UC Berkeley

The United States’ population is rapidly changing, but the ways in which political scientists measure and understand representation have not kept apace. Marginal shifts in descriptive representation over the past two decades have run counter to widely espoused ideals regarding political accessibility and democratic competition. A central assumption often made by academics, and the public, has been that groups which are otherwise disadvantaged in politics may leverage their communities’ numerical size as a political resource to gain influence. To this end, many studies of racial descriptive representation find that a larger minority population is associated with a higher likelihood of a racial minority running for and/or winning. However, these positive relationships between population growth and descriptive representation are tempered by an extensive literature documenting limits on racial minority groups’ political incorporation. Moreover, current frameworks for understanding group competition or patterns of descriptive representation are silent about whether shifts in racial demographics may also have an effect on the balance of representation between women and men.

These contradictions in debates over representation, and how groups gain influence, undermine the notion that eventually, marginalized groups will be fully incorporated into politics. White women have had de jure access to the voting franchise in the United States since 1920. In the intervening period, women have made up approximately half the population, and outnumbered male voters in every presidential election since 1964. Yet, women have held a quarter or less of all state legislative seats across the country for well over two decades, and only reached 100 members of Congress in 2014.

The case for eventual incorporation is similarly dubious when we consider the racial composition of elected bodies. The racial balance of American communities is in flux largely due to Asian and Latina/o immigration, which will continue to be the case into the forecastable future. Presently, Asian Americans and Latina/os make up 23 percent of the U.S. population and are the two fastest growing racial groups in the country. Members of these immigrant communities hold less than ten percent of all state legislative seats, and a similar fraction of seats in the 115th Congress. Taken together, these yawning gaps between presence in the population and representation in elected office strongly suggest that “time” alone may be an insufficient remedy for underrepresentation.

Moreover, for those who are living in the United States now, the current demographic makeup of state legislatures—which includes over 7500 elected seats nationally—raises doubts about their representative legitimacy. Asian American and Latina/o women and men typically have socioeconomic experiences, political perspectives and policy priorities that are distinct from that of their most likely descriptive representative—a White man. At the same time, state legislatures have been veritable policy engines for bills and resolutions related to immigration and immigrants in recent years. The National Council of State Legislatures reports that in 2015 state legislatures enacted 216 laws and passed 274 resolutions related to immigrants and immigration. Even as these bodies write, debate, and pass legislation targeting immigrant communities, Asian Americans and Latina/os are rarely in the room. Researchers increasingly point to the scarcity of female or racial minority candidates as a key explanatory factor, but seldom examine issues related to race and gender at the same time. As a result, the extant scholarship obscures the outsized effects that White men’s candidacies have in defining American elections, and overlooks the distinct challenges that women of color face in getting on the ballot.

This dissertation examines the intersecting roles of race and gender in elections, with particular attention to how they may be changing as immigrant communities become a larger proportion of the American population. I analyze the Gender Race and Communities in Elections dataset, which encompasses all state legislative general election winners and candidates from 1996-2015, and includes demographic information for candidates and their district populations. This original dataset provides the first opportunity to simultaneously analyze descriptive representation in state legislatures, for women and men in the four largest racial groups, at the national level. I also present the results of a national survey of state legislators, and in-depth interviews with political elites, in order to reveal race-gendered, informal, processes of candidate development and deterrence.

I also show that practical opportunities to compete in elections are sharply, and simultaneously, constrained by candidates’ race and gender. These constraints are most evident in the lopsided distribution of racial populations across districts, the uneven candidate development efforts of civic and political organizations, and the dominance of men in elite political networks, across racial groups.

Based on my examination of Asian American and Latina/o candidates in elections, I advance a Race-Gendered Model for understanding the persistence of underrepresentation in state legislatures. I conceptualize elections as competitions for descriptive representation, and account for the disparate social and political experiences of women and men from different racial groups. Within this framework, race and gender simultaneously constrain potential candidates’ access to elections, producing a frequent absence of competition for descriptive representation. This model uses an intersectional approach to explain why Asian American and Latina/o women and men do not run more often, and why the majority of ballots are made up exclusively of White male candidates.

I demonstrate that the increasing “strength in numbers” of Asian American and Latina/o communities has primarily served as a resource for increasing the racial diversity of men in statehouses—to the limited extent that racial diversity has increased at all. I also show that the most advantaged descriptive group, White men, benefits from an absence of competition in most electoral contests. At the same time, the fastest growing groups of women—Asian Americans and Latinas—are also the groups most frequently excluded from competing.

The Race-Gendered Model expands the intellectual terrain available to answer longstanding questions in the study of women and racial minorities’ underrepresentation. Along the way, I argue that it is necessary to simultaneously consider why White men’s overrepresentation is similarly persistent. More broadly, the theory of competition presented in this dissertation shifts away from a central focus on the advantages and disadvantages groups face during election campaigns. Instead, I argue that the choices voters face in electing a descriptive representative are limited long before election day.