The 2020 US Census unfolded against a contentious national political backdrop marked by protests seeking racial justice, unstable immigration policies, partisan rancor, and growing distrust in government institutions. Under these conditions, the power and potential long-term effects of upcoming battles over redistricting loom large. At the same time, within the halls of almost every American state legislature where these debates will occur, there is scant descriptive evidence that large-scale population changes have occurred since the late 1990s. The stagnant character of the race and gender composition of most state legislatures runs counter to widely espoused ideals of political access, and it undermines the legitimacy of these bodies as democratic institutions.

Many researchers assert that the main shortfall producing underrepresentation is at the candidacy stage—women and people of color are competitive candidates, but too few throw their hat into the ring (Lawless Reference Lawless2015). However, studies of female and racial-minority candidates often are animated by two assumptions that tend to speak past each other. On the one hand, the literature on women in politics often focuses on individual- and intimate-level reasons why women want to run for office less often than men, including attitudes toward officeholding and close personal relationships (Carroll and Sanbonmatsu Reference Carroll and Sanbonmatsu2013; Fox and Lawless Reference Fox and Lawless2014; Lawless and Fox Reference Lawless and Fox2005). On the other hand, race and politics scholars emphasize the importance of district racial composition, including majority-minority districts (MMDs), in facilitating minority candidacy and success (Barreto, Segura, and Woods Reference Barreto, Segura and Woods2004; Branton Reference Branton2009; Juenke Reference Juenke2014). Scholars of Black women and Latinas in politics have long asserted that these types of approaches treat women and minorities as parallel social groups and fail to account for the ways in which race and gender simultaneously shape candidacy (Hardy-Fanta et al. Reference Hardy-Fanta, Lien, Pinderhughes and Sierra2016; Smooth Reference Smooth2006; Takash Reference Takash1993). To more precisely explain variations in ballot presence across groups, the multiple levels of context that inform potential candidates’ decisions—individual motivations, domestic arrangements, membership in marginalized groups, local political and social networks, and broader political opportunities in states—must be more coherently integrated.

My book, Nowhere to Run: Race, Gender, and Immigration in American Elections (Phillips Reference Phillips2021), advances an intersectional account for why descriptive representation in state legislatures has not kept pace with changes in the American population. It focuses on members of the two fastest growing racial groups in the United States: Latinas, Latinos, and Asian American women and men. Using an original dataset encompassing every state legislative general election for almost two decades, as well as new interview and survey data from 42 states, I demonstrate that factors in candidate emergence long treated by political scientists as exclusively “racial” or “gendered,” are shaped, in fact, by race and gender simultaneously. To illustrate, I find that increases in a minority group’s proportion of a district population are much more robustly related to the election of men from that group than women. This has direct bearing on debates about the utility of MMDs as a tool for expanding descriptive representation. Prior scholarship and discussions by elites often hinged on the (often-unstated) assumption that the mechanisms driving the increased likelihood of racial descriptive representation are the same for minority women and men (Juenke Reference Juenke2014; Shah Reference Shah2014). My research suggests that debates around expanding representation must move beyond reliance on MMDs as a standalone measure. Majority-minority districts are just one tool in a portfolio of possible mechanisms and institutions that bring the race and gender composition of legislatures closer to that of the women and men that they serve.

By understanding state legislative districts as different types of electoral opportunities, I can investigate how these group-level dynamics are integrated with personal decision-making processes for potential candidates. Survey and interview data with Asian American women and men and Latinas and Latinos reveal that individual-level concerns (e.g., the lack of ambition and the impact of public service on close relationships) do not fully explain the underrepresentation of women across racial groups on the ballot. On average, women do not necessarily have lower levels of ambition than men, and intimate relationships are highly salient in candidacy decisions among both genders. Among Asian Americans and Latinas and Latinos, these issues are but a narrow slice of a larger set of social and institutional constraints that push women away from the candidate pipeline. These constraints include recognition among political elites of color; a sense of obligation to represent racial, gender, and immigrant communities; and group-level costs of trading professional success for public life.

Nowhere to Run makes sense of the political processes driving these results by arguing that state legislative elections are opportunities for descriptive representation that are shaped by two simultaneous processes. At the national level, the distribution of majority-white populations across most districts sharply constrains the number of competitive opportunities for nonwhite women and men to appear on the ballot. In majority-white districts, white men and white women who are thinking about running for office are unfettered by a consideration that potential candidates of color must grapple with—that is, appealing to a majority of constituents who do not share their racial background. There currently are 41 states where more than two thirds of state legislative districts are majority white. This includes 19 states where more than 90% of state legislative districts are majority white; this subset alone encompasses almost 40% of all state legislative seats in the United States.Footnote ¹

At the local level, the scarcity of “minority” seats is a constraint shared by women and men of color. However, that condition also exacerbates race-gendered processes of “secondary marginalization” (Cohen Reference Cohen1999; Strolovitch Reference Strolovitch2006) among political elites because men of color tend to dominate the informal groups and networks that plan and negotiate to maintain or win the one or two “Latino” or “Asian” or “Black” seats in a state or metropolitan area. Secondary marginalization describes processes within communities that are excluded from mainstream politics, whereby the political activities and leadership of members of a dominant subgroup render multiply disadvantaged subgroups politically invisible (Cohen Reference Cohen1999). I show that Latinas and Asian American women often struggle to be recognized as viable candidates by political elites in these networks. As a consequence, their ability to leverage electoral resources that are concomitant with a sizable minority population—and often necessary to make an opportunity realistic—tends to be less robust than that of co-racial men.

These national and local constraints are overlapping and interactive, resulting in systemic absences of opportunities for descriptive representation for certain groups, most acutely Latinas and Asian American women. Importantly, my book emphasizes that this dearth of representation opportunities is not occurring in a vacuum. Rather, it is driven in part by the abundance of electoral opportunities facing white men in particular because they comprise most of the incumbents and are relatively unrestricted by race in their access to realistic district opportunities.

In another project with Paru Shah, we are applying a similar intersectional approach to a series of analyses of the substantive representation of immigrant communities by legislators who are members of those groups. As such, we are studying the representation practices of these legislators as processes of immigrant incorporation. Immigrant incorporation typically has been a topic for scholars of mass publics; however, we contend that using this framework to understand the behavior of elites in legislatures allows a more accurate picture of the race-gendered (Hawkesworth Reference Hawkesworth2003) dynamics and power relationships across and within groups to emerge.

During 2019, we conducted 44 in-depth interviews with Asian American women and men and Latina and Latino state legislators at national gatherings and assembled a new database of almost four million observations of state legislative bill sponsorship. Our preliminary analyses of these data indicate that similarities between women and men in these racial groups on topics contained in sponsored bills may be obscuring as much as they reveal. The qualitative data we gathered point to distinctions in legislators’ understandings of the underlying issues driving policy choices and “race-gendered” (Brown Reference Brown2014; Hawkesworth Reference Hawkesworth2003; Smooth Reference Smooth2006) inequalities in access to legislative processes.

I mention this second project because across both parts of my scholarship on representation in state legislatures, I find that the women and men who carry out this work are constantly facing a complex mix of questions about their opportunities and limitations. Is there a real chance I can win where I live? Is the legislature a place where a person like me can actually get important and urgent things done? Particularly for the Latinas and Asian American women I interviewed, the latter question is daunting. Many view themselves as representatives who are embedded in their community and who must make the most of the rare opportunity to have someone “in the room” who looks like them and has lived as they have lived.

Against the backdrop of powerful mass political movements in recent years—immigrant-rights actions, Black Lives Matter protests, #MeToo activism, and others—a salient question for scholars of representation is whether women of color who are passionate about these issues will have less reason to try to advance their work through legislative officeholding. As one Latina legislator I interviewed stated, state legislatures “were built for other people” and have been slow to change.

Perhaps the most significant regular opportunity to enact change in legislatures is close at hand: US Census–based redistricting. Our understanding of the consequences of these district-drawing processes on representation must move beyond single-dimensional identity categories. Instead, I propose that we strengthen the study of representation by using frameworks centered on simple concepts that are complex in their ramifications for democratic processes: that is, individuals are simultaneously members of more than one social group, and their opportunities for political leadership are shaped by processes and institutions both large and small.