Abstract
This paper describes Latinx stratification economics (LSE) as a scholarly approach to studying the economic status of Latinas/os/es/xs primarily in the United States. We coin the term LSE to refer to work that draws on and is in conversation with both the emergent, interdisciplinary subfield of stratification economics (SE) and the interdisciplinary field of Latinx studies (LS). SE and LS have distinct intellectual traditions and drawing on both leads to strong theoretical and empirical scholarship on Latinxs, on the operation of race across space and historical time, and on the intersection of race with other systems of domination. We discuss how, based on these perspectives, it is misguided to expect racial/ethnic categories like Hispanic to be consistent over time and space and to correspond reliably with phenotypical characteristics or culture. We argue that a good faith reading of the LS literature would result in the recommendation to subordinate models of migration to models of colonialism and imperialism. We discuss the significance of normative goals and social justice to complement “gap analysis” comparisons to non-Hispanic whites. Lastly, we discuss deficiencies of the dominant models of discrimination and, as an alternative, we highlight rational models of racism that involve strategic identifications with whiteness, blackness, and mestizaje, including by members who identify as Latinx or those with Hispanic ancestry.
The purpose of this paper is to describe Latinx stratification economics (LSE) as a scholarly approach to studying the economic status of Latinas/os/es/xs 1 primarily in the United States. We coin the term LSE to refer to work that draws on and is in conversation with both the emergent, interdisciplinary subfield of stratification economics (SE) and the interdisciplinary field of Latinx studies (LS). In putting forth LSE as a distinct perspective, we do not suggest that SE has in some way failed to account for the Latinx experience. 2 Rather, we argue that much of the work in SE focused on Latinxs is consistent with LS, that drawing on LS has been helpful to stratification economists, and that, going forward, it would be a mistake for scholarship about Latinxs, particularly economists, to systematically fail to draw on the insights of LS. We are making the specific argument from within the economics literature that SE and LS have distinct intellectual traditions and that drawing on both leads to strong theoretical and empirical scholarship on Latinxs, on the operation of race across space and historical time, and on the intersection of race with other systems of domination. 3
Despite the close connection between racist and xenophobic ideology and the material consequences of racialized social, economic, and political systems, economics has generally avoided creating rational models of racism (Agesa & Hamilton, 2004; Darity & Mason, 1998; Lewis, 1985; Stewart, 1976). SE is a growing subfield that addresses this gap by recognizing that individual members of social groups have an interest in the relative position of their group in a social hierarchy (Blumer, 1958; Darity, 2005). SE explicitly draws on the disciplines of social psychology and sociology, and African American economists have played a prominent role in its development (Francis et al., 2022). SE has been recognized and incorporated in academic work outside of economics (e.g., Painter et al., 2016), and it has contributed significantly to several important policy debates (e.g., around baby bonds and reparations). SE argues that there are flaws in how core elements of mainstream, modern economics are applied to the study of race and racialization that make this work poorly suited to explaining observable, persistent disparities between ascriptively marked groups. As described in Darity et al. (2017, p. 38) and illustrated in Darity et al. (2006), the SE perspective is a variant of real conflict theory: The critique is not that mainstream economics is too abstract or too theoretical or too mathematical but that many of the dominant frameworks used are insufficient for developing a nonhegemonic understanding of race and racialization (and conflicts between social groups more broadly).
LS is an interdisciplinary field most commonly housed within the humanities and social sciences. 4 Engaging with LS has the potential to enrich the quality of economics research about Latinxs; notably, it may simply encourage scholars to explicitly value the lives of Latinx people through their scholarship, a shift that should not be taken for granted in the largely race-blind and politically “neutral” discipline of economics. The main themes from LS relevant to economics research on Latinxs are the nature, meaning, and limitations of Latinidad; the role of the nation-state in creating subjectivities and determining the distribution of resources; and the transnational experience of different regimes of racialization. In this paper, we unpack each of these themes in light of relevant threads of economics work on Latinxs. The disciplines of economics and LS contrast sharply in terms of prestige and institutional support, with economics generally ascribed higher status and economists provided with greater resources. Academic hierarchies and incentives within academic disciplines may partially explain the low level of engagement with LS, even by Latinx economists writing about Latinxs.
The intellectual contribution of this paper goes beyond creating a bridge between SE and LS—that is, identifying which themes in LS are most relevant for SE and demonstrating how the SE framework is consistent with much of the literature in LS. We identify specific concepts and debates from LS that conflict with dominant strains in the economics literature on Latinxs. For example, reading mainstream economics research about Latinxs through the lens of LSE reveals the glaring absence of the nation-state as an important site through which groups compete for material privilege and relative position in a social hierarchy. This type of analysis builds on but is distinct from technocratic policy analysis. As both sites of conflict and powerful producers of racialized living conditions, the nation-state and interactions between nation-states become all but indispensable in an authentic knowledge seeking project about Latinxs. Many of the arguments made here coincide with critical interventions that other individuals or intellectual traditions have made vis-à-vis mainstream economics. What is new in our argument is that LS and SE in combination generate specific critiques of economics research that can be used as criterion for economists’ work. 5
In the following section, “Two Related Literatures,” we present brief overviews of both SE and LS, the purpose of which is to establish key ideas, introduce the fields to those who may be new to the literatures, and highlight concepts that are used in the next section. Next, in “Toward a Latinx Stratification Economics,” we explain and illustrate the distinctive perspective of LSE compared with mainstream economics by discussing “Hispanics” as an analytical category and the three major theoretical frameworks that economists use to understand Latinxs: migration, assimilation, and discrimination. Taken together, these sections characterize the LSE perspective, demonstrate new insights that come from its application, and illustrate the LSE approach through discussion of specific examples of economics research.
Two Related Literatures: Stratification Economics and Latinx Studies
Stratification Economics (SE)
SE, a growing subfield in economics, seeks to explain intergroup inequality by analyzing the collective self-interest of members of social groups in maintaining or improving their group's relative position in a social hierarchy. SE provides a theoretical framework that can be used to develop new approaches in economic theory and new empirically testable hypotheses. SE has been applied to understand the origins of racial identity (Darity et al., 2006; Lewis, 1985), caste, skin color (Goldsmith et al., 2007), and intersections of multiple socially salient identities. Two recent reviews of the SE literature can be found in Chelwa et al. (2022) and Darity (2022). 6
In contrast to the dominant alternatives, SE focuses on group identities and the strategic behavior of individuals in groups by drawing on insights from economics, sociology, and social psychology. In mainstream economics, racism is usually modeled as discrimination taking one of three forms. Models based on Becker (1957) cite an irrational “taste for discrimination,” in which some individuals are willing to pay in order to avoid contact with members of another group. Another common approach, “statistical discrimination,” is based on information asymmetry, in which discriminatory behavior arises from the application of perceived or actual group average qualities to individuals from that group (Arrow, 1973; Phelps, 1972). Finally, behavioral economists and others have studied how unconscious bias, often presented as split-second decision-making, can affect economic outcomes (Bertrand et al., 2005). SE presents a fundamental alternative to these models by recognizing the rational, self-interested motivation for discrimination as well as the structural, institutional, and social dimensions of intergroup rivalry; in contrast with identity economics (Akerlof & Kranton, 2000) SE consciously rejects explanations that rely on the cultural dysfunction or genetic inferiority of subordinated groups to explain group stratification.
The basic idea of SE is that individuals belong to ascriptive social groups that confer privileges and penalties based on their relative position in a social hierarchy. Individuals have incentives to improve or maintain their group's relative position in order to reap the rewards of higher status, especially for the most privileged group, but also to avoid the many consequences of the bottom rung or “last place” which extend through time intergenerationally. Any group's collective ability to improve its position is itself contingent on the resources available to it. Because members of the dominant group benefit from their group's relative position, they may choose to take actions that have no specific direct individual benefit (or indeed that may seem individually harmful) but that benefit them in a relative sense by maintaining their group's collective privilege and relative positioning. These actions may be overt, such as extrajudicial killings at the border between Mexico and the United States (Mirandé, 2019; Shadowen, 2018), or they may be covert, with no reference to group membership. Actions that maintain racial hierarchy may take the form of market interactions, support for particular public policies or actions to influence their implementation, or social decisions operating at the level of interpersonal relationships, impersonal exchange, institutional formation and evolution, or cultural hegemony. The characteristic and distinct SE emphasis on material benefits does not negate the fundamental role of social psychology, since the SE focus on relative position means that an individual may prefer to maintain intergroup inequality rather than allow material improvements for a comparison group, even when the latter course would not change (or may even improve) their own individual and group's material conditions.
The key point for the SE framework is that rewards and punishments exist that are contingent on group membership. Not all members of the dominant group receive the same rewards from their group membership. In fact, not all members of the dominant or subordinate group will have the same attachment to their group. This is expected. Life chances related to one social category (race) may be mediated by another category (gender), further contributing to heterogeneity. The SE framework accommodates these complexities not as auxiliary qualities but as fundamental to an understanding of stratification. In order to explain intergroup disparities, SE explores the sociohistorical formation of group identities, how these identities change over time, and the ways in which individuals and groups use these identities strategically in between-group and within-group contexts. From the perspective of economic systems, interesting coordination issues arise; for example, in the United States, it is not necessary for every white person to actively be involved in maintaining white supremacy for there to be positive, material returns to whiteness in a variety of settings (Harris, 1993).
Because SE treats intergroup disparities in terms of relative position, a zero-sum game is generally operative. Gains for disadvantaged groups will necessarily imply relative losses in privilege for the dominant group. Not all members of the dominant group will act in every instance to forestall this loss of relative privilege. For those members of privileged groups who do not have a strong sense of group identity, who will not face penalties for disloyalty for declining to contribute to the group asset (that is, their group's advantageous relative position), and who choose to act on social justice or moral principles, gains for disadvantaged groups and loss of relative privilege may be welcome. Historically, at any one time the number of people who work to eliminate their own unjust material privileges is low and this group is not well represented in positions of power in society. The zero-sum nature of competition for relative position helps to justify and explain the importance of stratification economists’ normative goals for group-based justice and fairness. This stands in contrast with the mainstream of the discipline, reflective of the transformation that occurred in academic economics in the twentieth century whereby moral and ethical concerns were largely purged from the discipline in an effort to achieve the status of a “positive” science. Pareto optimality and welfare analysis do not generally have a way of adjudicating conflicting claims that arise from status goods. To quote Reverend Dr. William Barber II and a mantra from the Poor People's Campaign, “Economic justice is a moral imperative.” The SE approach frees the economist-researcher to explore questions involving the intergroup distribution of life chances, including questions around public policy tools to achieve more equality, independently of efficiency gains, grounded instead in concepts such as human rights and justice (Hamilton, 2020).
Cultural explanations for group inferiority are an integral part of the racial formation process—that is, the historical process by which groups are created and imbued with meaning. The problem with many of these explanations is that they reverse the causality by using culture, sometimes measured using observable behaviors, to explain economic outcomes (Hamilton et al., 2015; Steinberg, 2011). In the Latinx context, there are efforts to explain the perceived economic success of some groups (e.g., Cuban Americans) relative to others (e.g., Mexican Americans) without considering the different socioeconomic and wealth starting positions of each group, nor considering the role of the state, quasi-state, or international organizations in propelling groups onto different material trajectories. Few of these studies incorporate an explicit analysis of skin shade and opportunity structures within Latinx communities (Darity et al., 2005). While SE forms a theoretical core that can be used to understand racial inequality, LS constitutes a rich intellectual tradition that comments specifically on the Latinx experience and has developed theoretical frameworks specific to understanding Latinxs.
Latinx Studies (LS)
While a complete overview of the research and writing in LS is beyond the scope of this paper, the purpose of this section is to establish a working understanding of topics that will help connect our proposals for LSE to LS, especially for those who may be unfamiliar with LS. For this purpose, we also highlight differences between the LS approach and the dominant approaches in work by economists. We argue that SE is largely consistent with the research in LS. Rather than viewing this overlap as an impediment to providing a clear scholarly taxonomy with sharp distinctions between subfields and methods, we see it as an encouraging quality for our knowledge seeking project in the social sciences. In this section, we touch on the history of LS as an academic discipline and identify some of its major themes or lines of inquiry, including the nature of Latinidad, the role and operation of the nation-state in shaping the material conditions associated with different groups, and the process of racialization across time and space.
LS has developed as an interdisciplinary academic field (Delgado & Stefancic, 2010; Ramos-Zayas & Rúa, 2021). We can trace its origins to Chicanx and Puerto Rican civil rights and economic rights movements of the 1960s, 7 including Chicanx student walkouts in 1968 (Muñoz, 2007), Puerto Rican student and nationalist movements in New York (Cabán, 2022), and Cuban American politics in South Florida (Stepick, 2003). Scholarship on peoples in the United States and Canada with ties to Central American, South American, and Caribbean countries is sizable, though somewhat in the shadow of work on the “big three” (i.e., Mexicans, Puerto Ricans and Cubans). As part of the aforementioned civil rights movements and student protests, communities banded together to sign petitions and otherwise demand the creation of new departments and faculty positions aimed to enable Latinxs (and others) to create knowledge about their communities and their lives, and knowledge that could be used to transform the circumstances of their lives.
This origin story stands in contrast with that of the discipline of economics in several ways. LS has an indispensable grounding in political struggle, an epistemological concern with the process of legitimizing knowledge claims, and a marginalized position within the U.S. university system, in which LS is low-status and expendable whereas economics is (usually) high status, high enrollment, and relatively well remunerated. The LS scholarly agenda is tied to community well-being; for many Latinx scholars, professional success that marginalizes or leaves behind Latinx community members is a contradiction in terms. There is an ideal of politically engaged scholarship in LS that stands in contrast to a commitment to so-called “positive” economics. (See Cabán, 2003, for a critical retrospective discussion after three decades of LS about the field's antihegemonic goals.)
LS is not usefully described in terms of any particular or dominant methodology. As an interdisciplinary field, it regularly engages people with many different types of scholarly and nonscholarly training in conversation with each other. 8 LS is defined by its subject of research, Latinxs in the United States, people or phenomena related to them, and the different theoretical, empirical, political, and moral questions related to the lives of Latinxs. 9 The knowing subject in LS—the scholar, learner, writer, or reader—is usually Latinx themselves; some scholars would argue that this is characteristic of the field and has epistemological significance, but even those skeptical about scholarly subjectivity will recognize that most LS scholars are Latinx and that this has shaped the field into a pursuit of knowledge about “us” rather than creating knowledge about “them.” 10 For example, the question “Who counts as Hispanic?” is less interesting for those in LS than are the questions “How does Latinidad operate?” or “What subjugated knowledges can surface when we think from our position as Latinas/os/xs?” The LS knowledge seeking process is collective because (a) the LS scholar recognizes and operates as if the research matters to them as a member of the community being studied and (b) the LS scholar seeks a greater level of understanding than that which comes from narrowly applied disciplinary norms. The imperative to answer these questions “for ourselves” emerges from the recognition that the most widely circulated knowledge about subaltern groups has historically been produced by the dominant group, whether by individual members of the dominant group or by people in institutions primarily serving the interests of the dominant group.
The most enduring and prominent topic in LS is likely investigations into the nature of Latinidad itself. Scholars have done pioneering work in this area using many approaches in conversation with each other, including through fieldwork, textual analysis, archival work, and theorizing about the operation of power in the lives of Latinxs. Since these terms emerge, develop, and are contested over time, historical work is particularly helpful. Mora (2014) argues that the interests of various constituencies resulted in the general adoption of the label “Hispanic” for a group of people in the United States. For one, private firms, particularly TV and radio stations, were interested in “selling” their audience to advertisers, and they strategically grew their market by encouraging a larger group identity, a pan-ethnic identity, across disparate populations (Mexican Americans, Puerto Ricans, Cuban Americans, etc.). Civil rights organizations competed with other groups for legislative priorities: by employing a pan-ethnic identity, they were able to argue that they were the largest minority group in the United States, attempting to rhetorically displace African Americans in terms of deservingness of attention in the aftermath of the civil rights era (roughly 1954–1968). Finally, the U.S. government had an interest in responding to these changes, but as with the other decentralized processes, the decision to “update practices based on community identification,” on its face neutral, helped to reify “Hispanic” as a pan-ethnic identity.
Dávila (2008) argues that competing interests have driven common understanding(s) of who Latinxs are: For example, individuals and civil rights organizations working to improve Latinxs’ economic status carried out public messaging campaigns to associate Latinxs with being family oriented, hard workers, and interested in material economic mobility, characteristics that make them “good” consumers from the perspective of profit-seeking corporations. Conscious, strategic efforts were undertaken to associate positive cultural markers with Latinxs. The conclusion one can draw from these distinct but largely complementary (from the perspective of economics) accounts is that Latinx identity is shaped by intergroup competition and by local and large-scale economic systems such as capitalism. In contrast with the bulk of economics research, which seems to be ahistorical by design, the most influential works in LS draw heavily from history. In the next section, we discuss how this historical approach could inform economists’ study of Latinxs, with particular attention to how race—whiteness, blackness, mestizaje, etc.—operates for and within the imagined community of Latinxs.
Another major concern in LS is the nation-state and how people navigate multiple nation-states, particularly in the western hemisphere, including the dynamics of colonialism, imperialism, and national citizenship. Those writing in LS largely agree with the critical race theory perspective articulated by Ray et al. (2017, p. xxii): “Rather than seeing the law as separate from racial power—a neutral arbiter or umpire just calling balls and strikes— [scholars] understood the law as a tool that occasionally empowered people of color but usually advanced whites’ racial interests.” Chicanx scholars have written about how historical and contemporary U.S. policy toward Mexico produces subjugated peoples and identities. The economic development of the United States came at the cost of the underdevelopment of people with ties to land that was formerly Mexico. Puerto Rican scholars have written about how imperialism underdevelops the archipelago. Reading across this literature shows us how the nation-state is key to understanding both Hispanic identity and the economic status of Latinxs in the United States.
We can use this perspective to critique disciplinary norms in related fields. For example, economists routinely ask public policy questions that are treated as neutral (e.g., “What are the policy implications?”; “Which among X policies achieves Y most efficiently?”), but LS and other scholarship remind us that governments as emergent phenomena are the products of intergroup competition, that the interests of all citizens or even all minority groups are not weighted equally or justly, and that the interests of noncitizens and subaltern groups are often not valued or may be targeted for exploitation. 11 The voluminous research on so-called illegal migration in economics, insofar as it does not engage with LS or ethical questions and uncritically reproduces unjust hierarchy, reflects this problem (see Hanson, 2007; and Orrenius & Zavodny, 2015, for overviews).
One methodological conclusion that can be drawn from reading across the LS literature is a call to work carefully, strategically, and historically with existing national borders (Wimmer & Schiller, 2002). The LS scholar knows that people(s) cross national borders and that methodologies that render these individuals or their border-crossing relations invisible or selectively visible is a research challenge for understanding peoples’ lives. At the same time, the politicized scholar understands that for Latinxs, the border represents a lot of things, from dispossession and violence to the relative safety of invisibility; thus, LS scholars work in a variety of ways to undermine injustice insofar as funding or data from nation-states in practice may reinforce the harm of national borders.
The last major theme we identify in this section is racialization of and within Latinx communities. Various approaches have been helpful for building multifaceted and useful understandings of racialization for Latinxs. Writers contributing to LS have written extensively about racial hierarchy schemes operating in the western hemisphere (Telles, 2014). In addition, there is a long-standing debate in the literature about the nature of race and ethnicity, with different camps arguing that real-world phenomena are best understood in terms of colonialism (Grosfoguel, 2004; Grosfoguel & Georas, 1996), ethnicized races, racialized ethnicities, or other schemes. For the purposes of LSE, it is not necessary to take a position in all of these inter-LS debates, particularly since systems of racial inequality and hierarchy are always multiple, contested and overlapping, geographically and chronologically specific, and intersectional; that is, racialization is inextricably linked with other systems of domination. There is considerable common ground. Historically, Europeans had a material interest, starting with early contact, in the natural and cultural resources of Indigenous peoples in the western hemisphere. 12 Racial hierarchy systems adapted in response to this material interest and the concomitant need to justify and rationalize brutal imperial policy and practices. One of the most significant strengths of the SE framework is that it can help to explain racial inequality through longer periods of time than the alternative frameworks used by economists: the same rational, self-interested behavior that drove colonialism, slavery, and de jure racism can explain racial inequality today, at least in part. Though the institutional contexts and outcomes may have changed, the relative position of various groups has changed much less, particularly with regard to African Americans, dark-skinned Mexicans, and American Indians.
Nation-building projects often involved producing and maintaining racial in-groups and out-groups. Along with Critical Race Theory and other fields, LS recognizes the state as a prolific creator of subaltern groups. For example, the writer and public intellectual José Vasconcelos's self-conscious efforts to define a national race (raza cosmica) proper of Mexico have been critiqued and characterized, particularly in Indigenous and Afro-Latinx studies literature, as a form of group erasure, with implications for LSE considering the stratified conditions of Afro-Mexicans and Afro–Latin Americans writ large. The fact is that Latin America has a large Afro-descendent population for the same reason that the United States does, namely, the historical trans-Atlantic slave trade, and articulations of anti-blackness and positive identification with blackness are not limited to the United States. 13 Understanding how individuals navigate and express agency within multiple overlapping racialization schemes is central to LSE. 14
In the following sections, we apply these ideas to the main theories used in the economics literature to understand Latinxs, starting with the concept of Latinx identity, or Hispanics as an analytical category, and continuing with economic theories of migration, assimilation, and discrimination.
Toward a Latinx Stratification Economics
“Hispanic” as an Analytical Category
There is growing demand for information about “Hispanics” among economists, policy-makers, and Hispanics themselves, but the theoretical frameworks used by most mainstream economists to study this group are ad hoc and inadequate. The first problem is that mainstream economics does not have a workable theory of racial difference and racial hierarchy. For example, mainstream economics does not have much to say about the nature of Latinidad, what makes someone “Hispanic,” or how this identity itself matters (as opposed to proxies or characteristics associated with Latinxs such as immigration status or skin color). 15 Despite this, dummy variables for race and “Hispanic” are common in the literature, as is economics research on Latinx populations and issues that disproportionately affect this group.
Economists are typically not well trained in the modern view of racialization as reflected, for example, in LS. The graduate materials, mainstream economics resources, and general positivist orientation of the discipline do not offer much guidance for a serious study of racialization that can be brought into conversation productively with work outside the discipline. A serious study of race, for example, would understand it as a process or social phenomenon, not a static characteristic of individuals. It is common to hear researchers in economics express frustration about the category “Hispanic.” The authors of this paper have heard economists say that the label “Hispanic” is difficult to interpret, is inconsistently used across surveys, often changes over time even within the same survey, is unscientific, or has limited precision as a category. This reflects misguided expectations on the part of the researcher because the desire to have discrete, clear, externally perceptible, and inflexible racial categories is precisely a characteristic of the white-colonizer gaze (Robinson & Kelley, 2000).
We argue that the problem of theoretically organizing, empirically eliciting, and interpreting racial categories in research has important implications for work in economics. The status quo among economists is to use whatever racial or ethnic categories are provided in the data and to use race as a control variable as opposed to an outcome, typically to increase precision in estimating the coefficient of interest. If provided with greater detail about racial self-identification (i.e., more than a handful of possible categories), researchers usually follow U.S. government sources (e.g., the Census) and create categories such as white non-Hispanic, black non-Hispanic, and Hispanic. The implicit ideas here are twofold: that the federal government should receive deference in defining group membership and that race matters for every group except for Hispanics, where a panethnic identity supersedes race.
Most research on Latinxs goes against the latter claim (Flores & Román, 2009; Román & Flores (2010); Holder & Aja, 2021). It is not uncommon to see a regression with regressors for education, industry, state, and so on that also includes a control for Hispanic: Including controls for premarket factors that are themselves dimensions of racial inequality changes the interpretation of the Hispanic coefficient (and, for that matter, any similar social category) in ways that are not often discussed (VanderWeele & Robinson, 2014). The Hispanic variable seems to be a flexible proxy for any number of characteristics—language ability or accent, citizenship and legal status, skin color, culture, preferences—and the economics literature tends to see these as discrete, one-off deviations from a central, color-blind framework (mainstream economics). 16
The Latinx stratification economist, following the best work in LS, reads across disciplinary perspectives and evidence to integrate knowledge about the daily lives of Latinxs. Put differently, when “Hispanic” is only used as a proxy for specific other qualities, or when Latinidad is only explored through ceteris paribus studies that attempt to separate characteristics that are linked practically, representationally, and symbolically, it is easy to fail to understand how Latinidad itself operates or, more broadly, how racialization operates in the United States. Of course, this sort of theoretical move that naturalizes differential group outcomes with reference to supposed explanatory factors, a denial of racialization as a social phenomenon with real effects, is itself one way in which scholarship upholds white supremacy and racial hierarchy (Aparicio, 2017), however unintentionally. This is a question not only of epistemology and methodology but also of power and incentives in academic research: The qualities that legitimize mainstream approaches in economics today rely on the researcher not making general claims about the nature of racialization (i.e., engaging in social theory that strays from market fundamentalism) or the moral dimensions of race (i.e., having an explicit dimension to anti-racism that transcends efficiency arguments) so as to remain “objective.” 17 , 18
SE and LS both provide a great deal of guidance on the research questions and problems raised thus far. Both of these perspectives provide robust theoretical frameworks and evidence for how to think about “Hispanic” identity, but to the extent that these insights are incompatible with mainstream economics or inconvenient for advancing one's career in academic economics, they are largely ignored by the mainstream of the discipline. 19 The SE perspective tells us that group identity is a tool used by individuals and groups to maintain or augment their group's relative privilege in a social hierarchy. Whereas people associate “Hispanic” identity with particular culturally coded behaviors, family migration history, the Spanish language, or some other trait, the Latinx stratification economist sees racial and ethnic categories as flexible, evolving, and historically produced, used to structure strategic intergroup competition. One of LS's most significant contributions is doing the archival and other data collection “spade work” and theoretical work of historicizing identity formation (Mora, 2014; Rodríguez, 2000). LS documents and studies the rich spectrum of Latinidad and racialization of Latinxs in the United States, including black and/or Afro-Latinxs and/or Afro-descendiente, white Latinxs who benefit from and take actions to maintain racial hierarchy, and people who may be ascribed or racialized with a Latinx identity but who reject it for various reasons (those who may identify as Indigenous, Garífuna, “American,” etc.). As stated earlier, some of the central questions of LS consider not only what it means to be Latinx in the United States, but also how the term “Latinx” evolves over time, who gains or loses from its different formulations, and how this social category is related to colonialism, imperialism, and other examples of state power. This is where we see complementarities between the SE approach and LS: The history of “Hispanic” identity can largely be told through the lens of competition for relative position in a social hierarchy where material benefits are on the line, among other things, including discursive and hegemonic authority. 20
The LSE approach that we adopt, consistent with SE and LS, accepts race as a powerful force in everyday life in the United States that both works to the detriment of subaltern groups and opens up conditions for resistance and group-based social justice. Racialization and racial group categories—including Hispanic, Latino, Mexican, brown—have no reliable biological/genetic content, no universal consistency, and no logical coherence except the fundamental dynamics of intergroup competition. The stratification economist, informed by LS, does not unquestioningly use whatever race and ethnicity categories are given in the data. Instead, they seek to denaturalize racial categories and see race as a strategically produced hierarchy. At the material level (production, consumption, distribution, ownership, use, etc.), racial difference is primarily viewed as emergent from a process of intergroup competition (read: endogenous). Thus, the primary role that (stratification) economists can play in academic research on racialization is to emphasize the economic, material manifestation of race-based competition and oppression. As the debates within LS indicate, moves to reproduce racial categories or reify particular categories within research are political and moral (López, 2013; Telles, 2018). The corrective is to work with a normative orientation toward social justice, with curiosity about people and the real world that is not limited by deference to hierarchical forms of knowledge (e.g., limited to scholarly publication in a mainstream discipline's “top journals”) and that attends to the multidimensional nature of racialization, including phenotype, skin color, accent, name, legal status, generational wealth, and culture as dimensions of race.
A focus on racialization leads to a more rigorous and liberatory knowledge seeking projects for Latinxs. By racialization, as opposed to race, for example, we mean that race at any one time is treated analytically as the product of an evolutionary process, as socially constructed, and as contested. Note that the term “evolutionary” is used here to refer to a social process over time that has emergent properties, as in evolutionary economics, not with reference to the theory of biological evolution of populations. The significance of racialization as an evolutionary process is threefold: The historical changes in how race works in any society are neither teleological nor static. Also, race is constructed: It is part of a meaning-making process such that beliefs, expressions of human culture, and so on can help us to understand the competition over relative position in a social hierarchy and related competition over material interests. 21 Finally, it is contested: Although race has a large structural component, and the typical chain of causality runs from structural characteristics to individuals, individuals/groups also have agency to affect the structure (Arnsperger & Varoufakis, 2008; Jackson, 2003; Sewell, 1992), and there are commonly multiple overlapping racial regimes that are significant to the daily lives of individuals and communities. Both SE and LS view racial groups through this lens of racialization. To be clear, this is different from an argument that says that other formulations (static, intrinsic, unvarying, fixed) are simply lacking nuance: The problem is deeper and has a direct parallel to the problem in mainstream economics modeling of treating a key model variable as exogenous when it is in fact endogenous to the other variables of interest.
We conclude this discussion by suggesting a test to determine whether research on Latinxs is consistent with this argument: The Latinx stratification economist uses racial and ethnic categories in ways that enable or enact liberatory politics. The determination of what precisely is “liberatory” in any given context is purposefully not specified, but at a minimum one's reasoning that a research methodology is liberatory must be true (based on evidence and reasoning) and it must be politically oriented towards justice. In practice, this determination falls on individuals and communities: We can define liberatory politics as providing for capabilities for all peoples (Nussbaum, 2000), or equality, or ensuring legitimate economic human rights (Aja et al., 2018). Whether something passes this proposed standard depends on one's own values and reasoning based on current (political) circumstances; regardless, this proposal represents a step forward because working with questions about ethics or political strategy is so unfamiliar to most mainstream economists today. Just as dissent and critique are necessary for democracy, pluralism in terms of progressive visions and political strategy is necessary for liberatory politics. 22
Migration
In the previous section, we discussed “Hispanic” as an analytical category. In the following sections, we comment on three important theoretical frameworks commonly applied to Latinxs in the mainstream economics literature: migration, assimilation, and discrimination. We articulate a critique proper to LSE (i.e., informed by both SE and LS), and we offer criteria for evaluating research on Latinxs informed by this perspective. The core finding with regard to theories of migration is that an approach that draws on both SE and LS shifts the focus (a) from human capital models of migration toward approaches that account for wealth and privilege, including racial privilege, and (b) from individual selection models to models that subordinate migration to colonialism and imperialism, that is, models that account for development–underdevelopment relationships and political economy as applied to nation-states competing for international economic, political, and military hegemony.
When not viewed as a generic racial minority, Latinxs are primarily viewed in economics research as immigrants or descendants of immigrants from the Spanish-speaking Caribbean and parts of Central and South America. The human capital model of migration is the dominant tool used to study migration in mainstream economics (Amuedo-Dorantes & Pozo, 2015). In this framework, the migration decision is taken by atomistic individuals who perform a comparison between their wage at home and their possible wage in a foreign country, net of migration costs (Sjaastad, 1962). 23 Human capital is usually operationalized as years of formal education, sometimes with additional dimensions included such as years of work experience, and for each level of human capital, there is an associated expected wage determined by the decision agent's characteristics and the labor market characteristics of the sending and receiving countries. Relative differences in wage distributions between countries thus explain negative or positive selection of migrants by education (Borjas, 2008). LSE critiques both the grand narrative of Latinxs being immigrants or descendants of immigrants, drawing mostly from LS, and the human capital framework of understanding migration, drawing mostly from SE. The critique emergent from the LS literature is not that mainstream economics research is not sufficiently nuanced or that it is too quantitative or mathematical, but that much of economics research about Latinxs runs counter to a more rigorous (robust to multidisciplinarity) understanding of migration. The critique operates on multiple levels, including epistemology as well as model choice, variable choice, and treatment of different variables as endogenous or exogenous, and this critique is structural, which means that it cannot be fully described in reference to a single article or author but rather can be seen as a pattern in recent and contemporary, mainstream economics.
In reality, most Latinxs are not immigrants (in 2015, only 34.4 percent of the Hispanic population were foreign born, Flores, 2017). An alternative to the immigrant/foreigner and human capital paradigm describe above is one based on colonialism and imperialism. Large parts of the United States were illegally and unjustly seized through the colonial annexation of Puerto Rico and parts of Mexico, 24 and people with ties to these two countries represent the majority of Latinxs in the United States. 25 As these acts were designed to (and did) benefit white Americans and capitalists, their ramifications are thus properly measured not just through (Indigenous American and Latinx) dispossession but also through (white American) unjust privilege including, for example, excess wealth accumulation (Lui, 2006). 26 Not only did these acts of dispossession differentially enable or interrupt intergenerational transfers of wealth, with implications stretching to today due to insufficient policy responses to redress this injustice, but the acts of dispossession, exploitation and enslavement were accompanied by concomitant racist ideologies that have had their own independent effects. The “Hispanic/Latino immigrant” narrative obfuscates the violent history of land seizure against Mexicans during the era of U.S. expansion (Gomez, 2008; Gonzalez, 2001) and the longtime, violent colonial subjugation of Puerto Ricans since 1898 (Grosfoguel, 2003). Contemporary barriers to free movement and other barriers to connections between “sending” countries and migrants, develop out of political processes: “The role of both the sending and receiving states in fostering or hindering social spaces where immigrants can orchestrate their transnational activities is crucial” (Menjívar, 2021, p. 122). And state violence, material disparate treatment itself is what creates immigrants, not inherent qualities in those people: “The illegal immigrant cannot be constituted without deportation—the possibility or threat of deportation” (Oboler, 2021, p. 77). Speaking in the language of mainstream economics today, we might say that the economic status of particular Hispanic groups in the United States is a product not only of a selection process (migration) but also of other processes, namely colonialism and imperialism, that have had first order effects through multiple generations.
Besides Puerto Rico and Mexico, researchers have argued that broad patterns in the origin and timing of Latinx migration are inextricably linked to U.S. foreign policy in Central and South America and the Caribbean (Gonzalez, 2001). We will discuss the Harvest of Empire thesis separately, but the main point for now is that even for non-Mexican, non-Puerto Rican Latinx groups, the dynamics of the Good Neighbor policy, Cold War, and other expressions of U.S. nation-state domination have significantly accounted for the movement of peoples and thus should not be ignored by economists writing about migration (Mayblin & Turner, 2021). One critique we might expect from mainstream economists is that the political dynamics of colonialism and imperialism are better dealt with in other disciplines. The LSE response is to reaffirm a view of the economy as embedded in society. Failing to account for important political or social phenomena results in bad economics, not a better, purer form of economics that is isolated from other disciplines. The alternative LSE approach helps us to center wealth/dispossession, the nation-state, and history.
SE research points to selection based on wealth and privilege in place of or in combination with selection on the basis of education and years of experience (operationalized human capital). In other words, wealth and privilege matter in the migration process, and they are important dimensions of analysis for answering questions about migration patterns. The decision to focus almost exclusively on so-called human capital is, in part, a rhetorical move that has the effect of justifying market fundamentalism since migration based on human capital is efficient and thus increases the size of the metaphorical pie. The problem is that while education and skills do matter for migration, the human capital model almost entirely disregards initial conditions or history; the distribution of wealth and privilege, both of which often enable education, is passed down in families and social classes under conditions that have been shaped by a rational, intentional process of competition, exploitation, and domination based on socioeconomic class and social groups, including racial groups. Part of the disregard for wealth and privilege in neoclassical economics is justified theoretically with reference to credit markets. Assuming perfectly competitive credit markets and no market failures, it should be possible for anyone with high skills to migrate if it is optimal, regardless of their wealth. But, of course, this is not typically how migration works. Also, individual productivity itself is mediated by social and political processes and by institutions like race and gender, so that the returns to any given type of human capital are heterogenous: If we are interested in intergroup inequality, which is often where Hispanic identity comes up in economics research, human capital does not elucidate much.
In this discussion of migration, a richer model with a fundamentally different focus is emerging: From human capital models of migration to models that include wealth and privilege, and from migration as the main lens for thinking about Latinxs, we shift to racialization, colonialism, and imperialism as alternative critical lenses. These critiques that help form the LSE perspective follow from nearly any reasonable reading of the LS literature. LS emphasizes the role of the nation-state in Latinxs’ lives, which leads to a recontextualizing of much research in the economics of migration that focuses on different migration channels. The economics literature creates analytical distinctions between so-called economic migrants, refugees, asylum seekers, “illegal” immigrants, undocumented immigrants, deportees, and so on. These distinctions are legal and political: The nation-state creates the rules that people interact with, and these rules create and reinforce social hierarchies in ways that typically advantage dominant groups (Asad & Clair, 2018). The LSE scholar works critically with immigration categories created by the nation-state. The de facto and de jure rules that differently structure the lives of citizens and noncitizens extend to nearly every realm of life, not just citizenship laws, work permits, and travel restrictions but also, for example, occupational licensing and how credit scores are calculated and what they are used for. Migration laws are not exogenous, 27 and the options available to people are the product of a political system that is intended to bring benefits to those in the dominant group and to shift costs to those in marginalized groups.
U.S. government policies create different legal and de facto categories of migrants through differential treatment of people, and this treatment is racialized and linked with a foreign policy framework that aims to maintain (or regain) United States regional and global hegemony. Material incentives explain a great deal of this process because dominant groups and interests in the United States benefit from this state of the world. What separates an “economic migrant” from a “refugee” is the outcome of the nation-state's process of creating deserving and undeserving populations, not any difference in the characteristics of migrants (though in practice any two historically and geographically distinct groups may differ on average or along any number of different categories). 28 This way of thinking stands in contrast to an approach that maps a small number of characteristics into a choice-theoretic framework (exogenous preferences, common objective function and constraints) without engaging with history or dynamics through time. It is not that some groups better match the profile of theoretical economic migrants and others more closely match the theoretical profile of refugees: Cubans seeking to migrate after the Cuban Revolution were “seen” as different from working-class Mexicans crossing the US-Mexico border, different from Haitians seeking to move to the U.S., and similar perhaps to Vietnamese migrants (Aja, 2016; Grosfoguel, 2003). While LSE seeks to build accurate and liberatory knowledge, the role it plays in economics is to deconstruct, denaturalize and present alternatives for ways of thinking that justify, rationalize, or obfuscate inequality based on social group identity.
The most prominent story omitted by economists’ neoclassical models of migration is the one journalist Juan Gonzalez tells in Harvest of Empire and that many others analyze in detail (Gonzalez, 2015, 2020): U.S. policies have created conditions that lead people to seek lives and livelihoods in the United States. 29 Such policies have included Manifest Destiny, the Good Neighbor policy, the Cold War and proxy wars, structural adjustment policies, climate catastrophe policies, and ongoing intervention throughout the western hemisphere. The neoliberal policies of other governments in the region, too, affect migration patterns.
What economists should learn from LS is the imperative to add the nation-state to our analytical toolkits. The nation-state and interactions between nation-states are not likely to be well understood if we limit our tools to mainstream economics because the nation-state's relationship with subaltern groups is rarely best understood through neoclassical economics or causal inference (the nation-state's policies are rarely exogenous). LS draws on the work of Latin American structuralists and related anti-colonial thought (e.g., Di Filippo, 2019; Rodney, 1972) to understand the interrelationship between development in the Global North and underdevelopment in the Global South. The interconnected issues of imperialism and colonialism, neoliberal policies, global military and economic hegemony, capitalism, and racism are not just added nuance; our understanding is critically flawed if we only study one dynamic at a time without ever accounting for closely linked systems of domination. In particular, this analytical mistake limits our capacity to change these structures because they are mutually reinforcing. Indeed, members of the dominant group have an interest in making government action (or inaction) invisible by selectively taking notice of the nation-state.
Assimilation
Another framework often applied to Latinxs in the economics literature is assimilation theory. In this section, we summarize neoclassical economics work on assimilation and present a critique that draws from the SE and LS literatures. We explain how migration, assimilation, and discrimination in economics are conceptually linked to each other and to the monistic nature of economics and its neoclassical core, and we describe the ideological function of this framing, and describe our alternative framework that we label LSE. The first insight from this section is that, unlike mainstream economics literature, LSE does not expect that the market mechanism in the real world leads to the erosion of intergroup disparities. The second insight again involves turning the focus from supposedly normative questions around measuring outcome gaps to issues of social justice: The main concern is not whether groups are resembling the non-Latinx, nonimmigrant population, which is often framed as normative, but whether life chances and the material conditions of life are conducive for human flourishing through diverse lifeways for all groups.
The basic research question for economists working within an assimilation framework and studying Latinxs follows from the grand narrative of Latinxs as immigrants: If Latinxs are mostly a group of immigrants or people with family ties to immigration, how do they fare in the United States over time? Although the use of the assimilation framework in economics research is not as monolithic as the use of theories of migration based on human capital, we can nevertheless characterize several patterns. This literature aims to describe how well particular immigrant groups have assimilated and the factors that determine whether this assimilation happens quickly or slowly, or stalls, or reverses. More narrowly, this framework supposes that groups of people who come to the United States may differ from the nonmigrant population in average economic outcomes, but that over time and over generations, the distributions should resemble each other. This can also be stated as the idea that well-functioning, competitive markets drive outcomes that we see in the economy, and that any deviations from the prediction of nearly identical distributions of characteristics must be due to cultural differences, genetic differences, or (purportedly anachronistic) discrimination. 30 In other words, chronological time and the market mechanism are forces for good. 31
Assimilation also has a more common, nonacademic meaning: that people of different cultures who come to the United States eventually adopt mainstream American culture, perhaps with some differences that are acceptable to the mainstream or that can be thought of within the melting pot ideal. Assimilation in economics is usually defined empirically as a convergence in outcomes (e.g., years of schooling, hourly wage, employment rate) to the nonimmigrant population, or sometimes the white nonimmigrant population. On its face, then, assimilation in the economics literature is purely understood in terms of economic or demographic trends: A mainstream economist might claim that “assimilation” is more about achieving equality than it is about losing minority culture. Little explanation specifies mechanisms that lead to assimilation besides a change in culture (e.g., valuing education) or removal of barriers that could cause minoritized individuals to behave differently from members of the dominant group. Because race and ethnicity have no special places in mainstream economic theory, the assumption seems to be that, through using the same rational decision-making as everyone else, difference will quickly disappear. The denial that there are cultural assumptions embedded in the figure of homo economicus thus leads to an inability to recognize cultural assumptions in the application of mainstream economic theory.
According to some economists, relative differences in the achievements of particular racialized minority immigrant groups in the labor market can be explained by cultural differences (Chiswick, 1983). In contrast, SE proposes the lateral mobility hypothesis, which states that the relative socioeconomic status of an immigrant group is related to the relative position of the adults that constituted the bulk of the original members of that group (Darity et al., 2001). For Latinxs specifically, one example is the post-1959 Cuban American community. Often viewed as possessing the right cultural imputations that led to its so-viewed success (Chua & Rubenfeld, 2015), the lateral mobility hypothesis of SE helps explain why this relative “success” has only been true for early exilic arrivals (1959–73), their families, and subsequent generations, but not necessarily for subsequent arrivals, including Afro-Cubans of all stages of migration (Aja, 2016). 32 The SE perspective, through the lateral mobility hypothesis and other frameworks, allows for “importing” stratification. That is, while competition for relative position in a social hierarchy may look different in different settings, the conditions of that competition in the United States for many groups are based on what that competition primarily looks like in other countries. Aligned with other research on race, the stratification economist expects that the market mechanism, even given many generations, will not erode away supposedly inefficient biases like racism (Derenoncourt et al., 2022). SE tends to draw distinctions that other theories overlook. For example, while phenotypically similar groups are sometimes compared in order to isolate the role of culture separate from externally perceived race (or “street race,” see López et al., 2018), SE emphasizes intergenerational transmission of privilege, including wealth and status; for the Latinx stratification economist, race is a multidimensional social construct not reducible to skin color.
Using neoclassical theory to model assimilation by racial group, to understand inequality for example, is misleading because of inattention to initial conditions and faith in the equalizing effect of markets. Neoclassical economics assumes a great degree of rationality and agency; It presumes that individuals’ actions are rational from the individual's perspective, but it fails to adequately account for starting conditions or to allow for an individual's interest in the relative position of their social group. It is not (just) that individuals from subaltern groups face constraints that members of the dominant group do not face, but that the rules of our economy are structured by racial difference; even without different constraints, the same behaviors or achievements by minoritized Latinxs, black people, and other visible racial minorities often have different effects. The LS view fundamentally emerges out of a social justice or human rights framework. “Assimilation” is not uncritically accepted; the concerns are injustice and exploitation. In some cases, this means that assimilation itself is an inappropriate framework. The LS perspective recognizes human rights, including cultural rights, as fundamental to thinking about inequality and racialization.
The topic of language skills can illustrate the critique of neoclassical economics that emerges from SE and LS. Language skills in mainstream economics are treated as human capital, an embodied skill that is associated with returns in the labor market. The basic idea is that language skills usually take time to develop and that workers can invest in gaining these skills if they don’t have them already in order to increase their productivity in particular occupations. Economists studying Latinxs and language skills can learn from the extensive research in LS about what communicating in different ways means to Latinxs in the United States and how this has changed over time and differs across place. Recent work in raciolinguistics (Rosa & Flores, 2017) is largely consistent with SE in presenting an alternative framework: Certain types of language abilities and language use are labeled and treated as deficient, while others are valued as part of intergroup competition and racialized capitalism. 33 Rather than seeing language use as a particular human capital, we can see it as a form of whiteness as property (Harris, 1993). There are many examples of the racialized dimension of language in the United States, including the preference for Spanish from Spain (Zentella, 2007) and preferences for unaccented speech. We can see the nation-state involvement as efforts to make English an official language move forward, now the law in 29 states (Zentella, 2007, p. 34), and as fights over bilingual education continue to be decided on the side of limiting the material benefits of speaking Spanish in the United States. 34
Discrimination
With few exceptions, the bulk of the writing by mainstream economists theorizes racism and racialization narrowly as discrimination. 35 The most significant omissions in this work are the failures to develop theories of structural racism (Mason, 2023) and theories of race and racial inequality that are based on the material incentives that members of the dominant group have to maintain, expand, and strengthen a system of racial hierarchy. Put simply, whites in the United States benefit from racism, which gives them incentives to maintain or expand this system, and purposeful action, largely on the part of those who benefit from racism, explains the fundamental drivers of racial inequality. It is ironic that academic economics, which historically has treated motivations other than self-interest as suspect, produces so many works that argue instead that the underlying drivers of racial inequality are information asymmetries, unfortunate but apolitical fallout from distant historical events, unintended consequences of policy, or unconscious and irrational beliefs. The arguments in this section are (a) that economics conceives of racism too narrowly as discrimination, (b) that SE provides a more robust theoretical framework that, unlike neoclassical economics, is consistent with work in LS and other humanistic social sciences, and (c) that LS provides an important base of knowledge from which to analyze the ways in which some Latinxs can benefit from and perpetuate racist systems. We begin with a review of the three main theoretical frameworks used to account for racism in economics: discrimination based on animus, so-called statistical discrimination, 36 and unconscious bias. 37
Gary Becker–style models are neoclassical theories for discrimination based on animus. In these models, agents have preferences for avoiding interactions with members of particular groups, and these preferences are fully fungible, expressible in terms of utility and resources, as revealed by behavioral choices (for a critical perspective, see Figart & Mutari, 2005). In most circumstances, the individual who discriminates due to animus stands to lose either because they reject or undervalue workers who happen to be members of a disfavored group (employer side) or because they end up paying more to avoid purchasing goods and services from members of the disfavored group (consumer side). The main theoretical assumption is that discrimination only produces individual psychological benefits while producing material disadvantages and costs to the discriminator and the main prediction from Becker-style models based on irrational animus is that the market mechanism in most cases will drive out racism. It is not the level of abstraction or the mathematical nature of this work that should be the primary target of critique. Rather, the issue is that this abstraction and mathematical sophistication is put in the service of a model that is at odds with reality, part of an ideological framework that in practice opposes bold policies for achieving racial justice, and largely irreconcilable with work on similar topics from other disciplines.
Statistical discrimination models of racism involve information asymmetries that lead agents to apply perceived average group characteristics to individuals from that group (Phelps, 1972). For example, when faced with a hiring choice, an employer may not have full access to accurate information about a candidate's human capital, but they can observe characteristics that signal group membership (race, gender, age, etc.) and ascribe to the individual the perceived characteristics of their group. The social problems at the heart of this model arise for qualified individuals who are members of groups perceived as having low human capital and for all individuals from this group, the dynamics of statistical discrimination mean that predictions of low human capital attainment may be self-fulfilling as opportunities to build human capital are denied to qualified candidates or as individuals limit investment because they anticipate discrimination. Unlike animus-based discrimination, statistical discrimination usually has benefits for the discriminator and may represent gains in economic efficiency if the proported group traits accurate. On the other hand, statistical discrimination doesn’t explain historical racial inequality—chattel slavery of African Americans in the United States, Jim Crow laws, lynching, illegal land seizures, overt labor market discrimination, denial of political rights, whiteness and racial hierarchy as normative within many important institutions—and it doesn’t explain the social process wherein racial groups form or why it is that negative perceptions about characteristics are systemically associated with black people in the United States. As Mason (2023, p. 61) writes, “No African was enslaved because (s)he was raised in a pathalogical family, had dysfunctional values, had ruinous behavior, or had inferior skills.”
While the insights captured by the theory of statistical discrimination may be useful, it is not possible to explain racial inequality meaningfully using just this theory. Some scholars argue that the theory is not only marginal in its possible applications, but that it is harmful: Tilcsik (2021, p. 93) argues that “rather than simply providing an explanation,” exposure to the theory of statistical discrimination “can lead people to view social stereotyping as useful and acceptable and thus help rationalize and justify discriminatory decisions.” In contrast, LSE provides a realist explanation for why individuals take actions and maintain institutions that promote racial hierarchy that is largely consistent with the literature outside of economics—namely, that whites benefit from such a system—but this explanation, rather than justify such behavior, it can lead to reflection about ethical and moral principles that condemn racism and exploitation or harm of others, regardless of whether it is group-based. While members of the dominant group may benefit materially and, to some degree, psychologically (Roediger, 2007) from racism, they clearly do not benefit morally or spiritually.
The third dominant framework for discrimination, and the latest to get a spotlight in the mainstream of the economics profession, is so-called unconscious or implicit bias (Bertrand et al., 2005). The basic idea is that some of an individual's attitudes or preferences may not be conscious, that these attitudes may affect behavior (particularly when the decision needs to be taken quickly or in other circumstances that interfere with cognitive reasoning), and that this bias can be measured using procedures like the Implicit Association Test (IAT). Although independently developed and thoroughly different from “taste for discrimination” and statistical discrimination models, all three models share some features: none of the approaches accounts for the origin and change over time of racial groups and racial inequality, none is based on a rational model for racism, and none accounts for the material interest that members of the dominant group have in maintaining racism. Of the three dominant approaches to discrimination, none draws on the literature around systemic or structural racism. Bonilla-Silva (2021) argues against these types of approaches, saying, “by focusing on the ‘racists,’ we keep attention away from the structure and collective practices that maintain racial domination, which... is “the goal of racism.”” Whereas SE can be based on individual behavior, a focus on individuals within a social hierarchy always brings attention back to the structure within which individuals have agency. The IAT, which operationalizes implicit association theory and is by far the most popular psychometric test for implicit association in published economics research, has been criticized as a on conceptual grounds by psychologists (Fiedler et al., 2006) and by economists on the grounds of external validity (Lee, 2016). Bayer and Rouse (2016) shows how the theory of unconscious bias is being used by the top of the profession to inform policy as a response to racism in academic economics. While it is encouraging that leaders in the profession are considering evidence-based approaches, it is not clear whether a strategy based around implicit discrimination theory will ultimately lead to meaningful change for the better or whether, more significantly, it will forestall criticism while preserving as much of the racialized social system as possible (see Bell, 1980, on interest convergence).
Racism, and discrimination more narrowly, are routine in the lives of many Latinxs in the United States, particularly non-white Latinxs. According to a recent survey, nearly one in three Latinxs reported being personally discriminated against when applying for a job, 37 percent had experienced racial or ethnic slurs, and 17 percent had avoided calling the police in an emergency because they thought they would be discriminated against because they were Latino (NPR, 2017). Discrimination is an important phenomenon, but it is only part of a larger system an important part of racialization (Gans, 2017). It is more accurate and more politically challenging to established hierarchies to think of racialization as a system that goes well beyond discrimination on an individual basis (Cobas et al., 2009). To be clear, the only alternative to discrimination is not an identical version of current institutions ruled by meritocracy with no bias, where criterion and values remain the same but decisions are made with no regard for race. The fields of feminist theory and ethnic studies, including Latinx Studies and African American studies, have provided a critique of meritocracy, saying, in short, that “the very definitions of competence and worth that employers use are shaped by broader societal forces and are biased against certain groups” Tilcsik (2021, p. 97). This has direct implications for anti-discrimination work. Castilla and Benard (2010) ran three experiments in real world firms that adopted anti-discrimination programs and found that in all three cases, discrimination increased after firms shifted to a more meritocratic system. 38 The lesson from this paper is that social competition, including unjust discrimination and exploitation, can take place through the language of formal equality or other abstract liberalism (Bell, 1992; Bonilla-Silva, 2018).
Not all Latinxs and Latinx communities experience discrimination or racialization in the same way. While the mainstream economics literature focuses on disadvantage associated with being Latinx, stratification economists, like much of the recent scholarship in LS, allow for and study situations where Latinxs seek and benefit from racial privilege, either by strategically deploying Latinidad or by identifying as white. 39 Darity et al. (2002) presents evidence of a Latinx “preference for identification with whiteness” based on a survey where individuals with dark or very dark skin tone identify as white or as some other racial group other than black. There is significant scholarship documenting anti-black sentiment among Latinxs linked both to the racialization process affecting immigrants to the United States (McClain et al., 2006) and to “imported” anti-black attitudes and beliefs (e.g., Clealand, 2017), though this exists alongside countless examples of non-black Latinx solidarity with black people (Aja et al., 2012). Racism is important to the lives of Latinxs, but it would be a mistake to see Latinxs as only victims when they can be instead of also perpetrators or beneficiaries of racism. Whiteness itself is a flexible category, conditionally admitting and able to be deployed by some Latinxs (Beltrán, 2021; Márquez, 2023). The research implications for economists are to consider the multidimensional nature of race (including local racial hierarchies and non-United States based hierarchies), to see racial identification as strategic (an endogenous variable), and to not assume that interactions between co-ethnics are benign or mutually beneficial. The logic and contradictions of whiteness are just below the surface when studying racialization within the Latinx community: White or light-skinned Latinxs are overrepresented in media, but their whiteness is rendered invisible, it is either taken for granted (the face of Latinxs in the United States is fair skinned) or, alternatively, taken as ridiculous (someone can’t be both Latinx and white, Latinxs can be considered “brown” even with a very fair skin tone). These possibilities, seemingly at odds, constitute but one illustration of the logic of race as social hierarchy; they are all consistent with the idea that individuals employ multiple strategies while engaging in group competition.
Although the three mainstream economics formulations of discrimination are often contrasted, with attempts to empirically establish which of the three is most appropriate to explain different phenomena, all three have in common the study of discrimination based on race as a deviation from the predicted behavior of rational Homo economicus. In practice, neoclassical economics insists on seeing discrimination as a departure from the canonical model. The alternative is to understand racialization as a fundamental system structuring social and economic life in the United States since the country's inception; a system that has been studied and written about extensively outside of and within academia. We can orient our work as economists to this collective effort to understand and transform the world we live in. It is not just that Becker-style models, statistical discrimination, and unconscious bias models are partial understandings of a complex phenomenon; almost all research on race can be described this way. The problem is that there are patterns in this research by economists; the characteristic features of economics research on Latinxs compared with LS are the negation of the role of power and systemic racism, a consistent effort to depoliticize research, and a disciplinary chauvinism that manifests as a disinterest and inattention to work outside of economics that runs counter to neoclassical theory, and this pattern holds among many economists who primarily study racial inequality. These characteristics stand in the way of building oppositional power, they do not help to enable radical politics for social justice. LSE is interdisciplinary by necessity. Many branches of economics are essentially applied mathematics and statistics (Lawson, 2003). In the context relevant to this paper, there is nothing wrong with this except for when the authority and privilege accorded to this type of work is used to comment on racism in a way that contradicts work in related social science and humanistic social science disciplines and when it, in practice, works to support racist ideology and policies.
The stratification economist does not reduce racialization as a social phenomenon to animus-based discrimination, statistical discrimination, or unconscious bias. Instead, racialization is seen as dynamic, evolutionary, and historical. The phenomenon of racialization is misunderstood, and the most effective justice-based responses may not be pursued adequately, when empirical observations clearly linked to racialization are analyzed in a way that systematically excludes knowledge on race produced outside of mainstream economics. 40 Part of how we see this is that SE papers engage explicitly with frameworks for thinking about race or other ascriptive groups and intergroup competition. Many mainstream economics papers on race do not mention the word, “racism.” This is a curious thing. From the perspective of SE and LS, racism is a distinct social phenomenon, and studying racial inequality without reference to any larger structural framework for race is akin to studying prices and or quantities with no reference to market structure in neoclassical economics.
Discussion
In this paper, we articulate a critique of mainstream economics research on Latinxs and provide an alternative framework, Latinx stratification economics (LSE), that draws from both stratification economics (SE) and Latinx studies (LS). While a discussion of each of the subsections can be treated separately—Hispanics as an analytical category, migration theory, assimilation theory, discrimination theory in economics—we argue that identifying the specific methodological issues shows us something about the monistic core of neoclassical economic theory. Taken together, the grand narrative neoclassical economics paints for Latinxs is as follows. People move from one country to another based on human capital. Thus, groups of immigrants are negatively or positively selected. Once they come to the United States, they encounter labor and other markets that function more or less the same for everyone. So, while a group may have, on average, lower education or lower life expectancy than nonimmigrant (white) citizens, over time, these differences will disappear as migrants spend more time in situations facing the same incentive structures as the domestic population (investment in human capital). The main things that could prevent this convergence are differences in preferences (i.e., culture), differences in constraints (i.e., discrimination), or biological differences (i.e., genetics). In other words, barring genetic differences and discrimination, cultural differences may persist for a while, but eventually everyone should behave like Homo economicus. SE is based on a radically different view of society, one that is better supported when compared to ethnographic and other research in LS.
We briefly summarize key themes from the SE and LS literatures; we then explicate LSE by discussing first “Hispanic” as an analytical category and then three major theoretical frameworks from economics that are applied to Latinxs: migration, assimilation, and discrimination. With each in turn we discuss perspectives from LS and SE, how the two approaches complement each other, and how they stand in contrast with the dominant alternative in economics. The results of this work are a rich discussion of how the discipline of economics “sees” Latinxs, based on published research, and a number of emergent findings. We discuss how misguided it is to expect racial/ethnic categories like Hispanic to be consistent over time and space and to correspond reliably with phenotypical characteristics or culture. We argue that a good faith reading of the LS literature would result in the recommendation to subordinate models of migration to models of colonialism and imperialism. We discuss the significance of normative goals and social justice to complement “gap analysis” and endless comparisons to non-Hispanic whites. Lastly, we discuss deficiencies of the dominant models of discrimination and, as an alternative, we highlight rational models of racism that include strategic identifications with whiteness (or brownness), including by members who identify as Latinx or those with Hispanic ancestry. Throughout, we are clear that this critique is a response to patterns in published economics research, not to individual works or the collected works of any one economist. Indeed, taken individually, much of the empirical research by economists that we discuss could be informative from a LS perspective, but the limited or lack of engagement with LS on the part of economists who study Latinxs results in interpretations of this research that is not robust to multiple perspectives.
It is clear that this paper is just a starting point. Another way to articulate the research question is: “What are ways to do LS as an economist?” That is, how do we engage with the work in LS authentically rather than drawing on select ideas or evidence that can be incorporated in a neoclassical model? The intellectual contribution of this project, of which this paper is a start, is focused on the deeper methodological and epistemological questions that characterize the disciplines in question, not superficial differences. And this project is leading to emergent findings; for example, despite many differences in style, part of what leads LS and SE to be complementary and what distinguishes these approaches from the dominant approaches to intergroup inequality (especially that based on race) in economics is that both LS and SE consciously enable social justice politics through the knowledge production process, leading to greater counterhegemonic possibilities for racial justice.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
