Legislatures should reflect the diversity of the constituents they represent. In the United States, women and many racial minority groups lack fair representation at all levels of government. Similarly, independent and third party voters are often left without representation. While deep-rooted social factors have long stood in the way of reflective representation, our electoral structures also play a significant part in determining who is elected to office. And while social change is a long and slow process, reforming our electoral structures will have significant and immediate impact on who runs, wins, and serves in office.
Legislative bodies should reflect the diversity of the electorate they seek to govern. When the broad spectrum of political views, personal characteristics and experiences, and interests of Americans are not reflected in our elected institutions, groups in the minority lose their voice. Furthermore, governing bodies that do not reflect the diversity of the general public do not benefit from the full breadth of views, insights and talents of the community.
Electoral systems—the rules and structures that govern the way votes are cast and counted—have a large impact on the quality of representation.
This page summarizes findings about electoral systems and representation.
Currently, most federal, state, and local governing branches are elected using electoral systems that fail to produce quality representation because they are based on the "winner-take-all" principle. In a winner-take-all system, the candidate with the most votes wins, regardless of their vote share, and no other candidates are elected.
The historical use of winner-take-all in multi-winner districts to disenfranchise African-Americans groups notwithstanding, winner-take-all elections can produce quality representation in homogeneous single-member districts that consist of one overwhelmingly dominant political group—whether an ideological group or a racial minority group. In 1967, Congress even mandated that members of the House of Representatives be elected from single-winner districts, with the intention of ensuring minority groups, who mostly lived in concentrated urban communities, could gain descriptive representation.
However, single-winner districts can produce skewed representation in diverse, heterogeneous societies. In diverse places, such as the United States, there are many groups in a district that need representation and only one person to represent everyone in a single-winner district.
For more information, see:
One potential improvement to winner-take-all without abandoning single-winner districts is to use the single-winner form of ranked choice voting (RCV). In single-winner RCV, voters rank candidates in order of choice. Initially, every ballot counts as a vote for its highest ranked candidate. Then, the weakest performing candidates are eliminated one-by-one, and ballots counting for them are added to the totals of their highest ranked candidate still in the race.
Single-winner RCV is used in many localities across the United States, including four cities in the San Francisco Bay Area in California, Minneapolis and St. Paul, Minnesota, and Portland, Maine. Because American cities began turning to RCV only in the early 2000s, there jury is still out on how single-winner RCV improves the quality of representation. To investigate the impact of single-winner RCV, FairVote is currently undertaking an ambitious project examining the impact of single-member RCV on the election of women and people of color in the Bay Area in California.
Proportional representation describes multi-winner systems in which winners are allocated in proportion to their vote shares.
In European models, voters often vote for political parties rather than individual candidates. In a closed list system, political parties choose their candidates and the order in which they will be elected, while voters vote for a political party. In an open list system, political parties still choose their candidates but voters influence the order in which those candidates are elected by voting for the candidate from their party they most prefer.
Although European party-centric proportional representation is typically seen as out of step with American political traditions, there is a substantial history of candidate-based forms of proportional representation in the U.S. These systems have increased the quality of representation where they have been used. Systems currently in use include:
Cumulative Voting (CV) is a semi-proportional method of election in which voters have a number of votes equal to the number of seats to be elected. Voters can assign as many of their votes to a particular candidate or candidates as they wish. In a three-seat district, for example, a voter could give all three of their votes to one candidate, two votes to one candidate and one to another, or one vote to three different candidates. CV is used in over seventy different localities in the United States, mostly in southern states like Alabama and Texas as part of settlements in Voting Rights Act cases.
In Limited Voting (LV), voters have fewer votes than there are seats to be elected and may cast only one vote for each candidate. This system allows a majority group to control the majority of seats, but not all seats. The greater the difference between the number of seats and the number of votes, the greater the opportunities for proportional type representation. Versions of limited voting are used in Washington, D.C., Philadelphia (PA), Hartford (CT) and many jurisdictions across North Carolina and Alabama. Like Cumulative Voting, Limited Voting has been used successfully to resolve several Voting Rights Act cases.
Single Vote (SV), also know as single non-transferable vote (SNTV), is a form of limited voting. In SV, voters have one vote in a multi-winner district election. Counting is similar to plurality models of voting, where the top vote recipients win seats. However, the model is semi-proportional because the most popular candidates earn seats but, like LV, a majority block of voters can not win all the seats. While this does not ensure highly proportional representation, it prevents minority groups from being locked out from the system.
Studies by Greg Adams (1996), S. Bowler, T. Donovan, and D. Brockington (2003), Richard Engstrom (2011), and Charles Wiggins and Janice Petty (1979), to name a few, have found that cumulative, limited and single voting enhances representation of not only racial and ethnic minority groups and women, but also political or ideological minority groups as well. The drawback, however, is that for minority groups to gain representation under CV or LV, there needs to be a sizable population of that group and the group must vote coherently so that their votes count towards electing the same candidate(s). Another form of proportional type representation that requires less strategic voting by minority populations is Ranked Choice Voting in multi-seat districts.
In Ranked Choice Voting in Multi-Winner Districts, also known as the Single Transferable Vote, voters have one vote but are able to rank candidates in order of preference. Initially, every ballot counts as a vote for its highest ranked candidate. Those candidates who have enough votes to win are elected and the weakest performing candidates are eliminated. For instance, in a five-seat district, a candidate is elected if they receive more than 1/6 of all votes cast, as this threshold ensures that they will be one of the top five finishers. If not enough candidates as number of seats reach the threshold to win, then voters' second choices come into play.
Multi-member RCV is currently used for state elections in Malta, Ireland, and Australia. Domestically, however, it is only used in Cambridge, Massachusetts (and has been for forty years). Work by S. Bowler and B. Grofman suggests that multi-winner Ranked Choice Voting should increase the quality of representation by increasing the proportion of women and people of color elected to public office. Work by other scholars including Douglas Amy (2002) Kerstin Barkman (1995), and Susan Welch and Donley Studlar (1990) also suggests RCV in multi-member districts should increase descriptive representation.
Women are over 50% of the population of the United States but make up only about 20% of Congress. Only one state, New Hampshire, has reached gender parity in elected office, and no states have reached parity in both chambers of their state legislature. Women have served in the U.S. Senate in 27 only states. These statistics reflect the fact that women have a diluted presence in government.
infogram_0_women_in_the_senateWomen in the Senate//e.infogr.am/js/embed.js?pvVtext/javascript
Based on the percentage of women in the House of Representatives (19.3%), the Inter-Parliamentary Union ranks the United States 94th out of 188 countries in the world for representation of women (as of November 2015). Countries ahead of the United States include Rwanda, Cuba, South Africa, Vietnam, Pakistan, China and Bosnia.
infogram_0_world_rankings_of_womens_representationWorld Rankings of Women's Representation//e.infogr.am/js/embed.js?OOltext/javascript
Many women in government, advocacy groups, and scholars observe that the number of women in office needs to increase. While many look at social reforms to address gender disparity, FairVote's work examines the structural barriers to representation, specifically single-winner electoral systems.
Studies by Douglas Amy (2002), Jennifer Hayes Clark and Veronica Caro (2013), James King (2002), Richard Matland and Deborah Brown (1992), Michael Minta (2012), Heather Ondercin and Susan Welch (2009), Lilliard Richardson, Brain Russell, et al (2004), and Jessica Troustine (2008), to name a few, find that winner-take-all elections in single-winner districts are a barrier to women in office.
Numerous comparative studies, by American and international scholars, conclude that women are best represented in proportional systems with multi-winner districts (Welch (1990), Studlar and Welch (1996), Welch, Clark, and Darcy (1985), Zimmerman (1994), Kaminsky and White (2007)).
These findings hold even when we take into account the cultural expectations about the role of women. So while ingrained social attitudes might be a significant barrier to gender parity, our current electoral system also plays a large role. The bottom line is that use of single-winner districts and winner-take-all have allowed little opportunity for women to do more than squeeze in at the edges.
In their studies of U.S. jurisdictions, Matland and Brown (1992) and Hughes (2013) find multi-winner districts create better opportunity for women’s representation over single-member districts.
Numerous studies reach similar conclusions. For example:
Multi-winner districts increase women's representation for two key reasons: voters tend to balance their tickets; and political parties seek to appeal to as many voters as they can. Amy (2002), Zimmerman (1994), and Troustine (2008) find that in the multi-winner environment voters are more likely to vote for male and female candidates to balance their choices. Therefore, in multi-winner systems parties have greater incentives to run more female candidates. As much of the scholarship finds, parties run more female candidates in multi-winner systems because parties diversify their candidates to appeal to more voters (Barkman, 1995). This leads to more recruitment of female candidates and, consequently, more women in elective office.
Moving from single-winner, winner-take-all districts to multi-member proportional representation models has the best chance of increasing women’s representation (King, 2002). Of all the systems, party-list types of proportional representation, common in Europe, tend to be the best for women's representation.
While the U.S. is unlikely to adopt European model of proportional representation, it has a long history of using candidate-centered forms of proportional representation like cumulative voting districts or ranked choice voting in multi-winner districts. In particular, ranked choice voting in multi-winner districts should increase the number of women elected. The international use of multi-winner RCV provides clues as to its likely impact in the United States. The Australian Senate, which is elected using multi-winner RCV, has relatively high proportions of women members as does Malta's legislature (also elected using multi-winner RCV). In Ireland, however, the Dail Eireann, elected using multi-winner RCV, has below average proportions of women. But many scholars posit that Ireland's lack of parity has more to do with its conservative political culture and recruitment of women in local governments than electoral mechanisms (Bowler, S. and B. Grofman (2000), Buckley, Fiona, Mack Mariana, et al (2015), McElroy, Gail and Michael Marsh (2010)).
In the United States, multi-winner RCV is only used in Cambridge, Massachusetts. FairVote is currently working to further examine the effect of ranked choice voting, in both its multi- and single-winner forms, on representation.
Based on the percentage of women in the House of Representatives (19.3%), the Inter-Parliamentary Union ranks the United States 94th out of 188 countries in the world for representation of women (as of November 2015). Countries ahead of the United States include Rwanda, Cuba, South Africa, Vietnam, Pakistan, China and Bosnia.
Many racial minority groups are severely underrepresented in local, state, and federal government. All discussions of improving descriptive racial minority representation in the United States are guided by the Voting Rights Act of 1965, and a later 1967 law that legally embeds the notion that single-winner districts are better for minority representation than multi-winner districts, especially block voting (Carroll and Sanbonmatsu, 2013).
Contemporary scholarship supports the idea that minorities are most fairly represented in well-drawn single-winner districts only when the minority population in question is politically homogeneous and geographically concentrated. In recognition of these problems, inherent in single-winner districts, judges increasingly uphold multi-winner proportional electoral systems, like Cumulative Voting and Limited Voting, as legitimate alternatives to majority-minority single-member districts under the Voting Rights Act of 1965.
The literature shows:
Because multi-winner RCV is not yet widely used in the United States, and single-winner RCV has gained popularity recently, we do not yet know much about the impact of RCV on minority representation.
There is good reason to expect that multi-winner RCV will fairly represent ethnic and racial minorities, for the same reasons other proportional systems do (Amy, 2002).
Early, anecdotal, evidence shows that RCV in single-winner districts has been accompanied by high levels of representation for ethnic and racial minorities. Currently (2015), three of the four mayors of the Bay Area cities using ranked choice voting in their elections are female. Women hold half or more of the offices elected by RCV in the Bay Area in three cities: Oakland, Berkeley and San Leandro. Women and people of color hold 47 of the 52 elected offices filled using RCV.
To explore the link between RCV and the representation of ethnic and racial minorities FairVote has launched two ambitious projects:
We anticipate completion of these projects in 2016.
The effects of alternative electoral systems such as Ranked Choice Voting on third-party and independent representation in the United States is a major topic of FairVote's research. Although there are examples from abroad (for instance, the multi-winner use of RCV in the Australian Senate is thought to slightly increase the representation of minor parties (Bowler and Grofman, 2000), the unique two-party, candidate-centered system of the US makes extrapolating the overseas experience difficult.
While RCV might not increase the election of third party or independent candidates in the US, as candidates must clear a threshold of the total vote (a threshold that decreases the more candidates that are being elected), RCV allows supporters of third parties and minor candidates to sincerely rank their preferred candidate first without feeling like their votes are wasted (Lewyn, Michael (2012)). RCV also minimizes the spoiler effect of third party votes. Under RCV, third party and independent supporters can rank their most preferred party or candidate first. If their candidate is eliminated, their votes immediately transfer to their second choice preference. Third parties supporters are thus free to elect their favorite candidate with minimal chance that that support will spoil the election outcome (Bartholdi III and Orlin,1991, Bowler and Grofman, 2000).
Along with preventing spoilers, RCV may also help keep the major parties more accountable to the electorate. Candidates are incentivized to court a broader range of voters than they normally would, including asking for second and third choice rankings from minor-party supporters.
In multi-winner RCV, it also becomes possible for Democrats or Republicans that live in a district with the opposite majority to gain representation. As long as the Democrat or Republican population is equal to or greater than the threshold to win, people can gain representation where they currently feel left out.
We need your support. Help us spread the love by pitching in $5 today!