5

Studying Mobility by Race, Ethnicity, and Immigration Status

This session featured presentations and discussion about studying the influence of race, ethnicity, and immigration status on mobility. The first presentation provided a broad perspective on the challenges and opportunities for studying mobility by race. The second presentation examined how researchers can measure concepts such as structural racism and institutional discrimination, and identified research challenges and opportunities in this area. The third presentation considered immigration status and mobility, while the fourth focused on the diversity of the Asian American population.

STUDYING MOBILITY BY RACE AND CLASS

Mobility can be upward or downward, said William (“Sandy”) Darity (Duke University). There is a general consensus that good societies have a high degree of upward mobility. However, he said, an alternative view is that a good society is one in which a floor is set for well-being; that is, there is a point below which no one can fall. In the current U.S. system, there is no boundary to downward mobility; however, there is some degree of rigidity in upward mobility. The rigidity of a system, said Darity, is predicated on the boundaries on upward mobility, which in turn rely on the extent of structural stratification in the economic system. In order to understand this system of mobility, it is critical to understand and be able to measure the conditions that determine the degree of rigidity, for example, structural racism. There are efforts underway to measure structural racism, he said, but there are some serious conceptual issues. One of the central difficulties is “bouncing between” attempting to measure an outcome of structural racism and a component of structural racism.

Studying social mobility by race and class, said Darity, can mean looking at how social groups fare over the life course, or what happens intergenerationally for these groups. The term social class, he continued, is nebulous and can be defined a number of ways. It is conventionally defined by sociologists with a combination of occupation and income to create groups of bottom, working, middle, and upper classes. Alternatively, social class can be defined by educational attainment and asset ownership. Darity commented that he prefers a definition based solely on occupation, in which the “working class” are those engaged in productive labor, that is, who are neither business owners nor hired managers. Similarly, mobility can be measured in different ways, including by wealth or income. Darity has become “increasingly convinced” that wealth is a more compelling standard to assess individual or group opportunity and economic security.

Based on the definitions of social class and the use of wealth as a measurement of mobility, Darity shared his recent research on recovery after the 2007-2009 Great Recession.¹ In this research, Darity and his colleague examined intragroup mobility patterns over the course of the recession and its immediate aftermath. Darity shared three key findings

___________________

¹ Addo, F.R., and Darity, W.A. 2021. Disparate recoveries: Wealth, race, and the working class after the Great Recession. The ANNALS of the American Academy of Political and Social Science, 695(1), 173-192.

with workshop participants. First, during the recession, Black and Latino² households lost 48 percent and 44 percent of their wealth, respectively, while White households lost 26 percent. Second, whether professional or working class, Black- and Latino-led households were less likely than White households to reach the three upper wealth quintiles (“middle class” status or above). Third, the proportion of wealth-poor families generally decreased between 2010 and 2019, but it increased among Black professionals.

What is particularly noteworthy, said Darity, is the fact that the White working class has a net worth that significantly exceeds the net worth of professional class families who are Black or Latinx. He shared a graph that illustrates this “striking” disparity (see Figure 5-1). It is often asked, he said, why White working class people do not join with oppressed minorities in order to transform American society. This graph, he said, provides “very strong evidence as to why that doesn’t typically occur”—the White working class is in a dramatically advantaged position relative to all other class groups of other races.

___________________

² Usage of the terms Latino, Latinx, and Hispanic in the workshop proceedings reflects use by individual speakers.

MEASURES OF STRUCTURAL RACISM AND INSTITUTIONAL DISCRIMINATION

When it comes to mobility, said Tyson Brown (Duke University), place matters, and place is closely intertwined with race, racism, and opportunity structures. However, he said, understanding of these processes has been hindered by both conceptual and methodological gaps in the literature. Brown shared a figure (see Figure 5-2) that shows the evolution of the field of research on social mobility and race. Descriptive studies yield important information about racial differences in social mobility patterns, he said, and the field has largely moved beyond discredited biological and cultural deficit explanations. Research has also demonstrated the limited utility of behavioral economic and human capital explanations for racial inequities. Some researchers suggest that residual disparities that exist after accounting for behaviors are attributed to discrimination. However, these approaches operate under the “untenable assumption” that racism does not shape economic decisions and the acquisition of human capital in the first place. With this evolution in understanding, it is becoming clear that focusing solely on individual factors leads to an incomplete and biased understanding of the drivers of racial inequities and social mobility.

Researchers have increasingly moved toward a conceptualization of racial inequities in social mobility as a consequence of structural racism, said Brown. However, relatively few studies have empirically tested this proposition; skeptics point to the dearth of quantitative research on the topic, and dismiss structural racism as a slippery concept for which robust empirical evidence is lacking. As a field, said Brown, there is an opportunity to gain empirical traction by measuring structural racism and its effects on social mobility. This will enable researchers to answer new and salient questions, to use more rigorous research designs, and to build a knowledge base to inform more efficacious racial equity solutions.

In order to move forward toward this potential, Brown offered several research priorities for the field. First, researchers should utilize core tenets of structural theories to guide measurement approaches. Second, there is a need to develop novel, theory-informed, multi-sectoral measures of structural racism. Third, mapping structural racism can build a better understanding of the geographic variation in racial exclusion and subordination. Fourth, the impact of structural racism on mobility patterns should be estimated by linking contextual structural racism measures with geocoded data on mobility. Finally, he said, building a publicly available data infrastructure on structural racism could lower barriers to this type of research and catalyze research on differential mobility processes. Brown focused the remainder of his presentation on the first three research priorities.

Utilizing Core Tenets

There are a number of challenges to measuring structural racism, said Brown, because it is a “complex and insidious de facto phenomena” that is often hidden and not directly observed in modern society. It is therefore unsurprising, he said, that there are relatively few empirical studies that explicitly measure the impact of structural racism on social mobility. However, there are several theoretical frameworks that offer foundational grounding for understanding how dimensions of structural racism intersect and work together as a system. Brown shared a study in which he and his colleague scoped the broader literature in order to identify central tenets of structural racism theories.³ They distilled these foundational ideas into a definition that provides conceptual and analytical clarity: “structural racism involves a multifaceted, interconnected, and institutionalized system of racial insubordination for people of color, and superordination for Whites, and this is observable in manifest concrete racial inequalities in life out-

___________________

³ Brown, T., and Homan, P. 2022, March 21. Structural racism and health stratification in the U.S.: Connecting theory to measurement. SocArcXiv. https://doi.org/10.31235/osf.io/3eacp

comes.” In short, said Brown, structural racism refers to the systemic, racial exclusion from power, resources, opportunities, and well-being.

Structural racism is embedded in political, economic, medical, criminal, legal, and social institutions, and is observable as racial inequities in outcomes in education, occupation, wealth, health, political representation, incarceration, and housing. Structural racism can be considered a fundamental factor that drives racialized disadvantages, because it shapes access to both immediate economic resources and opportunities, as well as exposure to social and economic risk. Because of its multifaceted, interconnected, and institutionalized nature, it is likely that structural racism impacts mobility through many different intervening mechanisms, he said.

Developing Measures

Empirical studies on structural racism are often out of step with theoretical insights, said Brown, and they have a number of limitations. For example, studies have typically relied on single indicators of structural racism, which are useful but often subject to measurement error. Alternatively, they have examined several indicators, but done so separately, thereby overlooking relationships among structural forms of racism and how they operate as a system to shape life outcomes. These approaches, said Brown, fail to capture the extent to which indicators of structural racism are interconnected and reflect an underlying latent construct, and they also lead to an incomplete understanding and biased estimates of the impact of structural racism on mobility. Furthermore, he said, much of the research on structural racism measurement has been focused at the meso-level, in particular the neighborhood and county levels. While these levels are important, the literature has largely ignored or overlooked more macro-level units, such as states. However, a few recent studies use administrative and publicly available data to examine state-level structural racism; these studies span domains including judicial, educational, economic, political, and segregation. Investigating structural racism at the state level is a “major advance,” said Brown, yet few studies have examined the joint consequences of multi-sectoral structural racism. Brown shared details of his recent work that aims to measure structural racism in ways that align with structural theories. Brown and his colleague used nine indicators of structural racism to develop a latent measure of structural racism, using confirmatory factor analysis to measure the extent to which structural racism across domains is reflective of an underlying latent construct.⁴ Brown conveyed his hope that these sorts of approaches to capturing and measuring structural racism will provide a proof of concept to “get empirical traction on the drivers of racial inequalities.”

___________________

⁴ Ibid.

Mapping Structural Racism

Theory and evidence both suggest that the manifestations, as well as the modalities, of structural racism vary spatially. For example, Figure 5-3 maps structural racism across states, using the structural racism latent scale Brown and his colleague developed. The patterns in this map are consistent with the idea that states, which operate as political, social, cultural, legal, and administrative units, vary in their degree of structural racism and function as institutional actors contributing to the unequal distribution of resources. Brown noted that recent events, including the pandemic, have “laid bare” the role of states in shaping inequality in general and racial inequality in particular.

To test the utility of the latent measure of structural racism, Brown and his colleague linked it to geocoded health and demographic survey data. The results revealed that state-level structural racism is predictive of an array of health outcomes; specifically, exposure to higher levels of structural racism at the state level is associated with worse health outcomes for Black people, but not White people.⁵ There is a growing body of research on developing novel measures of structural racism in the population health literature, said Brown. For example, a recent study⁶ conducted a latent class analysis on structural racism at the county level, spanning several domains, including segregation, home ownership, education, employment, and in-

___________________

⁵ Ibid.

⁶ Hardeman, R.R., Homan, P.A., Chantarat, T., Davis, B.A., and Brown, T.H. 2022. Improving the measurement of structural racism to achieve antiracist health policy. Health Affairs (Millwood), 41(2),179-186.

come; the study found that structural racism in these areas is predictive of health outcomes. Brown argued that this line of inquiry could be expanded by empirically testing the extent to which there are distinct typologies of structural racism, at multiple levels, and how they affect racialized mobility processes. This would be an “innovative direction for the field.”

It is also important, he said, to directly measure how social mobility is affected by policy contexts. A 2021 project⁷ developed a publicly available database of laws related to structural racism. Another project looked at how state-level immigration policies may impact health outcomes for Latinx people.⁸ One question that can be examined by mapping existing data sources is the extent to which social mobility is impacted by cultural and ideological forms of anti-Blackness. While some measures such as racial animus and racial biases are established, there is an opportunity to utilize innovative data and methods to capture new aspects of anti-Blackness. For example, geocoded data from internet search engines can be used to capture racial animus in a state’s population.

Mapping historical racism can help illuminate how history directs, constructs, and molds contemporary structural racism. In that vein, said Brown, several empirical studies⁹ have found that regions that had larger enslaved proportions of population in 1860 have greater present-day inequalities in poverty and economic mobility, as well as higher levels of pro-White bias. Other findings of these studies include that historical redlining practices underlie contemporary residential segregation patterns, and that New Deal policies expanded the White middle class and are “directly implicated” in modern racial inequalities in wealth. Further research, said Brown, should explicitly test the extent to which contemporary social mobility is affected by historical racism—for example, how variations in exposure to slavery, Jim Crow laws, or racialized voter suppression impact mobility.

___________________

⁷ Agénor, M., Perkins, C., Stamoulis, C., Hall, R.D., Samnaliev, M., Berland, S., and Bryn, A.S. 2021. Developing a database of structural racism-related state laws for health equity research and practice in the United States. Public Health Reports, 136(4),428-440.

⁸ Philbin, M.M., Flake, M., Hatzenbuehler, M.L., and Hirsch, J.S. 2018. State-level immigration and immigrant-focused policies as drivers of Latino health disparities in the United States. Social Science and Medicine, 199, 29-38.

⁹ Bloome, D., and Muller, C. 2015. Tenancy and African American marriage in the postbellum South. Demography, 52(5), 1409-1430; Darity, W., and Mullen, A.K. 2020. From Here to Equality: Reparations for Black Americans in the Twenty-First Century. Chapel Hill: University of North Carolina Press. Williams, J.A., Logan, T.D., and Hardy, B.L. 2021. The persistence of historical racial violence and political suppression: Implications for contemporary regional inequality. The ANNALS of the American Academy of Political and Social Science, 694(1), 92-107. Muller, C. and Wildeman, C. 2016. Geographic variation in the cumulative risk of imprisonment and parental imprisonment in the United States. Demography, 53(5), 1499-1509.

Brown shared an example of a research project¹⁰ that used a structural intersectionality approach to population health; researchers examined the relationships between macro-level structural racism, structural sexism, and economic inequality, and linked these data to state-level measures of individual health. Their analysis, he said, showed that structural forms of justice intersect in a variety of ways, but do not strongly covary across states. Furthermore, they found that the joint effects of structural oppression were most deleterious to Black women, followed by Black men and White women. White men’s health, said Brown, was “largely unaffected” by intersecting forms of structural oppression. This study could serve as a springboard and potential data source for similar studies aimed at understanding how intersectional oppressions shape mobility.

STUDYING MOBILITY BY IMMIGRATION STATUS

Studying mobility among immigrant populations continues to have relevance, said Tomás Jiménez (Stanford University), as one-quarter of today’s population is either a first- or second-generation immigrant. The framework for this research goes back to the origins of the study of immigrant groups in the United States, he said, introducing workshop participants to the definition of the word generation as it is used by immigration scholars. First generation refers to immigrants themselves; their children are referred to as second generation and grandchildren third. This conceptualization of generation is based on assimilation theories that sought to explain the experiences of late-19th- and early-20th-century southern and eastern European immigrants. This theory holds that there is change between immigrant generations across time; each new generation born in the United States tends to have more education and more income, be less residentially segregated and more likely to intermarry, and their ethnic identity becomes less important over time.

Jiménez described how immigration patterns have changed over the years. First, starting in 1965, immigrants largely come from non-European origins; most are from Latin America with a sizable proportion from Asia (see Figure 5-4). Second, there is a large undocumented population among today’s immigrants. The size of the undocumented population peaked shortly before the 2007-2009 Great Recession, declined precipitously and has remained relatively flat. One misconception about the population of undocumented immigrants, he said, is that they are newcomers to the country. In reality, about two-thirds of undocumented immigrants have lived in the

___________________

¹⁰ Homan, P., Brown, T.H., and King, B. 2021. Structural intersectionality as a new direction for health disparities research. Journal of Health and Social Behavior, 62(3), 350-370. https://doi.org/10.1177/00221465211032947

United States for more than 10 years. This fact has significant implications for understanding mobility, said Jiménez. The third important characteristic about today’s immigrant population is that immigrants are distributed across the country “in ways that we have never seen before.” There are still significant concentrations of immigrants in California, the eastern corridor, the southern part of Florida, and the U.S.-Mexico border, but there are now new areas of concentration in the Midwest and the South.

In spite of these changes, the approach for studying mobility among immigrant populations still involves a close examination of the differences between immigrant generations, said Jiménez. The typical way of doing this is comparing all first-generation immigrants to all second-generation immigrants and so on. Jiménez, in contrast, advocates for a model that takes into account not just the immigrant generations, but also the time period that these immigrant generations come from. He shared a figure from a 2008 book¹¹ that illustrates this model (see Figure 5-5).

There are several challenges involved in studying immigrants and mobility, said Jiménez. There is a lack of large government or non-government surveys that track immigrant generations. The Current Population Survey is an exception because it asks whether people and their parents were

___________________

¹¹ Telles, E., and Ortiz, V. 2008. Generations of Exclusion: Mexican Americans, Assimilation, and Race. New York: Russell Sage Foundation.

born in the United States; however, it still remains difficult to track the third generation and higher. For example, if a respondent said that they and their parents were born in the United States, they could be post-1965 third-generation immigrants, or they could be fifth-generation descendants of immigrants who arrived more than 100 years ago. This difference is very important in a society “where we have experienced tremendous demographic, social, political, and economic change because of the recent arrival of immigrants.” In the 1990s, social scientists made particular effort to understand the children of immigrants, said Jiménez, but there has been no similar effort to understand the grandchildren of immigrants.

There are also intellectual challenges in immigration studies, he said, such as the common distinction that is made between race and immigration scholarship. Jiménez argued that this is a false distinction, and that “one cannot begin to understand race in the United States today and mobility without understanding differences in immigrant generations, and one cannot begin to understand differences in immigrant generations without understanding race.” Another challenge is an increasing tendency to homogenize ethnoracial groups and to treat them as if they are “hermetically sealed units” that do not change internally. Instead, he said, these groups should be seen as dynamic and internally differentiated, and research should focus not just on mean outcomes of an ethnoracial group but on the distribution of differences within the group as well. The third challenge in this area, he said, is that the ideas of assimilation, integration, and incorporation are “unfashionable”; he argued, however, that they remain essential to the study of immigrants in the United States.

Studying immigration and mobility is more important than ever, said Jiménez, because the “new third generation” has arrived. These grandchil-

dren of post-1965 immigrants can give great insight into the status of immigrants and how they are doing compared to their parents and grandparents. Jiménez shared details from his work in this area¹²; the study compared second generation immigrants from 1980 to third-generation immigrants from 2010. Jiménez and his colleagues looked at measures such as family structure, educational attainment, household income, and poverty status. This study was in part a “proof of concept” project to show the importance of studying this new third generation because they are the generation in which there is often a huge shift in mobility outcomes. Jiménez closed by advocating for a concerted effort to track down the actual second-generation immigrants who were studied in the 1990s and their third-generation children; this would allow researchers to track changes between individuals and their parents and grandparents without constructing synthetic cohorts.

MOBILITY AMONG ASIAN AMERICANS

Asian Americans are one of the fastest-growing populations in the United States, said Jennifer Lee (Columbia University); the percentage of Asian Americans in the United States nearly doubled between 2000 and 2020, and is expected to double again by 2060. Unlike some other ethnoracial groups, the population of Asian Americans is growing primarily through immigration, and by 2055, Asian Americans will surpass Hispanic Americans as the largest immigrant group in the country. Two out of three Asian Americans are first-generation immigrants and 90 percent are either immigrants or the children of immigrants. Moreover, said Lee, one in seven Asian immigrants is undocumented, and this population is growing at a faster pace than the population of undocumented immigrants from Mexico or Central America. Asian Americans are a diverse group, coming from countries including China, Korea, the Philippines, India, Pakistan, and Cambodia. Lee noted that this heterogeneity is often not acknowledged in the United States, which leads to biased assumptions about Asian Americans’ opportunities and outcomes. At one end of the distribution are highly educated, hyper-selected Asian immigrants from areas such as Taiwan and India, and at the other end are Southeast Asian refugees from countries such as Bhutan and Laos.

This diversity, said Lee, is often obscured by means and medians; she emphasized the importance of disaggregating data in order to understand, serve, and protect the rights of the Asian American population. For example, a detailed count of the Asian population is essential to providing

___________________

¹² Jiménez, T.R., Park, J., and Pedroza, J. 2018. The new third generation: Post-1965 immigration and the next chapter in the long story of assimilation. International Migration Review, 52(4), 1040-1079.

voting materials in appropriate languages. Furthermore, she added, Asian Americans are the group most likely to be concerned that answers they provide on the Census will be used against them. This institutional distrust also affects Asian Americans’ willingness to report hate crimes and violence, which surged during the COVID-19 pandemic. “When we consider opportunities and challenges related to inclusion and mobility among Asian Americans,” said Lee, “building institutional trust must be part of the conversation.” Unfortunately, said Lee, there is a “glaring lack” of investment in Asian American communities, from both the federal government and private foundations. For example, the National Institutes of Health invested only 0.17 percent of its budget between 1992 and 2018 in Asian American communities,¹³ and foundations awarded only 0.20 percent in 2018 to this community.¹⁴

Lee closed with several recommendations for improving research, and ultimately outcomes, for the Asian American population. First, collection categories should be expanded for Asian subpopulations, Native Hawaiian and Pacific Islander subpopulations, and gender identity. Second, researchers should adopt the American Community Survey practice of collecting information on country of birth of both the respondent and the parent. Lee’s third recommendation was to seek data equity in all stages of research design, analysis, and dissemination. For example, data collection instruments should be linguistically and culturally appropriate in order to serve a population in which two-thirds are immigrants and one-third is limited in English language proficiency. Fourth, Lee emphasized the importance of building institutional trust among Asian Americans; this involves including community leaders and scientific experts at all stages of data collection, as well as creating scientific advisory committees and community advisory committees. Finally, Lee called for more investment in the Asian American population. She noted that after the surge in anti-Asian violence during the COVID-19 pandemic, there has been “unprecedented interest and surge in investment.” This is an opportunity to “break from the past” and to include Asian Americans in research and investment.

___________________

¹³ Đoàn, L.N., Takata, Y., Sakuma, K., and Irvin, V.L. 2019. Trends in clinical research including Asian American, Native Hawaiian, and Pacific Islander participants funded by the US National Institutes of Health, 1992 to 2018. JAMA Network Open, 2(7), e197432. https://doi.org/10.1001/jamanetworkopen.2019.7432

¹⁴ Asian Americans/Pacific Islanders in Philanthropy (AAPIP). 2021. Seeking to Soar: Foundation Funding for Asian American and Pacific Islander Communities. https://aapip.org/resources/seeking-to-soar-foundation-funding-for-asian-american-and-pacific-islander-communities

DISCUSSION

Following the presentations, Trevon Logan (The Ohio State University) and Kathleen Mullan Harris (University of North Carolina at Chapel Hill) moderated a discussion with speakers and workshop participants.

The Concept of Race

Logan began the discussion session by noting that while presenters talked at length about definitions of structural racism and racial inequities, there was no discussion of “what race is in and of itself.” Logan argued that in this context, race should be defined as a political variable, because it influences the distribution of resources and has an inherent dimension of power. He asked speakers to comment on whether using a political definition of race would alter the study of mobility, and whether it could open up new avenues to think about solutions to inequities. Darity responded that he personally views race as a political construct because it is an instrument “for the purposes of giving one social group an advantage over others, and to rationalize that advantage.” However, Darity said he was not sure that thinking of race as political rather than a social variable would change research or analysis; researchers generally rely on self-reported race, which is a construct that emerges out of social experiences. Brown agreed with Logan that race is a historical and political construct, rather than merely a demographic characteristic. Race was made through a political process, and it is dynamic, fluid, and “highly contested.” Brown said that this does have implications for research, and that there is an opportunity to rethink how to collect and interpret self-report measures of race by taking a more contextual approach. Darity added that this historical and political approach to race can be examined through the lens of stratification economics, which looks at how relative group position plays a role in the behavior of individuals and social groups. Brown agreed and said new data and tools are available for mapping aspects of historical oppression and violence in order to better understand patterns and drivers of inequality and mobility.

Mobility in Immigrant Populations

Harris asked Lee and Jiménez to elaborate on what past mobility research in immigrant populations has shown, and where the research should go moving forward. Jiménez responded that the patterns of mobility for immigrants today look similar to the patterns of 100 years ago, and for some measures, upward mobility is happening faster. For example, he said, intermarriage rates among second-generation immigrants are high, with about 30 percent of third-generation Hispanic American children growing

up with one Hispanic parent and one non-Hispanic parent. Intermarriage is a “key measure of integration,” said Jiménez. When looking at the research, said Lee, it is important to keep in mind the difference between mobility and outcomes. For example, second-generation Chinese Americans have high educational attainment, but the rate does not vary significantly from their first-generation parents. On the other hand, second-generation Mexican Americans have much higher educational attainment than their parents, being two to three times as likely to graduate from college. When considering “success” among immigrant populations, the conversation is often focused on outcomes rather than mobility, she said. Another interesting pattern, said Lee, is that women in the immigrant community are more likely to graduate from college than men, particularly in the more disadvantaged groups.

Jiménez added that an undocumented status can have a large effect on mobility, even for future generations who are born in the United States. Research has found that for immigrants who were able to legalize, their children’s and grandchildren’s outcomes were “far and away” better than the outcomes of descendants of immigrants who could not legalize. Undocumented status is a “penalty” to the classic mobility outcomes, he said.

Structural Measures in Immigration

Noting that Darity and Brown discussed the importance of structural factors in studying race and mobility, Harris asked Lee and Jiménez what types of structural measures are important in the immigration space. Jiménez said that immigration scholars have been calling for large government surveys to ask for parent place of birth, and ideally grandparent place of birth as well. He acknowledged that if surveys included all of the questions that social scientists wanted, they would be inappropriately lengthy, but emphasized the importance of capturing this information in order to understand the generational shifts that are occurring. If large government surveys do not do this work, he said, there needs to be investment in nongovernmental efforts. Lee told workshop participants about her work with the STAATUS (Social Tracking of Asian Americans in the United States) index, and its questions about perceptions of Asian Americans. For example, the survey asks questions about whether respondents think that Asian Americans are “at least partly responsible” for COVID-19. Surveys that focus on racial attitudes, said Lee, often ignore attitudes toward Asian Americans, and Asian Americans are often not included in conversations about race or immigration. It is imperative, said Lee, that future research looks at the experiences of this population as well as the attitudes toward them.

Mobility Over the Life Course

Previous speakers in the workshop discussed how measures of mobility—for example, income or wealth—often change significantly over the life course, said Harris. She asked speakers to comment on how the life course approach applies in the context of studying mobility and race and immigration. Darity responded that the wealth differential by race widens sharply as people age; it is at its smallest when people are in their 20s and 30s. If policies designed to close the wealth gap are based on data collected early in life, they can be “quite misleading.” Darity suggested that measuring wealth at age 60 would be most appropriate, because it reflects the cumulative consequences of their life experiences. Furthermore, he said, the best predictor of an individual’s eventual wealth position is the wealth position of their parents and grandparents. An individual’s net worth is not generally a consequence of “careful and deliberate acts of personal savings,” but instead is impacted by the advantages they receive from the previous generations. A final point, said Darity, is that educational attainment has a very weak relationship to reducing wealth disparities by race. For example, Black heads of household with a college degree have a lower median net worth than White heads of household who never finished high school. When it comes to wealth accumulation, he said, it is the intergenerational transmission mechanisms that are decisive. Brown added that this point is important when considering approaches for reducing racial inequality. Measuring wealth later in life is useful for capturing cumulative consequences, he said, but it is unclear where interventions would be most helpful. For example, should “baby bonds” be distributed when people reach the ages of 18 or 20, or would interventions at age 30 or 40 be more impactful? To answer these types of questions, said Brown, researchers need longitudinal data to capture all of the factors that matter to outcomes, and when they occur. He asked: are there “sensitive periods” that make a big difference, and when are these periods?. The next iteration of research in this area, he said, needs to go beyond the static snapshot to follow people in place and time, both within and between generations.

Darity gave his opinion on what policy would be appropriate for addressing racial wealth differences. He said that reparations for Black American descendants of slavery is the “only way to do it.” The “baby bonds” proposal is directed at all Americans, so it would not have a powerful effect on the racial wealth gap.

When considering the life course perspective for immigrant populations, said Lee, there are two major factors that impact mobility. First, limited English language proficiency is a critical issue for some immigrant populations, and directing resources to this area is necessary to help them have the resources to be “a part of our society.” Second, a pathway to

citizenship is critical to inclusion, both for the first generation and for subsequent generations. Having a parent who was not able to naturalize significantly affects mobility outcomes for the second and third generations. Jiménez added that the understanding of the second generation is “stuck in late adolescence.” He further stated that investing in research to understand where these individuals are now, and when they are in their 30s, 40s, and 50s, is “incredibly important.”

Reparations

Following up on Darity’s comment that reparations for Black American descendants of slavery are the only way to address racial wealth differences, a workshop participant asked how eligibility would be determined, given that many Americans have mixed ancestry or are descended from more recent immigrants. Darity said that he has long recommended two criteria for eligibility. First is a “lineage standard,” in which an individual would have to demonstrate that they have at least one ancestor who was enslaved in the United States. The second criteria would require individuals to show that they self-identified as Black, Negro, African American, or Afro-American on a legal document at least 12 years before the adoption of a reparations plan. These criteria, he said, would eliminate the issue of White Americans who have an enslaved ancestor being eligible for reparations. To facilitate this process, Darity recommended that the federal government provide genealogical services to individuals seeking to make a claim.

Geography of Structural Racism

A workshop participant asked Brown to elaborate on the map of structural racism that he shared (see Figure 5-3), noting that some of the results were counter to what he expected to see, for example, that structural racism was higher in the Northeast than in the South. Brown responded that the structural racism measure includes factors that are particularly acute in Northeast and Midwestern states, such as inequalities in education, housing, political domains, employment, and criminal justice. While the reasons for these inequalities are not completely understood, scholars posit that the contemporary structural racism in the North has its roots in institutionalized policies and practices that were created in response to the Great Migration. These policies—such as redlining, racial covenants, and discriminatory policing—were codified because northern White populations perceived the incoming Black populations as a threat. There is a growing body of evidence, said Brown, that shows that Black/White inequalities are greatest in the Northeast and Midwest. However, he cautioned, this does not mean that individuals in these areas are doing worse than individuals

in other areas; the South is “unique and pathogenic in all sorts of ways that are problematic.” The greatest overall disadvantage is often found in the South, particularly in Appalachia. It is important, he said, to research and understand both absolute well-being and inequalities between groups.

Measures and Outcomes of Structural Racism

Harris asked Darity and Brown to comment on the need to distinguish between structural racism itself and its outcomes, which are often used as measures of structural racism. Specifically, she asked how they theorize about the causal mechanisms that link mesolevel indices with outcomes of interest. Brown responded that he thinks of the multifaceted scale of structural racism as reflecting a hidden or latent complex system. In this system, there are an array of pathways through which structural racism impacts outcomes, including health, economic capital, autonomy, power, and risk. For example, social stressors such as toxic living conditions and stigma are also drivers of inequality, he said. Brown argued for an approach in which structural racism is not measured by single indicators, but instead is considered as a system. This approach avoids certain types of measurement error and bias, he said, and also more realistically reflects the phenomena observed. Research in this area is still in its infancy, and there are opportunities to further explore specific causal pathways as well as the indirect effects of structural racism.