the cusp of seeing the first minority white generation.” 19
In brief, since 2000 the Census Bureau has publicized the majority – minority outcome in its projections, as well as in population estimates for children that are consistent with the majority – minority forecast. The bureau’s efforts have been greatly amplified by the media attention its data releases receive. This has established in the public mind the notion of an inevitable transition in the not too distant future to a nation where whites will be a demographic minority, with essentially unknowable but presumably huge consequences for the relative positions of whites and nonwhites.
Extrapolating from the Data
Of course, many commentators, on both the political right and left, have taken these ideas much further. On the far right, the common assertion is that immigration is a mechanism for forcing unwanted demographic change on American whites. The Daily Stormer , for example, takes as an “obvious and admitted fact that all White nations are being purposefully multiculturalized to the end of making them racial melting pots.” 20 Fox commentator Laura Ingraham opines that “massive demographic changes have been foisted upon the American people,” with the consequence that “in some parts of the country, it does seem like the America that we know and love doesn’t exist anymore.” 21
An atmosphere of crisis bordering on hysteria pervades the discussion of immigration and demographic change among some conservatives. The right – wing commentator Pat Buchanan has evoked the bogeyman of “Third – World America.” In 2018, he identified as “the great issue of our time” the demographic changes wrought by immigration, and he wondered whether America could stop the “invasion” of other countries before “they change the character — political, social, racial, ethnic — character of the country entirely.” However, Buchanan is more worried about “Western suicide” than about what others seem to see as cultural and demographic homicide. The New York Times columnist Charles Blow has not unfairly characterized these views as “white extinction anxiety.” 22
This anxiety has bubbled up in a nightmare vision that goes far beyond the demographic projection of a white minority. Known as “replacement theory,” or “the Great Replacement,” this fevered vision is not limited to the United States but is also believed by anti – immigration extremists in western Europe. 23 The core claim of replacement theory, popularized in recent years by the French thinker Renaud Camus, is that a demographic process is under way to replace native white populations of wealthy Western countries with immigrants and their descendants who come from the Global South. An especially sinister twist is the additional idea that this process is being directed from above by Western elites pursuing their own interests, or even by Jews. These ideas echoed, for instance, in the chant by white nationalists demonstrating in Charlottesville, Virginia, shortly after Trump occupied the Oval Office — “Jews will not replace us.” 24 More ominously, they have motivated multiple acts of mass violence; the shooter who in 2019, targeting Mexicans, killed twenty – two people at an El Paso Walmart posted a manifesto just before the violence referring to population replacement.
To the left of center, by contrast, there is generally a preternaturally calm confidence in the inevitability of a majority – minority society in the near future. Blow’s column on white extinction extinction anxiety continues by asserting that “white people have been the majority of people considered United States citizens since this country was founded, but that period is rapidly drawing to a close.” Citing reports about the harbingers of profound demographic shift from the Brookings Institution, among others, Blow concludes, “This is happening. America will soon be a majority – minority country.” The panic about this transition, he notes, is limited to those “who conflated America with whiteness.” 25
The journalist Ezra Klein, then the editor of the webzine Vox , presents the projections from the demographic evidence in a similarly matter – of – fact way, while underscoring its political significance. Calling demographic shift and whites’ reaction to it “the most important idea for understanding American politics in 2018,” Klein ticks off the indicators of imminent shift as identified by the Census Bureau and demographers: the majority of infants now are nonwhite; the white population is aging, and its most common age is nearly sixty; and accordingly, in most states, deaths among whites outnumber births. Only whites are projected to decline in numbers in the future; other groups will grow. The unavoidable conclusion is that in a few decades, “for the first time in the nation’s history, non – Hispanic whites will no longer make up a majority of the population.” 26
Oddly, given the facticity of this presentation, Klein notes, if briefly, that “race is what we make of it, and what we make of it shifts and mutates.” But he does not pursue this thought, which could undercut the apparent implications of demographic shift…Americans, both white and minority, tend to exaggerate minority population sizes and to underestimate the size of the white majority, to the point that many appear to believe that whites are already a numerical minority in the nation. The General Social Survey, a highly regarded national poll, revealed in 2000 that roughly half of Americans perceived whites as a numerical minority, or at least as a smaller group than blacks and Hispanics combined; perceptions of either sort were more common among minority – group members than among whites. Respondents had more realistic estimations of minorities in their local communities, but those living in communities with a larger minority presence perceived minorities as larger in relation to whites in the nation as a whole. 39
The perception of the relative sizes of minority and white populations is linked in turn to whites’ attitudes toward immigrants and racial minorities. This is consistent with the social – psychological research described in the previous chapter, which used population projections of whites’ future as a demographic minority to show the effects of status threat on whites’ political stances and attitudes toward minorities. The role of status threat equates with the notion that whites’ prejudice toward minorities is often defensive — part of a posture by many whites to maintain their sense of superior group position in relation to nonwhites. 40
An imaginative study by Ryan Enos, a Harvard political scientist, illustrates vividly the impact of diversity in the everyday on whites’ political attitudes. 41 He sent pairs of Spanish – speaking, Latine – appearing confederates over the course of several days to suburban train stations around Boston. Surveying the white commuters on the train platforms before and after exposure to these “strangers,” Enos found a shift in their attitudes, toward more exclusionary positions concerning Mexican immigration and the children of undocumented immigrants. The experiment powerfully demonstrates the impact of even subtle ethno – racial changes in the everyday environment on white attitudes toward minorities…The question is unavoidable: why should anyone not believe the data and projections of the Census Bureau? These point to a rapidly unfolding shift in the ethno – racial complexion — a “browning,” or “beiging,” it is sometimes said — of the United States. 1 This change appears in population projections, but these are, after all, just calculations about a hypothetical future, though based on current data and current patterns in basic demographic processes like fertility. 2 The calculations could easily prove wrong because these patterns are not fixed but shift over time. For example, white fertility may reverse course and go up; or immigration to the United States could decline, as some Americans appear to want to happen.
But the changing complexion of the population is also apparent in data about the present, especially in differences between older and younger Americans. Older Americans are disproportionately white, and younger Americans nonwhite. In particular, children under age ten — “generation Z – plus,” as they have been dubbed by the demographer William Frey — have been characterized by him as minority white, the first cohort of Americans for whom this is true. 3 Since this will likely be true of future birth groups, and since the white population apparently has already begun to decline, isn’t the future majority – minority nation an inevitability?Collecting data is one thing; processing the data into presentable form can be something else. The two questions, with the detail they allow — especially the option of identifying with more than one race — can give rise to numerous complex combinations, such as persons of Mexican ethnicity and Chinese race. On its website, the Census Bureau provides counts for some of these detailed and complex categories and makes data available for large samples of individuals (the so – called PUMS data, or Public Use Microdata Samples) from which other counts not provided by the bureau can be calculated.
However, the public pronouncements about census data take advantage of a far more condensed set of categories and generally focus on mutually exclusive ones. Six of the categories stand out: single – race whites; single – race blacks, Asians, and American Indians; persons of mixed race; and Hispanics. (There is also a small category for Hawaiians and other Pacific Islanders, and another small “other race” category for those who cannot be placed anywhere else. 14 To avoid overburdening the discussion, these two categories are not considered.) As is frequently noted in Census Bureau publications, Hispanics “may be of any race.” Hence, individuals who indicate any Hispanic ancestry are classified in this group, whatever their race. The Census Bureau’s public statements that minorities outnumber whites among US babies and that projections show whites becoming a minority of the population in the mid – 2040s are undergirded by these categories. In other words, the “whites” who are or will be a minority are non – Hispanics who check only “white” on the race question. Others are also white in some sense — for instance, they may be white and Asian by background, or have one white and one Hispanic parent — but are counted as not white for the purpose of describing the population in public statements.
Table 4.1 , which is taken from the Census Bureau’s report on the 2017 projections, illustrates this data scheme and its usual interpretation. The subheading to the table is actually the headline: “The non – Hispanic White population is projected to shrink.” But in fact the headline concerns only those whites who select white as their sole race; those who are white and another race are not deemed relevant to the claim. There are two lines for the single – race white population, but the subheading directs attention to the second of them, to non – Hispanic single – race whites. In other words, the first line, which shows the white population increasing over time, includes Hispanics, as do all of the other lines for racial groups. Accordingly, the note at the bottom of the table found in the report warns that “percentages will not add to 100 because Hispanics may be of any race.”
Why not interpret the line for whites including Hispanics? It tells a very different story, indicating that whites will still be a strong majority of the population, about two – thirds, in 2060. This story gets no traction for two reasons. For one, there is a strong proclivity in census data presentations, as well as in the public understanding of them, to see the ethnic and racial data as reports of identities and group affiliations and to presume that one identity must be dominant for individuals whose ancestral origins supply multiple possibilities. To put the matter differently, group belongings tend to be seen as mutually exclusive: affiliation with one precludes affiliation with another. There is also a conceptual tidiness involved: mutually exclusive, exhaustive categories classify every individual in one and only one of them.
The second reason has to do with empirical knowledge about the difficulties that many Hispanics have in answering the question about race. Since their primary identity is often within the umbrella of Latine identities, they can be unsure how to answer the race question. For example, immigrants can be confused by US racial categories, which do not match those at home, and often write a nationality label in the “other” race space. Hispanics may also choose “white” for a variety of reasons, including as an attempt to signal their Americanness. 15 (For example, many later – generation Mexican Americans in south Texas identify as “white” because of family memories of Mexicans’ past legal status as white, which originated in the citizenship guarantee of the 1848 Treaty of Guadalupe Hidalgo at the end of the Mexican – American War.) In other words, in the current two – question format for recording ethno – racial background, it is a mistake to assume that Hispanics’ choice of “white” for their race generally conveys the same information about identities and group affiliations as it does for Americans whose ancestors immigrated from Europe.
This ambiguity in race reports by Hispanics has a deeper implication. The current questions make it impossible for individuals filling out census forms to indicate meaningfully — in a way that the Census Bureau can recognize — that they or their children have mixed Hispanic and non – Hispanic ancestry. There is a fundamental lack of parallelism in the reporting of race and Hispanic origin. Matters would be different if Latines were not forced to choose a race but could do so only when they thought it relevant, when, for example, they come from mixed Anglo – Hispanic families.
The OMB’s role is especially critical to this book’s story because it defines standards for the collection of ethnicity and race data. This may seem peculiar because, as its name implies, the OMB is not an agency that is primarily concerned with the generation of scientific knowledge. One mission of the OMB, however, is to ensure that important kinds of data are consistent across federal agencies and departments, and ethnic and racial data are crucial for civil rights enforcement. 16
The standards that regulate census ethnic and racial data today were promulgated by the OMB in 1997, in Directive 15. The standards, to be fair to the OMB, were not developed in a remote and bureaucratic way, but on the basis of extensive discussion with stakeholders, including the Census Bureau. And they took into account concerns that were already affecting the bureau’s thinking about the impending 2000 Census, such as the introduction of a multiple – choice option for the race question. The directive stated that
respondents shall be offered the option of selecting one or more racial designations. Recommended forms for the instruction accompanying the multiple response question are “Mark one or more” and “Select one or more.”
That memorandum identified the major categories to be used for the reporting of such data. 17 These are the categories in table 4.1 . The memorandum also defines these labels in terms of the ancestries or origins underlying them: for example, whites are equated with “origins in any of the original peoples of Europe, the Middle East, or North Africa.” The directive does not require that the original data be collected solely in terms of these categories; more detailed ones may be used, but these must be capable of being aggregated into the required categories for reporting.
The OMB standards mandate a two – question format, one question for race and the other for Hispanic ethnicity. Both questions are to be answered by self – reports. However, there is an unresolved ambiguity about this reporting. Often, the census form for a household is completed by a single household member, and the degree to which that person’s reports on other household members would coincide with what they would claim themselves is unknown. Since they all live together, and most often constitute a family, the correspondence is presumably very high, but it is probably not perfect.
The multiple – race option raised concerns among civil rights groups and in Department of Justice discussions with stakeholders prior to the 2000 Census. The census counts of minority groups are essential for civil rights jurisprudence, such as litigation over employer discrimination and enforcement of the Voting Rights Act. With a mixed – race option, civil rights advocates feared that minority counts could be reduced, with the consequence of weakening arguments, say, that employers were discriminating because they had failed to hire an appropriate number of minority workers given the minority presence in geographically proximate labor markets. Any removal of mixed – race individuals from minority counts seemed especially unjust in light of the past relegation of mixed – race blacks to the black group via the so – called one – drop rule.
The OMB issued “guidance” on this issue in March 2000, just prior to the census. 18 The guidance was tailored specifically to civil rights monitoring and enforcement and laid out principles for classifying individuals of mixed race for those purposes. In particular, the OMB said that individuals of mixed minority and white origins should be allocated to the minority category. Just as with the one – drop rule, in other words, the minority side was to take precedence over the white side in the federal statistical system, at least for civil rights purposes. (The irony involved was not lost on some critics.) Though this intent might seem to limit the principle, it is easy to see why its impact would be larger within the broad scheme of census data reporting. How could the census publish two contradictory versions of the data for the same place? This guidance tipped the scales against a more expansive definition of whites in particular. If two versions of ethno – racial counts existed for the same geography, one with higher counts of whites than the other, and hence lower counts of minorities, the consequences would be great confusion (which one is right?) and the undermining of civil rights litigation (why should the court accept these data, the defendant could argue, when the Census Bureau has published another version more favorable to me?).
With this OMB recommendation, a track switch had been thrown, perhaps unintentionally, but thrown nevertheless. Data, at least in the form shown to the broad public and reported by the media, were to be shunted in a distinct direction. The decision to give preference to the minority side in reporting data has had huge consequences, as we will shortly see.
The role of the OMB in preparation for the 2000 Census may appear to suggest a largely collaborative relationship with the census, rather than a constraining one. A different picture appears in the run – up to the 2020 Census. Aware of shortcomings in the two – question format — especially because many Hispanics, forced to answer a race question that they do not find meaningful, place themselves in the “other” race category — the Census Bureau tested in 2015 a new composite question. In the one – question format, a “Hispanic, Latino, or Spanish” category would be added to the list of races, and respondents would still have the option of marking one or more categories. An additional element in the test was a new race category for “Middle Eastern or North African” (or MENA). This part of the test responded to demands from individuals and interest groups that the groups from this part of the world, whose members frequently frequently experience prejudice and discrimination in American society, not least because many are Muslim, must be distinguished from whites, so that their disparate status could be analyzed.
The test was a success in the eyes of the Census Bureau. 19 Latine use of the bothersome “other” race category (which, because it is not an OMB category of reporting, has forced the Census Bureau to reallocate its occupants to the OMB race categories for other governmental agencies) was greatly reduced. Moreover, most Hispanics no longer checked any racial category, corroborating the widespread belief that “Hispanic, Latino, or Spanish” is the main identity for much of this population. A side benefit was that Latines who checked racial boxes — mostly “white” — did so meaningfully, according to interviews conducted by the Census Bureau. These Latines were generally reporting mixed family origins; those who reported “white” in the new one – question format were usually indicating a family background that included European – descent forebears. In the two – question format, mixed backgrounds could not be reported because of the assumption that a Hispanic “may be of any race”; as noted, a Hispanic identity therefore took precedence over everything else.
The new MENA race category also worked as anticipated. Consequently, the Census Bureau staff recommended using the new composite question in the 2020 Census. 20 However, the OMB appears not to have cooperated. Use of the new question would require a change in the OMB standards for race and ethnicity, but no new standards were issued. In January 2018, the Census Bureau announced that it would retain the two – question format for the 2020 Census and no new categories would be added. The National Public Radio journalist Hansi Lo Wang reported that, according to census experts, the decision “suggests the Trump administration will not support Obama – era proposals to change how the US government collects information about race and ethnicity.” 21
If so, politics has constrained the census on a very consequential issue…The mixing of white with minority parents has a profound and little – known impact on the public reporting of demographic data. Understanding it requires a bit of wading into the weeds of census ethno – racial classifications. Consider, for example, the widely publicized statement by the Census Bureau, discussed in the preceding chapter, that the majority of babies born in the United States are now members of minority groups. This statement relies on a classification in which most mixed babies with a white parent are categorized as minority — not white, in other words. In identifying children as mixed, the Census Bureau does not look to the ethno – racial backgrounds of parents, since one or even both may be missing from the household (the latter in the case of foster children or children being raised by grandparents). Instead, it looks to how the ethno – racial background of the child is reported by adults on the census form…
The surprising fact is that not all mixed children, even when they are infants, are reported as such to the census. 34 (Birth certificates are not helpful here because they do not indicate how parents view their newborns.) Table 4.2 shows the distribution of infant identities as reported by one white and one minority parent in 2017. Overall, about one – third of the infants were reported as mixed, though this figure is lowered by the distribution of identities when the minority parent was Hispanic. By conventional census coding, the identities of Anglo – Hispanic children are mostly Hispanic — that is, they are assigned the identity of the minority parent — and they are therefore reported as minority in table 4.2 . They are coded this way essentially by default because the two – question format currently used by the Census Bureau requires that any indication of Hispanic ancestry lead to classification as Hispanic regardless of what is reported on the race question and regardless of the intentions of the parents. If we remove the white – Hispanic parent pairs, where census conventions create confusion about parental intentions, then the percentage of infants identified as mixed jumps to 70 percent.
Since the children reported as mixed are classified as nonwhite by the census, the table makes apparent how few mixed children appear on the white side of the ledger: just 17 percent. If we apply this figure to what we know from the birth certificate data to have been the share of mixed white – minority births in 2017 — 10.6 percent — the implication is that almost 9 percent of all 2017 births involved white parents but would be nevertheless classified as “minority” by the Census Bureau. Based on this classification decision, the Census Bureau would conclude that whites were a minority among 2017 infants. But the number of “minority” births to white parents is clearly large enough to alter the conclusion with a different classification rule, one that recognizes that partly white infants are not necessarily “babies of color.”
A priori, these babies are in between: they are both white and minority, with family roots on both sides of a still salient societal divide. In this respect, the idea that US babies are now mostly minority is an illusion fostered by arbitrary classification decisions — arbitrary at least with respect to the daily lives of these young children. The reader will remember that this classification decision has its origins in an OMB – issued “guidance” to be used for civil rights purposes.
The decision has had far – reaching consequences for census data and lies behind the population projections that yield a majority – minority society by the mid – 2040s. Understanding this statement again requires wading into the weeds. In a projection, an individual’s ethno – racial classification is fixed: it does not vary over time. The classification as white or not white occurs at first encounter, which can be: (1) when the individual appears in the base, or initial, population data; or (2) when he or she is projected to be “born” (in the future obviously). In either case, the projection relies on an initial ethno – racial description. In the case of an existing individual, this comes from actual data (responses to the two – question format); in the case of a projected birth, the Census Bureau projection program invokes the current patterns by which parents describe their children in ethno – racial terms to make a classification. 35 Either way, the great majority of individuals from mixed majority – minority family backgrounds will be classified as not white. And incidentally, according to the 2017 projections, most of these individuals will be born in the future, some of them of course to present – day Americans with mixed white and minority ancestry. Because most of these prospective parents are already classified on the not – white side of a binary division, their children are likely to be classified this way as well. The classification decision will propagate.
Consequently, individuals of mixed majority – minority parentage are the pivot in the projections. That is, whether or not a majority – minority society is forecast by 2060 depends on the classification of the mixed group. This unsettled outcome emerges when the projections are recast with alternative category placements of the mixed majority – minority group…Consider the infant population, a strong indicator of the future. Figure 4.7 shows its projection under alternative assumptions. The lowest line represents, again, the census – defined white group, which appears to plummet as a percentage of all infants, starting out below 50 percent in 2016 and dropping to 34 percent by 2060. However, the upper line, which again includes the projected number of non – Hispanics who report both white and nonwhite races and the estimated number of Hispanics with Anglo ancestry, presents a very different picture. Although it descends, it does so slowly, dropping by just five percentage points over the life of the projection. And in 2060, white and partly white infants still make up a majority.
What the example of infants demonstrates is the powerful growth of the mixed – white population in the projections. By the 2050s, one of every three babies with white ancestry will also have Hispanic or racially nonwhite ancestry when second – generation mixes are counted. These mixed infants will be almost one – fifth of all infants, of any ethno – racial background. Consequently, assumptions about the ethno – racial assignment of mixed minority – white individuals have a large impact on the projections. The assumption in public demographic data that they are not to be counted with whites determines the outcome of the majority – minority society by the mid – 2040s…I am not advocating that the mixed group be counted exclusively with whites. Their position in a majority – minority binary scheme is profoundly ambiguous: they start life with close family links to whites and to minority kin. In this respect, they are an in – between group that through kin relationships spans a societal chasm.
In any event, it should be apparent that our understanding of what the United States might look like in coming decades hinges on what we are willing to assume about the mixed minority – white group. This raises the obvious question of what we know about them. The United States is still a very racially segregated society, and even if Americans from mixed majority – minority backgrounds span its divides through their family connections, as adults they may still lean in one direction or another in choosing where to live and whom to have as friends and, even more pertinently, as a life partner, as well as how they identify themselves. The next chapter delves into these critical questions.
What we have seen so far calls into question the appropriateness of a binary ethno – racial scheme applied to the US population. The public presentation of census data and projections has emphasized the division between the white majority and minority groups, implying that these two blocs are exhaustive and mutually exclusive — everyone fits in one and only one category. Admittedly, a scheme like this can suffer from some degree of error and still provide a very useful approximation. However, the findings in this chapter show that the degree of error is no longer small, and it is growing. At present, more than 10 percent of infants born in the United States have one white and one minority parent. That fraction has been increasing steadily and is virtually certain to increase further in the near future. As the current youngsters from mixed families age and new births continue to feed into the mixed population, the share of all Americans from such backgrounds will increase. With their numbers, they are guaranteed to have a major impact on our society, and the ambiguity of their social location between the majority and minority social formations will firmly refute any binary scheme for predicting the American future.
By relying so heavily on publicly presented census data, Americans have accepted a distorted picture of changing on – the – ground social realities. The census data in their public form emphasize a binary division, but a finer sifting of them uncovers much more mixing across this divide than is generally recognized. The headlines blare about minorities predominating among births, then among children under a certain age, and soon among all children. The social realities could be quite different: in stark contrast to the headlines, a substantial fraction of these “minority” children will have a white parent.
