Poverty and Affluence across the First Two Generations of Voluntary Migration from Africa to the United States,1990

Abstract

The first substantial waves of voluntary migration from Africa arrived in the United States in the last quarter of the twentieth century. The largest number of them hailed from Egypt, Ethiopia, Nigeria, and South Africa. Highly select in their educational aspirations and achievements, many of them settled and started families. By 2010, their U.S.-born children had begun to reach adulthood, offering us a first look at intergenerational mobility among voluntary migrants from Africa. The racial diversity in this group of immigrants allows us to gauge the impact of racial stratification on immigrant adaptation. 1990 U.S. census and 2008–2012 American Community Survey data are used to uncover patterns of affluence and poverty among young Egyptian, Ethiopian, Nigerian, and South African immigrants in 1990 and U.S.-born men and women of those ancestries in 2008–2012. White and Black cohorts of U.S. birth and stock serve as additional referents. I find that women of the African second generation have advanced faster than their male counterparts and that racial group membership is at least predictive of financial well-being as specific national origins, with Black Africans, and Ethiopians in particular, showing pronounced disadvantages compared with White Africans in both the immigrant and second generations.

Keywords

Africa immigrant adaptation international migration racial stratification segmented assimilation socioeconomic assimilation

Between 1970 and 1990, more than 250,000 Africans migrated to the United States, and by 2010, the U.S.-born children of these immigrants had begun to reach adulthood. Theirs are important stories for our understanding of racial stratification and immigrant adaptation in the post–civil rights era United States, as they include the first substantial waves of voluntary migration of Black Africans to the United States. The concomitant migration of their White African (e.g., South African) counterparts makes for a unique natural experiment: a chance to compare the intergenerational trajectories of Black and White voluntary immigrants from Africa. This is a critical test of whether the well-documented pattern of second-generation ascent is affected by the Black/White (or perhaps Black/non-Black) divide in American race relations. Any such test must also take into account gendered patterns of incorporation that have been illuminated in prior research (see Hondagneu-Sotelo 2003). International migration always engenders some amount of struggle, but those struggles may be compounded for Black African men entering a society characterized by a widespread apprehension of Black masculinity (Armour 2000; Bonilla-Silva 2001; Glassner 2000).

The four largest African national-origin groups before 1990 were Egyptians, most of whom identify as White but who are often identified by others as non-White and/or Arab; Ethiopians and Nigerians, who most often identify as Black; and South Africans, who typically identify as White.¹ In this article, I assess African immigrant socioeconomic assimilation (Neidert and Farley 1985) by examining patterns of poverty and affluence among young adult immigrants from Egypt, Ethiopia, Nigeria, and South Africa in 1990 and U.S.-born young adults of those ancestries 20 years later. Specifically, I address the following questions: (1) Was the African second generation any better or worse off in 2008–2012 than the African immigrants of their parents’ generation in 1990? (2) Do the answers to this question vary by race or national origin? and (3) What does this tell us about the respective roles of race and national origins in shaping immigrant socioeconomic trajectories in the post–civil rights era United States?

Background: Voluntary Migration from Africa to the United States

Because of immigration laws designed, in part, to prohibit Black migration to the United States (Bashi 2004) and the selective nature of migration flows from distant locations, the early waves of voluntary migration from Africa comprised highly educated and skilled people (Arthur 2000; Djamba 1999; Gordon 1998; Halter 2007; Kusow 2007; Marrow 2007). Portes and Rumbaut (2001), however, pointed out that the ability of immigrants to translate their human capital into a commensurate quality of life (and pass it on to their children) depends, in part, on how they are received by the government and by the general public in the receiving society. On these bases, there is good reason to believe that African immigrants of different national origins will exhibit different socioeconomic trajectories in the United States despite having similarly favorable human capital profiles.

The Immigration Act of 1965 and its now famous Hart-Celler Amendments were enacted in 1968, making way for the first substantial waves of voluntary migration from Africa to the United States (Reimers 2005). Because legal migration from most African countries had been barred in the 1920s (Bashi 2004; King 2000), the “family reunification” criterion was of no use for most prospective migrants from Africa. Instead, prospective immigrants from African countries relied upon occupational criteria. As a result, early African immigration flows consisted disproportionately of college and university students, highly skilled professionals and later, their families (Arthur 2000; Butcher 1994; Gordon 1998). The Immigration Act of 1965 was the first but not the only change to U.S. immigration policy that facilitated new flows of migration from Africa. The enactment of the Refugee Act of 1980 made way for the migration of more than 20,000 Ethiopians into the United States in the decade that followed. Most African countries, however, did not benefit from the Refugee Act before 1990. Ethiopians’ refugee status speaks to the context of their departures from Ethiopia and the context of their reception in the United States, both of which may distinguish their patterns of adaptation in the United States from those of other African immigrants (Portes and Rumbaut 2001).

All four countries (Egypt, Ethiopia, Nigeria, and South Africa) were profoundly influenced by British colonization efforts in the nineteenth and twentieth centuries, and the English language has been taught in all four countries since World War II. However, they each differ in their historical and geopolitical positioning vis-à-vis the United States in some important ways (Davidson 1994; Wallerstein 1986). The coastal and interior regions of modern Nigeria, for instance, were at the epicenter of the trans-Atlantic slave trade (Rodney 1981), meaning that many Nigerians share distant kin with some slave-descended Black Americans. Whether members of either group recognize or value this possible connection is not clear, but the high incidence of intermarriage between Nigerian men and U.S.-born Black (not Nigerian) women suggests that many do.²

At the other end of the spectrum on this count are South Africans who descend from mainly European settler colonists. Although some fled South Africa in response to apartheid policy, which they found morally reprehensible (Marrow 2007), they may find themselves least differentiated from and most comfortable with White people of U.S.-born parentage. Furthermore, given their mainly northern and western European ancestry, their most obvious proximal host group (Mittelberg and Waters 1992) is White. Egyptians and Ethiopians occupy an intermediate position in this respect, neither population having obvious linkages to the African communities ravaged by the trans-Atlantic slave trade. Rather, their historic ties to Mediterranean and Arabian Sea trades have left them with phenotypic traits and identities that defy easy categorization in the U.S. context (Kusow 2007). The entry of most Ethiopians as refugees further complicates matters.

In the second generation, many White South Africans may think of themselves and be thought of as White—nothing more, nothing less—whereas Nigerians may come to think of themselves and/or be thought of by others simply as Black. Egyptians and Ethiopians may have more ambiguous racial identities, and all of this may bear significantly on their experiences and socioeconomic trajectories as the impacts of racial stratification on opportunity in the post–civil rights era United States are well documented (Farley and Allen 1987; Pager and Shephard 2008).

Literature Review: Race, Origins, and Immigrant Adaptation

In this article I gauge the relative impacts of national origins and race on the prevalence of affluence and poverty among African immigrants, controlling for a few widely accepted predictors (see Iceland 2006): elaborating exhaustive statistical models for those outcomes is not among my objectives. My review of the literature, instead, focuses on the impacts of national origins and racial group membership on life chances of immigrants to the United States, not on the determinants of poverty and affluence.

The U.S.-born children of post–civil rights era African immigrants are part of the “new second generation,” whose prospects for integration into the American mainstream are the subject of much debate. Some argue that assimilation continues to draw immigrants into an ever evolving American mainstream even as the sources of immigration have shifted away from Europe and toward Asia, Africa, the Caribbean, and Latin America (Alba and Nee 2003; Alba, Kasinitz, and Waters 2011). Others suggest that race- and class-based patterns of exclusion may shut many members of the “new second generation” out (Bashi and McDaniel 1996; Haller, Portes, and Lynch 2011).

Evidence on this question is mixed, and depends partly on whether men, women, or both are being considered. Field work and interviews conducted in a New York City high school by Nancy Lopez (2002) uncovered evidence that the experiences and prospects of young men of color were different and, in important ways, less favorable than those of their female classmates. Young men were often criminalized by school personnel. Young women, on the other hand, were more often favored by teachers and school personnel, and steered into school-to-work programs. Feliciano and Rumbaut (2005) found corroborating evidence in survey data gathered in California that revealed greater educational expectations and achievements among the young women of the second generation than was true of their male counterparts.

African immigrants have been largely overlooked in debates about the incorporation of the new second generation. This may be due to their small absolute numbers and the fact that the majority of U.S.-born children of African immigrants are too young to have established their own careers, families, and/or households. However, there is a burgeoning African second generation now entering adulthood that is part of the new second generation whose patterns of advancement have been the subject of much scholarly attention. We should not be surprised to find that race, national origins, and gender influence their experiences and trajectories.

Using data from the 1998 and 2000 Current Population Surveys, Farley and Alba (2002) compared the socioeconomic outcomes of U.S.-born children of several immigrant groups with those of their native (i.e., “third+-generation”) White, Asian, Black, Hispanic, and American Indian cohorts. In their study, immigrant groups defined by regional origins were compared with native groups defined by race. As discussed earlier, there is good reason to do so, because the experiences of immigrants are surely influenced by where they came from and the characteristics of their particular nations of origin (Portes and Rumbaut 2001). However, the experiences of immigrants may be no less influenced by racial identities that are often imposed on them than by their national origins which are often more central to how they see themselves (Bashi and McDaniel 1996; Lopez 2002).

Portes and his colleagues (Portes and Rumbaut 2001; Portes and Zhou 1993) have argued that assimilation processes may lead different groups in different directions, both socially and economically, on the basis of how they are received by the government, the general public, and any established coethnic communities. Differences on these dimensions (government, societal, and coethnic community receptivity) may lead to very different paths for the four African immigrant groups under study here. Government and coethnic community receptivity are tied to the particular circumstances of specific national origins groups, but societal reception may not be. In the United States, there is little awareness or knowledge of national or ethnic distinctions among African peoples, so how they are received may rest heavily on their perceived belonging in some or another widely recognized racial group.

It is, therefore, of some consequence that most Ethiopians arrived as refugees after 1980 and were beneficiaries of government sponsored resettlement programs (Hein 1993) but may often have been seen as Black in a society whose institutions deny Black people critical opportunities as a matter of course (Feagin 2006). Furthermore, because of the lack of long-established coethnic communities to soften their landings in the United States, early Ethiopian immigrants may have found themselves particularly vulnerable to nativist and racist conventions. In these ways, their path may depend on both their specific national origins and their perceived position in the U.S. racial hierarchy.

Farley and Alba (2002), nonetheless, uncovered a nearly universal pattern of marked intergenerational improvement among post–civil rights era immigrants. Adult U.S.-born children of immigrants (25–39 years of age) from Asia, Europe and Canada, South America, Afro-Caribbean, and Central America outperformed immigrants (50 years of age and older) of those origins in terms of educational attainment, occupational attainment, and income. However, possible implications of race and gender are not explicitly addressed.

The striking images of upward mobility in nearly all of the immigrant groups in Farley and Alba’s (2002) study are based on the average achievements of men and women in those groups, when, in fact, achievements and trajectories of men and women in those groups differ. The gendered nature of immigrant socioeconomic assimilation was revealed in multivariate analyses run separately for men and women later in Farley and Alba’s study. They demonstrated that where occupational attainments are concerned, Afro-Caribbean men were at a significant disadvantage compared to other men in the study, but the same was not true of Afro-Caribbean women. When treated as a single group, Afro-Caribbean men and women appeared to be faring nearly as well as non-Hispanic Whites, but this aggregation (of men and women) obscures the fact that Afro-Caribbean men were at a substantial disadvantage. If women of a particular group exhibit a sharp upward intergenerational trajectory and men of that group exhibit an equally sharp downward trajectory, their movements taken together cancel each other out and leave us with the impression of no mobility in either direction. This aggregate trajectory would misrepresent the experiences of both women and men in that group. I am, therefore, compelled to examine the trajectories of African immigrant men and women separately in the analyses to follow.

On the matter of race, Farley and Alba (2002) found that “Spanish Caribbean” men and women between the ages of 25 and 39 years in the study outperformed, by substantial margins, foreign-born “Spanish Caribbean” men and women 50 years of age and older. However, it is not clear that this pattern of advancement is equally pronounced among Spanish Caribbeans whose U.S.-born members are identified as “Black” as opposed to those identified as “White.” Qualitative and quantitative accounts suggest that it is not (Denton and Massey 1989; Golash-Boza and Darity 2008; Ojito 2001). It has been argued that some Latinos and some Latino groups have become “racialized” (see Massey 2009; Telles and Ortiz 2008), which is to say that they will have to overcome not only structural barriers associated with their typically humble origins but also a prevailing perception of their inherited and immutable incompetence.

Racialization, whereby some immigrants are systematically excluded on the basis of the idea that they are fundamentally unassimilable, is sometimes offered as an alternative to assimilation (Golash-Boza 2006). Bashi and McDaniel (1996) pointed out that although much of the theoretical and empirical work on immigrant adaptation revolves around the incremental march of ethnic groups toward undifferentiated Americanness, racial groupings may be more salient as immigrants interface and adapt to their new society. They suggest that to assimilate in the United States is to find one’s place in its centuries-old racial hierarchy, and for many, “one’s place” is not chosen but imposed. Each immigrant group is pushed and/or pulled into a domestic proximal host with whom its members share (mainly) phenotypic traits, facilitating their integration into certain segments of U.S. society (Kasinitz et al. 2008; Mittelberg and Waters 1992). Like assimilation, racialization operates both within and across generations. Although Black African immigrants may sometimes be seen by others as “not really Black,” their U.S.-born children will tend to be seen and, perhaps, come to see themselves as “just Black” (Butterfield 2004).

To the extent that this is true, it is reasonable to predict that White South African immigrants will often settle into White contexts and have more or less “White experiences.” Nigerians and perhaps Ethiopians may more often settle into Black contexts and face many of the challenges associated with Black American group membership. By 2000, a clear pattern of residential segregation was emerging whereby West African (D = .79)³ and East African (D = .80) immigrants were substantially more segregated from White Americans than native Black Americans (D = .64) in their cities (Iceland 2009). This does not mean that Nigerians and Ethiopians share neighborhoods with each other or with other Black people, but it does mean that they do not share neighborhoods with White Americans. This may reflect an entrenched pattern of Black exclusion in urban housing markets that has less effect on non-Black immigrants. Research on other Black immigrants bears this out (Bryce-Laporte 1972; Butterfield 2004; Waters 1999). It has been established, for instance, that residential segregation is more pronounced among Black Caribbean immigrants than it is among White Caribbean immigrants (Denton and Massey 1989; Iceland and Scopolliti 2008), often leaving the former concentrated in neighborhoods with limited educational and occupational opportunities.

There is also a growing body of research literature that traces the differential monetary returns to education across Black and White immigrant groups. Dodoo and Takyi (2002) found that even when human capital and other pertinent background characteristics are held constant, African-born White men earn significantly more per hour (and per year) than their identically qualified Black African counterparts. The racial earnings gap among African immigrants in the United States has, therefore, not been due to Black African deficits but to the subpar “returns” Black Africans receive on their investments in education and other characteristics related to earnings. Nawyn (2012) traced the earnings of cohorts of White and Black African immigrant men across the 1990 and 2000 U.S. censuses and found that this Black African disadvantage persists long after their arrival in the United States. Djamba and Kimuna (2011) showed that the racial earnings gap uncovered in these studies remained significant as recently as 2008, even though Black African men are less often unemployed than others. It would seem that Black African men have relatively little difficulty finding employment—a fact that is probably reflective of employer preferences for immigrant workers in some sectors (Waldinger and Lichter 2003; Moss and Tilly 1999)—but are hard pressed to find jobs with pay that is commensurate with their educational attainments and occupational skills. Black African women, however, appear not to face such disadvantages (Djamba and Kimuna 2012)—more evidence of gendered socioeconomic trajectories among Black immigrants. Whether these patterns have been passed on to Black men and women of the African second generation is a question yet unanswered.

The disadvantages Black African immigrants face in U.S. labor markets may bear directly and indirectly on the experiences of their U.S.-born children as they age into adulthood. The African second generation includes U.S.-born Whites who were raised in predominately White communities by parents paid commensurately to their background characteristics; it also includes U.S.-born Blacks raised in predominantly Black communities (see Iceland 2009) by parents who often made considerably less money than their educational attainments would have predicted. These childhood differences may translate into more and better schooling for the children of White immigrants from Africa than for the children of Black immigrants from Africa, leading to human capital differentials in the second generation despite first-generation human capital similarities. As these young people venture into U.S. labor markets, they may be looked upon simply as young White and Black Americans despite their African origins. To the extent that this is true, documented patterns of Black exclusion in U.S. labor markets (Pager 2007; Wilson 2009) will compound the effects of uneven educational quality and attainment across White and Black members of the African second generation, leading, ultimately, to differing levels of poverty and affluence between them.

On the basis of previous research, it is reasonable to expect that whatever patterns of intergenerational advancement or stagnation we observe among African immigrants will be both “raced” and “gendered.” Patterns of upward mobility will likely be more pronounced across generations of women than men and more pronounced across generations of White African immigrants than across generations of Black African immigrants, which will translate into more favorable trajectories for Egyptians and South Africans than for Ethiopians and Nigerians. For members of the “African second generation,” patterns of racial identity may be no less influential than their specific national origins.

Data and Methods

Because people of African birth and/or known African ancestry make up less than 1 percent of the U.S. population, only the U.S. census and American Community Surveys are large enough to yield African samples sufficient for the purposes of this study. I use the 5 percent public-use sample from the 1990 census and an aggregate American Community Survey data file of 1 percent samples from 2008, 2009, 2010, 2011, and 2012 (Ruggles et al. 2014) to (1) generate first- and second-generation African immigrant samples and (2) carry out statistical tests to assess the impacts of race and national origins on the fiscal well-being across two generations of African men and women.

National Origins, Race, and Nativity Measures

The central independent variables in this study are race, national origins, and nativity. The census long form and American Community Survey ask respondents a closed-ended question about their racial group membership and open-ended questions about their ancestry or ethnic origins and their places of birth (which U.S. state or foreign country). I use answers to the closed-ended “race” question to fashion a trichotomous measure of racial identity. The vast majority of African immigrants identify as White or Black (only), with a small residual category composed of those who identify as Asian, “other,” or multiracial. I use open-ended questions to identify African first- and second+-generation samples, that is, to identify African immigrants (first generation) and U.S.-born adults of specified African ancestries (second+ generation).

Because the U.S. census no longer asks respondents where their parents were born,⁴ accurately identifying the children of immigrants is impossible once they no longer reside with their immigrant parents (Hirschman 1994). However, substantial migration from Egypt, Ethiopia, Nigeria, and South Africa has commenced only in recent decades, so U.S.-born adults who identify with any one of those ancestries are likely the children of immigrants from one of those countries and, therefore, members of the African second generation. For the purposes of this study, I treat immigrants who came from Egypt, Ethiopia, Nigeria, and South African before 1990 as members of the African first generation and U.S.-born respondents of those ancestries (observed in 2008-12) as members of the African second+ generation. Table 1 provides sample counts for each national origin across the first- and second+-generation groupings.

Table 1.

Sample Sizes by National Origins and Nativity.

Nationality	Men		Women
	First Generation	Second+ Generation	First Generation	Second+ Generation
	Foreign Born, 25–39 Years Old in 1990	U.S. Born, 25–39 Years Old in 2008–2012	Foreign Born, 25–39 Years Old in 1990	U.S. Born, 25–39 Years Old in 2008–2012
Egyptian	518	324	315	331
Ethiopian	226	62	187	51
Nigerian	552	185	266	200
South African	87	84	82	64
Total	1,383	655	850	646

Sources: 1990 U.S. census 5 percent public-use file and 2008–2012 American Community Survey aggregate file (Ruggles et al. 2014).

This method is not perfect. First, the immigrant or first generation observed in 1990 includes some parents of U.S.-born children who would age into adulthood by 2008, but it also includes some immigrants who were not and never would be parents, as well as some parents of children too young or too old to be included in my analyses of the African second generation. Second, the racial composition of the national origins/generational groupings poses some problems. Figure 1 shows that most but not all Egyptian and South Africans identify racially as White, while most Ethiopians and Nigerians identify racially as Black.⁵ This makes it impossible to separate the effects of national origins and racial identity. Furthermore, changing patterns of racial identification across the first and second generations in all four groups call into question what exactly is being captured by the closed-ended race question.

Figure 1.

Racial Identification by National Origins across First- and Second-generation African Immigrants.

Third and finally, the method identifies a “second+-generation” rather than a pure “second-generation” sample in that it may include a small number of third- and perhaps fourth-generation individuals who descend from the earliest immigrants from the countries of origin specified here but who cannot be identified as such. Of course, some children or descendants of Egyptian, Ethiopian, Nigerian, and South African immigrants may not answer the ancestry question at all,⁶ and those who do may not offer responses that would align with their parents’ places of birth. Some children of South African immigrants may say “British” when asked about their ancestry, some second-generation Nigerians may say that their ancestry is “African,” and Egyptian second-generation respondents may sometimes identify “Arab” rather than “Egyptian” ancestry. All three of these cases would be missed by the method of identification used here, and little can be done to detect such oversights. This could lead to an overestimation of second-generation achievement because more accomplished immigrants and children of immigrants have been shown more often to retain their immigrant (national) identities, while their less successful counterparts are more likely to cast off their (parents’) national origins in favor of pan-ethnic or racial labels (Portes and Rumbaut 2001; Waters 1999), rendering them invisible to studies such as this one.

The only thing we can say with certainty about the first generation is that they are U.S. residents who were born in one of the four African countries considered here. It is accurate to refer to them as the first or immigrant generation but not entirely correct to call them the parental generation. With regard to U.S.-born members of the four ancestry groups, it is accurate to refer to them as the second+ generation but not accurate to refer to them as the direct descendants (children) of the first generation. In the end, we cannot say whether second-generation children are doing better or worse than their own immigrant parents, but we can say with confidence how much better or worse the second+ generation is doing compared with the first. I proceed with this limitation in mind.

Poverty and Affluence Measures

The greater part of the analysis here consists of comparisons of affluence and poverty prevalence across two generations of men and women of the four African nationalities. Individuals in families whose incomes are at or below the federally determined poverty threshold will be treated as “poor,” while those in households whose incomes are five times the poverty threshold value or more (following Farley and Alba 2002) will be treated as “affluent.” In 2012, the poverty threshold for a family of four was $23,492 (U.S. Census Bureau 2014). Therefore, respondents living in families of four with incomes at or below this level are considered poor for the purposes of this study, and respondents living in families of four with incomes at or above five times that same threshold amount ($117,460) are considered affluent. Threshold values are based on family size and composition (i.e., number of children less than 18 years of age). I have chosen to use these outcomes because they take into account family rather than individual income as well as family structure to give a more complete sense of (fiscal) well-being than income alone. In this way, I compare patterns of poverty and affluence across first- and second+-generation Egyptians, Ethiopians, Nigerians, and South Africans with U.S.-born White and Black non-Hispanics as “native” reference groups.

Preliminary analyses revealed a universal pattern of intragenerational advancement whereby African immigrant men and women who were between the ages of 45 and 59 years in 2008–2012 were considerably more educated, had better jobs, and had higher annual incomes than they had (between the ages of 25 and 39 years) in 1990. Given these intragenerational advancements, the most valid intergenerational comparisons are to be made between young adult (25–39 years old) members of the African first generation in 1990 and young adult (25–39 years old) members of the African second+ generation in 2008–2012.⁷

Analytical Strategy

The analyses to follow consist, first, of simple comparisons of affluence and poverty across generational groupings in each national-origin group and each racial category. Second, I assess relative influences of national origins and racial group membership by taking a closer look at patterns of affluence and poverty across the generational groupings. Because both affluence and poverty are operationalized as dichotomous outcomes, I use logistic regression to identify statistically significant predictors of each. Without having sufficient numbers of “White Nigerians” or “Black Egyptians,” for instance, the question of which matters more, race or national origins, is impossible to adjudicate statistically. However, we can determine how much explanatory power we would miss if all we knew about African immigrants was how they identified racially and knew nothing of their national origins.

To answer this question, I first regress the dichotomous measure of affluence on age and education to establish a baseline (human capital) model with no consideration of race or national origins. Second, I repeat step 1 with race terms added to the model. Third, step 2 is repeated with the national-origin and ancestry indicators replacing the race indicators. These steps make it possible to see how much better affluence can be predicted when race or national origins or ancestry is added to the baseline model and come to an answer about which attribute is more predictive of affluence net of age and education. I repeat these steps for immigrant men in 1990, immigrant women in 1990, U.S.-born men in 2008–2012, and U.S.-born women in 2008–2012. Finally, I follow the same procedure with poverty as the dependent variable. In the end, I compare odds ratios and model fit statistics generated in the steps above to answer the following empirical questions: Is there appreciable improvement and/or convergence in the odds of affluence and poverty across African and non-African groups as we move from the first generation in 1990 to the second+ generation in 2008–2012? Does race contribute to model fit any more or less than national origins once the effects of age and education are accounted for?

To answer the second of these questions I calculate the proportional improvement to model fit (PIMF) associated with the addition of race indicators to a logistic regression model that estimates one’s odds of affluence (or poverty) solely on the basis of his or her age and education. I do so by locating the pseudo-r² value for the logistic regression model that includes race, age, and education and dividing it by the pseudo-r² value for the logistic regression model that includes only age and education. By subtracting 1 from the resulting value and multiplying by 100, we arrive at the percentage increase in pseudo-r² with the addition of race.⁸ If the baseline model yields a pseudo-r² value of .05 and the model that includes race yields a pseudo-r² value of .08, PIMF = (.08/.05 – 1) × 100 = 60 percent, meaning that the addition of race indicators improves the model fit by 60 percent. Next, I repeat these steps to assess the PIMF associated with the addition of national-origin and ancestry terms, and finally the two PIMF values, for race and national origins, are compared.

The contributions of race and national origins are not assessed net of each other. Instead, I compare the predictive power of the model with race added to the predictive power of the model with national origins (instead of race) added, and the largest pseudo-r² value and PIMF figures are taken to reflect a stronger effect. Because prevailing theories predict that immigrants will gravitate toward “host” or “proximal host” group characteristics, White non-Hispanic and Black non-Hispanic reference groups are included for purposes of comparison in all models. Ultimately, I draw conclusions about the extent of socioeconomic assimilation observed across generational groupings (Neidert and Farley 1985) and how it varies across racial and ancestry groups of recent African origin.

Results

Before turning to the achievements of African second+-generation men and women, it is instructive to examine the circumstances that shaped their childhoods. Table 2 displays select characteristics of the parents and households of the African second generation in 1990. There is significant variation across African immigrant groups and native White and Black referents. Although about 5 percent of U.S.-born children were being raised by foreign- born parents in 1990, no fewer than 60 percent of children with known African ancestry had immigrant parents. The vast majority of these children are members of the new second generation. Although two-parent families are typical for all African immigrant groups, Nigerian American children are considerably less likely to reside with both parents than children of other African origins. The relatively small share of Nigerian American children with foreign-born parents (61 percent) may reflect a pattern of intermarriage, childbearing, and divorce and separation between Black Americans and Nigerian immigrants. In 1990, more than a third of Nigerian married men were married to U.S.-born Black women with no known Nigerian ancestry.⁹ Many of these unions produced children and some surely ended in divorce, leaving some young members of Nigerian second generation with single U.S.-born Black mothers.

Table 2.

Household and Parental Characteristics of U.S.-born Children (0–17 Years Old) of Specified African Ancestries, 1990.

	% Living with One or More Foreign-born Parents	% Living with One or More College-educated Parents	% Living at or below the Poverty Line	% Living at or above 5 Times the Poverty Line	Median Family Income	n	Weighted n
Egyptian	90	70	9	34	$62,495	919	19,230
Ethiopian	76	43	18	14	$35,432	177	3,899
Nigerian	61	57	30	8	$31,327	1,211	28,117
South African	80	76	11	50	$77,533	158	3,217
U.S.-born non-Hispanic White	5	31	12	17	$43,234	2,326,963	43,785,072
U.S.-born non-Hispanic Black	4	11	41	5	$24,283	420,146	8,412,785

Source: 1990 U.S. census 5 percent public-use file (Ruggles et al. 2014).

Although children of the African second generation were more often raised by college-educated parents than other U.S.-born children in 1990, an inordinate number of them were also exposed to poverty. All of this translates into highly variable family incomes across the four immigrant and two native groups. Egyptian and South African children resided in households with median incomes considerably higher than those of native non-Hispanic White families; Ethiopian and Nigerian children resided in households with incomes lower than those of native non-Hispanic Whites but considerably higher than those of native non-Hispanic Black families. Have these disparities followed members of the African second+ generation into adulthood?

A Bivariate Look at Affluence and Poverty across African Immigrant Generations

Tables 3 and 4 display affluence and poverty rates for the African and non-African groups, along with percentage change between 1990 and 2008–2012. Because the standard errors are large for some of the African groups, many of the apparent intergenerational changes are not statistically significant; but there are several statistically and substantively significant patterns shown in the tables.

Table 3.

Prevalence of Affluence by Race, Nationality, Sex, and Origins.

	First-generation Men in 1990	Second+-generation Men in 2008–2012	First- to Second+-generation Change	First-generation Women in 1990	Second+-generation Women in 2008–2012	First- to Second+-generation Change
Race × U.S. Origins
White non-Hispanic	.259^b	.294^b	13%*	.236^b	.275^b	17%*
n	1,033,481	723,743		1,035,825	710,347
Black non-Hispanic	.117^w	.126^w	8%*	.091^w	.116^w	28%*
n	119,958	110,018		141,807	106,294
Race × African Origins
White African	.330^{w b}	.548^{w b}	66%*	.354^{w b}	.548^{w b}	55%*
n	601	372		395	363
Black African	.149^{w b}	.303^b	103%*	.130^{w b}	.379^{w b}	192%*
n	770	241		446	248
Nationality × African Origins
Egyptian	.307^{w b}	.519^{w b}	69%*	.324^{w b}	.541^{w b}	67%*
n	518	324		315	331
Ethiopian	.159^w	.210	32%^b	.182	.216	19%
n	226	62		187	51
Nigerian	.141^w	.351^b	149%*	.083^w	.410^{w b}	396%*
n	552	185		266	200
South African	.471^{w b}	.548^{w b}	16%	.500^{w b}	.484^{w b}	–3%
n	87	84		82	64

Sources: U.S. census 1990 5 percent public use sample and 2008–2012 American Community Survey aggregate file (Ruggles et al. 2014).

Group differs significantly from U.S.-born White non-Hispanics (α = .05).

Group differs significantly from U.S.-born Black non-Hispanics (α = .05).

Statistically significant intergenerational change (α = .05).

Table 4.

Prevalence of Poverty by Race, Nationality, Sex and Origins.

	First-generation Men in 1990	Second+-generation Men in 2008–2012	First- to Second+-generation Change	First-generation Women in 1990	Second+-generation Women in 2008–2012	First- to Second+-generation Change
Race × U.S. Origins
White non-Hispanic	.072^b	.106^b	46%*	.092^b	.116^w	26%*
n	1,033,481	723,743		1,035,825	710,347
Black non-Hispanic	.237^w	.377^w	59%*	.311^w	.316^w	2%*
n	119,598	110,018		141,807	106,294
Race × African Origins
White African	.107^{w b}	.097^b	–9%*	.106^b	.044^{w b}	–59%*
n	601	372		395	363
Black African	.149^{w b}	.237^{w b}	58%*	.148^{w b}	.113^b	–24%*
n	770	241		446	248
Nationality × African Origins
Egyptian	.106^{w b}	.124^b	16%*	.114^b	.051^{w b}	–55%*
n	518	324		315	331
Ethiopian	.212^w	.355^w	67%	.171^{w b}	.235^w	38%
n	226	62		187	51
Nigerian	.130^{w b}	.189^{w b}	45%*	.147^{w b}	.085^b	–42%*
n	552	185		266	200
South African	.103^b	.060^b	–42%	.061^b	.063^b	2%
n	87	84		82	64

Sources: U.S. census 1990 5 percent public use sample and 2008–2012 American Community Survey aggregate file (Ruggles et al. 2014).

Group differs significantly from U.S.-born White non-Hispanics (α = .05).

Group differs significantly from U.S.-born Black non-Hispanics (α = .05).

Statistically significant intergenerational change (α = .05).

Turning first to affluence, the percentage of men and women living at or above five times the poverty threshold increased across African and non-African groups alike, but improvements were more pronounced in some groups than in others. Whereas affluence ticked upward by just a few points for Whites and Blacks of U.S. stock, the improvements among White and Black Africans were large and statistically significant. Although affluence appears to have more than doubled among Black Africans, White Africans still held a considerable advantage over their Black African counterparts in the second+ generation. Nigerian men and women made the most impressive gains on this measure but fell well short of the affluence rates of Egyptian and South African second+ generation men and women. All three of these groups exhibit affluence rates that are substantially higher than those of White non-Hispanics of the same age. Ethiopian men and women of both generations seem to have stalled on this measure at levels between Black and White Americans of U.S. stock. In all, there seems to be good news in Table 3 for all groups except Black Americans and Ethiopians.

There is less good news where the prevalence of poverty is concerned. White Africans and, in particular, Egyptian women are the only groups to show statistically significant reductions in poverty across the generations. Black American and Black African men experienced large statistically significant increases in poverty across the generations. Also, although not statistically significant, the data suggest considerably more poverty among Ethiopian and Nigerian men moving from the first to the second+ generation.

In all, a story of increasing affluence and declining poverty emerges in Tables 3 and 4 for White African men and women. A story of more affluence but also more poverty emerges for Black Africans with the passing of generations, Nigerian women being a noteworthy exception. High (and increasing) levels of poverty seen among African men of the second+ generation are a cause for concern and call into question expectations that extend from older articulations of assimilation theory.

Race, National Origins, and Affluence among African Immigrants

Social scientists have often expected immigrant groups to move toward socioeconomic profiles resembling those of the dominant or host group with the passing of generations (Neidert and Farley 1985). Therefore, I turn now from asking whether the second+ generation men and women in 2008–2012 differ significantly from first-generation men and women in 1990 to asking (1) whether African immigrant groups compare any more favorably with White Americans in the second+ generation than was true in the first generation and (2) how the effects of race and nationality compare across the generations.

I use logistic regression to estimate affluence and poverty prevalence across the generational groups net of age and education. When the effects of those background characteristics are netted out, Black Africans seem to be approaching parity with White Americans with the passing of generations, and White Africans seem to have outperformed White Americans.

Figure 2 displays patterns of affluence across first- and second+-generation African immigrants with White and Black Americans as points of comparison. Each line on the figure originates at the level of advantage or disadvantage vis-à-vis White Americans in the first generation (1990) and terminates at the level of advantage or disadvantage observed in the second+ generation. The advantages or disadvantages are quantified as odds ratios (see Table A1 for logistic regression results). White triangles at either end of each line indicate statistically significant differences between the group in question and the White American reference group.

Figure 2.

Adjusted Odds Ratios of Affluence among Men, 1990 to 2008–2012.

In the left-hand panel of Figure 2, African immigrants are compared on the basis of their racial identities, with no attention to their national origins. The solid gray line in the middle of the graph represents the position of White Americans relative to all other groups. The solid black line running beneath and parallel to the solid line .5 above it (on the y-axis) indicates that Black American men were half as likely to be affluent as White American men at both points in time. The upward-sloping dotted lines suggest that the positions of both Black and White African immigrants improved vis-à-vis White and Black Americans with the passing of generations. Black African men were significantly less likely to be affluent than White American in 1990, but Black African men of the second+ generation were no less likely than White American men to be affluent. White African men were significantly less likely than White American men to be affluent in 1990, but White African second+-generation men were significantly more likely than White American men to be affluent two decades later.

These findings are echoed in the right-hand panel, which illustrates the effects of nationality and ancestry on affluence across the generations. All African nationalities seem to compare more favorably with White Americans with the passing of generations. The predominantly Black groups, Ethiopians and Nigerians, start with considerably lower levels of affluence than White American in the first generation but do not differ significantly in the second+ generation. The predominantly White Egyptian and South African groups both exhibit significantly higher levels of affluence than their White American counterparts in 2008–2012. These racial patterns are mirrored among women in Figure 3.

Figure 3.

Adjusted Odds Ratios of Affluence among Women, 1990 to 2008–2012.

These odds ratios, however, do not speak to the relative importance of racial identification and nation origins across generations. To address this question, I have included a measure of PIMF, as discussed earlier, to show how much improvement there is to the model fit when race or national-origin indicators are introduced into a human capital (age and education only) model predicting affluence. In Figure 2, we can see that the PIMF associated with race among first-generation African men is 143 percent, which is lower than the PIMF associated with national origins and ancestry (172 percent), suggesting that the latter is more predictive of affluence than the former. The same appears to be true among women in Figure 3. PIMF values decline with the passing of generations, suggesting that there is intergenerational decline in the importance of both race and national origins where affluence rates are concerned. However, both attributes continue to contribute significantly to patterns of affluence, and the PIMF figures suggest that race is equally if not more predictive of affluence among members of the second+ generation.

Race, National Origins, and Poverty among African Immigrants

The relationships among race, national origins, and poverty are depicted in Figures 4 and 5, which tell a more mixed story. Turning first to the effects of racial group membership among men, White and Black African men seem to experience net declines in poverty but are significantly more likely to be poor than their White American counterparts in both the first and second+ generations. Black African men, in fact, look much more like Black American men than any other group, with poverty rates more than three times as great as those of similarly educated White American men in both 1990 and 2008–2012. Shifting our view to the right of Figure 4, we can see that the Black African disadvantage that persists into the second+ generation reflects what appears to be increasing poverty across generations of Ethiopian men. By the second+ generation, there is still considerable difference between African men on the basis of both race and national origins. Replacing one with the other in a model predicting poverty yields very little improvement (or loss) with regard to model fit.

Figure 4.

Adjusted Odds Ratios of Poverty among Men, 1990 to 2008–2012.

Figure 5.

Adjusted Odds Ratios of Poverty among Women, 1990 to 2008–2012.

Figure 5 shows that among first-generation women, race does not significantly improve model fit despite statistically significant race coefficients. However, sharp declines in White African women’s poverty across the generations leaves them in a positon not significantly different from White American women in 2008–2012, while Black African women of the second+ generation seem to have stagnated at poverty levels about twice as high as those of White American women. In the left panel of Figure 5, we see that White African women’s advantage is reflective of a sharp decline in the Egyptian odds ratio across generations and the persistence of South African women at levels of poverty not significantly different from those of White American women. The disadvantage of Black African women in the second+ generation appears to be driven entirely by the stagnation of Ethiopian women’s poverty rates at levels twice or more as high as similarly educated White American women. Interestingly, the only statistically significant improvements to models predicting poverty among women are associated with race in the second+ generation. National-origin indicators improved overall model fit in neither the first nor the second+ generation. Although race indicators did not significantly enhance our ability to predict poverty among African immigrant women in 1990, they did among second+ generation African women, suggesting that the salience of race increased with the passing of generations.

In all, we are left with considerable evidence that both race and national origins continue to bear significantly on patterns of affluence and poverty into the African immigrant second generation in some ways that are fairly predictable and others that are not.

Discussion and Conclusions

Because African immigrant groups are among the most diverse and fastest growing in the United States, it behooves us to pay close attention to their patterns of adjustment. In this study, socioeconomic assimilation of African immigrant groups is assessed by comparing patterns of affluence and poverty among Egyptian, Ethiopian, Nigerian, and South African immigrants between the ages of 25 and 39 years in 1990 with those of U.S.-born persons of those ancestries and of those ages in 2008–2012. I find that there is considerable upward mobility across first and second+ generations of African immigrants, but that upward mobility is more pronounced for women than men, more pronounced for White Africans than for Black Africans, and less evident for Ethiopians than for all other groups. Additionally, I find that racial identities are at least as influential in processes of African immigrant adaptation in the United States as specific national origins.

For most African groups, there is more affluence but also more poverty with the passing of generations. In other words, there is more bifurcation evidenced in the second generation than in the first. White African immigrant men and women, most of whom were Egyptian and South African, did not differ greatly from White Americans in terms of affluence in 1990, and in the second+ generation they were significantly more likely to be affluent than their White American counterparts. For them the “convergence” that socioeconomic assimilation implies is absent, and instead, a more beneficial pattern that is perhaps reflective of selective acculturation (Portes and Rumbaut 2001) emerges. Interestingly, the affluence rates of Black African men and women of the second+ generation do converge with those of White Americans though this is more true of Nigerians than it is of Ethiopians. The movement of Black Africans toward the White American mean on this measure is encouraging, but it is too early to tell whether the success of the African second generation will be passed along to the third.

Socioeconomic assimilation is less evident when we turn to the prevalence of poverty across generations. Among men, the differences between White American and both White and Black African groups remain statistically significant into the second+ generation. The only group of second+ generation men able to reach parity with White American men are South Africans. Ethiopian men of the second+ generation appear to have fallen further behind White Americans with rates of poverty equal to or greater than those of Black Americans in both the first and second+ generations.

There is more encouraging news where African women are concerned. White African women of the second+ generation did not differ significantly from White American women in terms of poverty prevalence, nor did Nigerian women of the second+ generation. Among men, only one group achieved parity with White Americans on this measure: the predominantly White South African group. Among women, only one group did not achieve parity: the Ethiopian group. All of this suggests that there is considerable improvement across generations of African immigrants, but to the extent that socioeconomic assimilation is happening, it is driven more by intergenerational advancement into affluence than by advancement out of poverty.

Race seems to matter for African immigrants in the United States, but is it any more or less predictive of fiscal well-being than national origins and ancestry? My results suggest that racial identity is at least as salient as national origins in shaping patterns of fiscal well-being and perhaps even more. With regard to affluence, there is a precipitous decline in the salience of both race and national origins with the passing of generations. Still, both attributes are predictive of affluence in the second+ generation, even when age and education are held constant. For both men and women of the second+ generation, taking account of three racial distinctions (White, Black, and other) boosts model fit more than the inclusion of four national origins categories, suggesting that race is as salient as national origins, if not more so.

With regard to poverty, the effects of race and national origins appear to moderate only slightly across generations of African immigrant men, while they do not moderate at all among women. Among second+ generation men, the inclusion of national origins terms adds slightly more to model fit statistics than the addition of race terms, largely because of the dramatic divergence between Ethiopian and other African men of the second+ generation. In any case, knowing only the racial identities of African second+-generation men would result in a minute (and statistically insignificant) loss in predictive power. Among women, neither race nor national origins add to our ability to predict poverty in the first generation, but in the second+ generation, the inclusion of race adds significantly, whereas that of national origins does not.

Assessing the effects of race and national origins independently of each other is not possible given data limitations, but I can say with certainty that having knowledge of only one or the other does not result in significant losses in terms of our ability to predict affluence and poverty among African immigrants. For members of the African second+ generation, racial identity is no less influential than their specific national origins. This is an important finding because it calls into question the convention in immigration research of focusing on the characteristics and experiences of national origins groups, often referred to as ethnic groups, with no explicit address of racial identities that may be shared within and across them.

I should not, however, be taken as saying that national origins do not matter. As a prime example, there are important differences between Nigerians and Ethiopians, two largely Black groups, that must be considered by any serious student of immigrant adaptation. The Ethiopian disadvantage is large, more pronounced in the second+ generation than in the first, and defies easy explanation. The fact of their phenotypic Blackness is only a part of their story—the rest may have much to do with their mode of entry into the United States largely as refugees. Even this does not explain the second-generation decline observed here—a pattern not observed in other refugee populations—but it does highlight the importance of sending contexts and receiving contexts as experienced by specific national origins groups.

In all, we are left with results suggestive that socioeconomic assimilation is segmented along lines of gender, race, and national origins and ancestry, with Ethiopian men standing out as a particularly challenged group. Their struggle reveals the importance of the specific circumstances of national-origin groups present in immigrant streams from Africa and elsewhere but also the importance of more generalizable “raced” and “gendered” patterns of inclusion and exclusion: “gendered” in that women of the African second generation improved on the achievements of women of their mothers’ generation to a greater degree than men of the African second generation, and “raced” in that White Africans outperformed the White American majority in the second generation, and Black Africans exhibited less favorable and more varied patterns of upward mobility and stagnation. These findings call into question expectations of early assimilation theory and lend credence to those theories of immigrant adaptation which consider more carefully the multiple dimensions of diversity characterizing recent immigration and pay closer attention to the staying power of racism in U.S. society.

Footnotes

Appendix

Table A4.

Logistic Regression Predicting Poverty among Women 25–39 Years Old in 1990 and 2008–2012.

	Women 25–39 Years Old in 1990			Women 25–39 Years Old in 2008–2012
Variable	Exp(B)	Exp(B)	Exp(B)	Exp(B)	Exp(B)	Exp(B)
White non-Hispanic		Ref	Ref		Ref	Ref
Black non-Hispanic		3.892**	3.892**		2.756**	2.756**
African immigrants
White African FB		2.018**			.653
Black African FB		2.225**			1.806**
Other African FB		6.944**			1.906
Egyptian FB			2.298**			.811
Ethiopian FB			1.954**			2.363*
Nigerian FB			2.688**			1.560
South African FB			1.022			.684
Age	.968**	.968**	.968**	.965**	.965**	.965**
No HS diploma	Ref	Ref	Ref	Ref	Ref	Ref
HS diploma	.215**	.226**	.226**	.218**	.230**	.230**
BA/BS degree	.045**	.051**	.051**	.033**	.038**	.038**
Graduate degree	.037**	.042**	.042**	.019**	.022**	.022**
Constant	1.708	1.254	1.254	3.275
Nagelkerke’s pseudo-r²	.128	.179	.179	.186	.215	.215
n	1,178,482	1,178,482	1,178,482	817,287	817,287	817,287
Model fit statistics from analyses including only African first and second generations
Nagelkerke’s pseudo-r²	.100	.107	.108	.228	.262	.251
PIMF		7%	8%		15%**	10%
n	850	850	850	646	646	646

Sources: U.S. census 1990 5 percent public use sample and 2008–2012 American Community Survey aggregate file (Ruggles et al. 2014).

Note: FB = foreign born; HS = high school; PIMF = proportional improvement to model fit; Ref = reference.

p < .05. **p < .01. ***p < .001.

Acknowledgements

I wish to thank Jody Agius-Vallejo, Elizabeth Fussell, Charles Hirshman, Amaha Kassa, and Amanda Seres as well as anonymous reviewers and editorial staff at the Sociology of Race and Ethnicity for their helpful feedback on earlier drafts of this paper. It is a gift to have colleagues as generous as these.

Notes

Author Biography

Amon Emeka is an assistant professor of sociology at Skidmore College. His research focuses on racial inequality and immigrant adaptation.

References

Alba

Richard

Nee

Victor

. 2003. Remaking the American Mainstream: Assimilation and Contemporary Immigration. Cambridge, MA: Harvard University Press.

Alba

Richard

Kasinitz

Philip

Waters

Mary

. 2011. “The Kids Are (Mostly) Alright: Second-generation Assimilation.” Social Forces 89(3):763–73.

Armour

Jody

. 2000. Negrophobia and Reasonable Racism: The Hidden Costs of Being Black in America. New York: New York University Press.

Arthur

John

. 2000. Invisible Sojourners: African Immigrant Diaspora in the United States. Westport, CT: Greenwood.

Bashi

Vilna

. 2004. “Globalized Anti-Blackness: Transnationalizing Western Immigration Law, Policy, and Practice.” Ethnic and Racial Studies 27(4):584–606.

Bashi

Vilna

McDaniel

Antonio

. 1996. “A Theory of Immigration and Racial Stratification.” Journal of Black Studies 27(5):668–82.

Bonilla-Silva

Eduardo

. 2001. White Supremacy and Racism in the Post–Civil Rights Era. Boulder, CO: Lynne Rienner.

Bryce-Laporte

Roy S.

1972. “Black Immigrants: The Experience of Invisibility and Inequality.” Journal of Black Studies 3(1):29–56.

Butcher

Kristin

. 1994. “Black Immigrants in the United States: A Comparison with Native Born Blacks and Other Immigrants.” Industrial and Labor Relations Review 47(2):265–84.

10.

Butterfield

Sherri-Ann

. 2004. “‘We’re Just Black’: The Racial and Ethnic Identities of Second-generation West Indians in New York.” Pp. 288–312 in Becoming New Yorkers: Ethnographies of the New Second Generation, edited by Kasinitz

Philip

Mollenkopf

John

Waters

Mary

. New York: Russell Sage.

11.

Davidson

Basil

. 1994. Modern Africa: A Social and Political History. New York: Routledge.

12.

Denton

Nancy

Massey

Douglas

. 1989. “Racial Identity among Caribbean Hispanics: The Effect of Double Minority Status on Residential Segregation.” American Sociological Review 54(5):790–808

13.

Djamba

Yanyi

. 1999. “African Immigrants in the United States: A Socio-demographic Profile in Comparison to Native Blacks.” Journal of Asian and African Studies 34(2):210–15.

14.

Djamba

Yanyi

Kimuna

Sitawa

. 2011. “The Influence of Race on Employment Status and Earnings of African Immigrant Men in the United States.” International Journal of Population Research 2011:305873.

15.

Djamba

Yanyi

Kimuna

Sitawa

. 2012. “The Labor Force Participation and Earnings Gap among African Immigrant Women in the U.S.A.” International Migration and Integration 13(4):481–501.

16.

Dodoo

Francis Ni-Amoo

Takyi

Baffour

. 2002. “Africans in Diaspora: Black-White Earnings Differences among America’s Africans.” Ethnic and Racial Studies 25(6):913–41.

17.

Farley

Reynolds

Alba

Richard

. 2002. “The New Second Generation in the United States.” International Migration Review 36(3):669–701.

18.

Farley

Reynolds

Allen

Waler

. 1987. The Color Line and the Quality of Life in America. New York: Russell Sage.

19.

Feagin

Joe

. 2006. Systemic Racism: A Theory of Oppression. New York: Routledge.

20.

Feliciano

Cynthia

Rumbaut

Ruben

. 2005. “Gendered Paths: Educational and Occupational Expectations and Outcomes among Adult Children of Immigrants.” Ethnic and Racial Studies 28(6):1087–1118.

21.

Glassner

Barry

. 2000. Culture of Fear: Why American are Afraid of the Wrong Things. New York: Basic Books.

22.

Golash-Boza

Tanya

. 2006. “Dropping the Hyphen? Becoming Latino(a)-American through Racialized Assimilation.” Social Forces 85(1):27–55.

23.

Golash-Boza

Tanya

Darity

William

. 2008. “Latino Racial Choices: The Effects of Skin Colour and Discrimination on Latinos’ and Latinas’ Racial Self-identifications.” Ethnic and Racial Studies 31(5):899–934.

24.

Gordon

April

. 1998. “The New African Diaspora: African Immigration to the United States.” Journal of Third World Studies 15(1):79–103.

25.

Haller

William

Portes

Alejandro

Lynch

Scott

. 2011. “Dreams Fulfilled, Dreams Shattered: Determinants of Segmented Assimilation in the Second Generation.” Social Forces 89(3):733–62.

26.

Halter

Marilyn

. 2007. “Africa: West.” Pp. 283–94 in The New Americans: A Guide to Immigration Since 1965, edited by Waters

Ueda

Cambridge, MA: Harvard University Press.

27.

Hein

Jeremy

. 1993. “Refugees, Immigrants, and the State.” Annual Review of Sociology 19:43–59.

28.

Hirschman

Charles

. 1994. “Problems and Prospects of Studying Immigrant Adaptation from the 1990 Population Census: From Generational Comparisons to the Process of ‘Becoming American.’” International Migration Review 28(4):690–713.

29.

Hondagneu-Sotelo

Pierrette

, ed. 2003. Gender and U.S. Immigration: Contemporary Trends. Los Angeles: University of California Press.

30.

Iceland

John

. 2006. Poverty in America: A Handbook. Los Angeles: University of California Press.

31.

Iceland

John

. 2009. Where We Live Now: Immigration and Race in the United States. Berkeley, CA: University of California Press.

32.

Iceland

John

Scopolliti

Melissa

. 2008. “Immigrant Residential Segregation in U.S. Metropolitan Areas, 1990–2000.” Demography 45(1):79–94.

33.

Kasinitz

Philip

Mollenkopf

John

Waters

Mary

Hollaway

Jennifer

. 2008. Inheriting the City: The Children of Immigrants Come of Age. New York: Russell Sage.

34.

King

Desmond

. 2000. Making Americans: Immigration, Race, and the Origins of the Diverse Democracy. Cambridge, MA: Harvard University Press.

35.

Kusow

Abdi

. 2007. “Africa: East.” Pp. 295–306 in The New Americans: A Guide to Immigration Since 1965, edited by Waters

Ueda

Cambridge, MA: Harvard University Press.

36.

Long

J. Scott

. 1997. Regression Models for Categorical and Limited Dependent Variables. Thousand Oaks, CA: Sage.

37.

Lopez

Nancy

. 2002. Hopeful Girls, Troubled Boys: Race and Gender Disparity in Urban Education. New York: Routledge.

38.

Marrow

Helen

. 2007. “Africa: South Africa and Zimbabwe” Pp. 307–18 in The New Americans: A Guide to Immigration Since 1965, edited by Waters

Ueda

Cambridge, MA: Harvard University Press.

39.

Massey

Douglas

. 2009. “Racial Formation in Theory and Practice: The Case of Mexicans in the United States.” Race and Social Problems 1(1):12–26.

40.

Massey

Douglas

Denton

Nancy

. 1993. American Apartheid: Segregation and the Making of the Underclass. Cambridge, MA: Harvard University Press.

41.

Mittelberg

David

Waters

Mary

. 1992. “The Process of Ethnogenesis among Haitian and Israeli Immigrants in the United States.” Ethnic and Racial Studies 15(3):412–35.

42.

Moss

Tilly

Charles

. 2001. Stories Employers Tell: Race, Skill, and Hiring in America. New York: Russell Sage.

43.

Nawyn

Stephanie

. 2012. “Racial Differences in the Tempo of Assimilation for White and Black African-born Men in the United States.” Ìrìnkèrindò: A Journal of African Migration 6(1):69–90.

44.

Neidert

Lisa

Farley

Reynolds

. 1985. “Assimilation in the United States: An Analysis of Ethnic and Generation Differences in Status and Achievement.” American Sociological Review 50(6):840–50.

45.

Ojito

Mirta

. 2001. “Best of Friends, Worlds Apart.” Pp. 23–40 in How Race Is Lived in America: Pulling Together, Pulling Apart, by Correspondents of the New York Times. New York: Times Books.

46.

Pager

Devah

. 2007. Marked: Race, Crime, and Finding Work in an Era of Mass Incarceration. Chicago: University of Chicago Press.

47.

Pager

Devah

Shephard

Hana

. 2008. “The Sociology of Discrimination: Racial Discrimination in Employment, Housing, Credit, and Consumer Markets.” Annual Review of Sociology 34:181–209.

48.

Portes

Alejandro

Zhou

Min

. 1993. “The New Second Generation: Segmented Assimilation and Its Variants.” The Annals of the American Academy of Political and Social Science 530:74–96.

49.

Portes

Alejandro

Rumbaut

Ruben

. 2001. Legacies: The Story of the Immigrant Second Generation. New York: Russell Sage.

50.

Reimers

David

. 2005. Other Immigrants: The Global Origins of the American People. New York: New York University Press.

51.

Rodney

Walter

. 1981. How Europe Underdeveloped Africa. Rev. ed. Washington, DC: Howard University Press.

52.

Ruggles

Steven

Alexander

J. Trent

Genadek

Katie

Goeken

Ronald

Schroeder

Matthew B.

Sobek

Matthew

. 2014. “Integrated Public Use Microdata Series: Version 5.0” [Machine-readable database]. Minneapolis: University of Minnesota.

53.

Telles

Edward

Ortiz

Vilma

. 2008. Generations of Exclusion: Mexican Americans, Assimilation, and Race. New York: Russell Sage.

54.

U.S. Census Bureau. 2014. “Poverty Thresholds.” Retrieved November 7, 2015 (https://www.census.gov/hhes/www/poverty/data/threshld/).

55.

Waldinger

Roger

Lichter

Michael

. 2003. How the Other Half Works: Immigration and the Social Organization of Labor. Los Angeles: University of California Press.

56.

Wallerstein

Immanuel

. 1986. Africa and the Modern World. Trenton, NJ: Africa World Press.

57.

Waters

Mary

. 1999. Black Identities: West Indians Immigrant Dreams and Realities. Cambridge, MA: Harvard University Press.

58.

Wilson

William J.

2009. More Than Just Race: Being Black and Poor in the Inner City. New York: W. W. Norton.

Poverty and Affluence across the First Two Generations of Voluntary Migration from Africa to the United States,1990–2012

Abstract

Keywords

Background: Voluntary Migration from Africa to the United States

Literature Review: Race, Origins, and Immigrant Adaptation

Data and Methods

National Origins, Race, and Nativity Measures

Poverty and Affluence Measures

Analytical Strategy

Results

A Bivariate Look at Affluence and Poverty across African Immigrant Generations

Race, National Origins, and Affluence among African Immigrants

Race, National Origins, and Poverty among African Immigrants

Discussion and Conclusions

Footnotes

Appendix

Acknowledgements

Notes

Author Biography

References