Power at the Polls

Service included, for the first time, a gay/lesbian/bisexual self-identifier question on its presidential exit poll. Based on a national sample of 15,488 voters, the VNS polling data offers the first statistically reliable portrait of self-identified gay/lesbian/bisexual voters in relation to the overall voting population.<br><br> 1 Analysis of this data leads to the following four major conclusions: 1) Gay/lesbian/bisexual self-identified voters constituted more than three percent of the total voting population in the 1992 elections. Because of the fears that keep gay people in the closet, and the consequent constraints upon self-identification in a public setting, this figure must be considered a floor rather than a ceiling. In other words, the total number of these voters is certain to be higher than three percent, and likely to be significantly higher.<br><br> 2) The gay/lesbian/bisexual vote can be decisive in highly urbanized states. This vote is not evenly distributed throughout the population. Rather, it is heavily concentrated in urban areas, with self-identified voters constituting more than eight percent of the vote in medium-sized cities.<br><br> This means that it can play a critical role in key states whose electoral votes are essential for victory. 3) The gay/lesbian/bisexual vote is a younger vote, and is growing in size. Reflecting the changing social climate of the last generation, younger voters are significantly more likely to self- identify than are older voters.<br><br> They already constitute about five percent of the under-30 age group. And, as the issue of homosexuality continues to have high salience in national and state politics, more gay, lesbian, and bisexual Americans are likely to be mobilized and to self-identify. 1 2 4) The gay/lesbian/bisexual vote displays a clear political profile.<br><br> It is heavily Democratic in its affiliation, leans clearly toward liberal positions on issues such as health care, taxes and active government, and is strongly swayed by a candidate 9s position on so-called csocial issues, d such as abortion and, of course, gay rights. In sum, there is a gay/lesbian/bisexual vote with a clear political profile, concentrated in high population urbanized states, that can be mobilized in even greater numbers than it has in the past. In light of this evidence, how it votes in 1996 can make a critical difference in the outcome of the Presidential election.<br><br> The kind of difference it makes will depend on how each of the major parties responds to the needs and aspirations of the gay, lesbian and bisexual community. Introduction The 1992 elections witnessed the emergence of homosexuality as a major issue in American politics. The Democratic nominee, Bill Clinton, openly courted the gay, lesbian and bisexual community, seeking votes and financial support, and promising action on issues important to this constituency.<br><br> Meanwhile, the Republican national convention saw a major rhetorical assault against gay, lesbian, and bisexual Americans, and the Republican party incorporated explicitly anti-gay planks into its platform. Statewide anti-gay ballot initiatives in Colorado and Oregon focused even more attention on the issue. Since 1992, the issue of homosexuality has continued to receive significant attention.<br><br> The national debate over the military exclusion policy, the Supreme Court consideration of the Colorado Amendment 2 case, ballot initiatives in other states and cities, the debate over same-gender marriage sparked by a Hawaii Supreme Court case, and an unprecedented level of state legislative activity on gay, lesbian, and bisexual issues, have kept homosexuality in the political spotlight. With the 1996 election campaign well underway, and with candidates and parties seeking the support of critical voting blocs, questions about the size, distribution, and loyalties of the gay, lesbian and bisexual vote assume new importance. Because gays, lesbians and bisexuals cannot be easily identified 4as can groups based on gender, race, and religious affiliation 4their role in the political process has been understudied and, too often, dismissed.<br><br> This background paper seeks to remedy this problem. Because gays, lesbians and bisexuals cannot be easily identified, their role in the political process has been understudied and, too often, dismissed 3 A New 4and Still Hidden 4Constituency Over the last generation, the lesbian and gay community has become increasingly visible and well organized. Not only have many individuals come out of the closet, but they have built organizations, fought to change laws and public policy, run for office, and participated actively in electoral politics.<br><br> In a democratic society, electoral politics is one critically important means for citizens to express their wishes and work to shape laws and public policies that address their needs. The importance of voting, both symbolic and real, has been demonstrated by the long hard-fought campaigns in the 19th and 20th centuries of the propertyless, of African Americans, and of women to have access to the franchise. Since 1972, when gay activists first addressed the national convention of the Democratic party, lesbian and gay Americans have played a visible role in national elections.<br><br> They have formed Democratic and Republican clubs, have testified at platform hearings, have served as delegates at the conventions of both major parties, have participated actively in the campaigns of presidential and congressional candidates, and have run successfully for office themselves. The political mobilization of the gay, lesbian and bisexual communities reached a qualitatively new level in the 1992 national elections. For the first time, they were not simply struggling to be heard.<br><br> Rather, they and their issues were spotlighted throughout the campaign. Bill Clinton, the Democratic nominee, met with members of the community, promised to take action on key issues such as the military exclusion policy and AIDS, and incorporated gay, lesbian and bisexual supporters into his campaign staff. Meanwhile, at the Republican National Convention, prominent Republicans such as Pat Buchanan targeted the gay community as symptomatic of a decline in values in America.<br><br> The platform included explicitly anti-gay provisions on issues such as gays in the military, same-gender marriage, civil rights protections, and AIDS education. Throughout the campaign, Republican candidates made homosexuality an issue. The ability to demonstrate cvoting power d is a key road toward winning the attention of political parties and their candidates.<br><br> Many groups in the population are able to make reliable claims about their total numbers and about their voting strength because they are readily identifiable. But gay, lesbian or bisexual identity is not immediately apparent, and gays, lesbians, and bisexuals have a long history of remaining cin the closet d as a way of escaping the penalties attached to homosexuality. Despite the growth of a movement over the past 25 years, and despite the progress toward public acceptance, the large majority of gays, lesbians, and bisexuals remain in the closet.<br><br> Thus, until recently it has been difficult to offer any kind of count of the size of this voting bloc. 4 The 1992 Exit Poll: The First Reliable National Sample of a Gay, Lesbian, and Bisexual Vote Since the late 1980s, various news organizations have made some effort to enumerate gay, lesbian and bisexual voters. In 1988, for instance, ABC News collected such information during the New York and California Democratic Presidential primaries.<br><br> There are also some state election exit polls (New York, New Jersey, and California), and samples from mayoral elections in New York City, Los Angeles, and San Francisco, cities in which the gay and lesbian community is particularly visible and well organized. In 1992, Voter News Service, a well-respected polling organization, conducted a Presidential Election exit poll on behalf of the major television networks. For the first time in a national presidential exit poll, VNS provided a question that allowed respondents to self-identify as gay, lesbian, or bisexual.<br><br> Sensitive to the reaction that voters might have to such a question, VNS placed it at the end of the poll, in order to deter as few people as possible from completing the questionnaire. The exit poll conducted interviews with 15,488 voters, selected to provide a statistically accurate portrait of the voting population. Of those who answered the self-identifier question, 3.2 percent identified themselves as gay, lesbian, or bisexual.<br><br> This is not an insignificant figure. To place it in comparative perspective, it is approximately the size of the national Latino vote in 1992; more than twice as 5 large as the Asian-American vote; slightly less than the Jewish vote; and larger than the family farm vote. It also cannot be emphasized strongly enough that the 3.2 percent figure obtained from the 1992 presidential exit poll constitutes a floor, not a ceiling, and a basement floor at that.<br><br> It represents the percentage of voters willing to reveal a stigmatized identity in a public setting. Given the penalties that have historically attached to homosexuality, and the extreme fear of self-disclosure that still shapes the lives of the majority of gays, lesbians and bisexuals, one can confidently assume that the total vote is higher than 3.2 percent, and probably significantly higher at that. Who Are These Voters: Some General Demographics The VNS exit poll provides an opportunity to draw a demographic profile of self- identified gay, lesbian and bisexual voters, and to compare them to the total population of voters.<br><br> The portrait obtained from the VNS poll of 15,488 voters points to both similarities and differences between them and other voters. A majority of gay, lesbian and bisexual voters were male [55 percent] whereas a majority of all voters were female [53 percent]. The educational levels of gay and bisexual men were roughly comparable to those of all male voters, while those of lesbian and bisexual women were somewhat higher than those of all female voters.<br><br> Income levels reveal a different pattern. Among both men and women, family income was lower for gay, lesbian and bisexual voters than for all voters. The total size of the sample in the VNS exit poll does not permit drawing reliable conclusions about race and ethnicity for the population as a whole.<br><br> But among urban voters drawn from the VNS poll for cities of 50,000 or more, the proportion of African Americans and Latinos who self-identified as gay, lesbian, or bisexual [17.3 percent and 5.1 percent, respectively] was virtually equivalent to the proportion of African Americans and Latinos [16.2 percent and 5.6 percent, respectively] in the sample as a whole. Thus, the gay, lesbian, and bisexual vote in at least some respects seems to mirror the ethnic and racial diversity of the voting population. The 3.2 percent figure obtained from the 1992 presidential exit poll constitutes a floor, not a ceiling, and a basement floor at that 6 Who Are These Voters: Two Significant Characteristics In two important ways, the group of self-identified gay, lesbian, and bisexual voters differs from the total population of voters.<br><br> They are geographically concentrated in urban areas and they are concentrated heavily among the young. An Urban Vote Gay and lesbian activists have long proclaimed cwe are everywhere. d In one important sense this statement is incontrovertible: every reliable study has found that gays, lesbians, and bisexuals are raised in every religious grouping; they are born into families of every race and ethnicity; they come from families of every income and educational level; and they come of age in every part of the United States. But a more accurate rendering of the statement would read cwe come from everywhere, but we do not remain everywhere in equal numbers. d As gays, lesbians, and bisexuals have come to an awareness of their identity, they have often chosen to migrate to escape the disapproval of family, neighbors, and friends, and to find safety in numbers.<br><br> Historically, this has meant that they often leave small towns and rural areas, and that visible communities have tended to coalesce in larger cities. Though rural and small-town gays, lesbians, and bisexuals are now organizing, it nonetheless remains true that the most visible, highly organized, and numerically concentrated communities are to be found in large urban areas, which provide both anonymity and the safety of greater numbers. The VNS polling data dramatically bear this out: 7 "An analysis of sample precincts in towns ranging in population from 5,000 to 10,000 reveals a self-identification rate of 1.3 percent for men, and 1.2 percent for women.<br><br> But in cities between 250,000 and 500,000, the figure rises to 8.3 percent for men and 8.4 percent for women. A vote this large in states that are highly urbanized is of critical significance. The geographical concentration of the gay, lesbian, and bisexual vote means that it can exercise an influence in elections out of proportion to its total numbers.<br><br> A Growing Segment of the Vote The phenomenon of openly gay, lesbian, and bisexual people is, historically, a comparatively recent one. Only since the early 1970s have significant numbers of them chosen to identify publicly and act collectively in the political process. Over the last twenty five years, this process has spread, accelerated, and intensified, changing the social context in which younger Americans come of age and shaping their expectations of how they ought to be able to live.<br><br> This accelerating generational shift is clearly displayed in the self-identification rates of the VNS exit poll: "Whereas only 11 percent of gays, lesbians, and bisexuals were 60 years of age or older, 21 percent of all voters were. "By contrast, 61 percent of gays, lesbians, and bisexuals were under 40, and 30 percent were under 30; the comparable figures The geographical concentration of the gay, lesbian, and bisexual vote means that it can exercise an influence in elections out of proportion to its total numbers 8 for all voters were 44 percent and 20 percent. "For the sample as a whole, self-identified gays, lesbians, and bisexuals represent five percent of the 18-29 age group, a concentration two-thirds higher than in the voting population as a whole.<br><br> These findings are consistent with other available data. In ABC 9s 1988 Democratic presidential primary poll for New York State, the New York City sample revealed 45 percent of the gay and lesbian group to be in their thirties, but only seven percent over sixty. Similarly, in the 1988 California Democratic primary, 46 percent of the gay and lesbian sample were in their thirties.<br><br> When one remembers that the youngest of these voters are at an age when sexual identity is often still in flux, the figures for under-30 voters become even more impressive. The VNS data clearly point to a voting bloc that will continue to grow over time. Who Are These Voters: Some Political Characteristics While all self-identified gay, lesbian and bisexual voters are not alike in their political views, and while they share some characteristics with the pool of all voters, they nonetheless display a distinctive political profile.<br><br> First and foremost among these was their preference for a presidential candidate. In the three-way race for the presidency in 1992, in which Clinton received only 43 percent of the total popular vote, 72 percent of the self- identifying voters chose Clinton. The remaining 28 percent were equally divided between Perot and Bush.<br><br> Only African-American and Jewish voters provided Clinton with a larger margin of support. 9 Gay, lesbian, and bisexual voters also diverged from the general voting population in their level of activity in the electoral process. Data from the VNS exit poll found that gay, lesbian and bisexual voters were three times as likely to wear a campaign button and twice as likely to have been contacted by a campaign.<br><br> When we look at where they stood on issues, some clear differences emerge. While gay, lesbian, and bisexual voters share with all Americans a core concern about jobs and the economy, in certain other areas there is a clear divergence. Health care and access to it ranks higher in the list of concerns of gay voters.<br><br> "While 55 percent of the non-self-identifying sample prefer fewer government services and lower taxes, 60 percent of the gay, lesbian, and bisexual sample prefer more government services and are willing to pay higher taxes to fund them. "Almost 60 percent of the non-self-identifying sample rated deficit reduction as the most important priority for the next president, in comparison to only 36 percent of gay, lesbian, and bisexual voters. "Only 25 percent of non-self-identifiers rated expanded domestic programs as their highest priority, in comparison to 44 percent of gay, lesbian and bisexual voters.<br><br> Gay, lesbian and bisexual voters were three times as likely to wear a campaign button and twice as likely to have been contacted by a campaign 10 Differences are especially prominent in the realm of social issues: "Only 35 percent of non-self-identifying voters believe abortion should always be legal, in comparison with 56 percent of the self-identified gay, lesbian, and bisexual group. "Three-quarters of the non-self-identifiers believe the government should encourage ctraditional values, d whereas 70 percent of gay, lesbian and bisexual voters believe the government should encourage tolerance of diverse values. "When asked whether the main reason for our social problems was the breakdown of the family or government neglect, non- self-identifying voters split almost evenly, but 70 percent of the gay, lesbian and bisexual sample attributed the problem to government neglect.<br><br> The positions on individual issues were reinforced when voters were asked to describe their political ideology. Roughly half of the self-identified gay, lesbian, and bisexual sample characterized themselves as liberal in comparison to less than a quarter of all voters. 70 percent of gay, lesbian and bisexual voters believe the government should encourage tolerance of diverse values 11 Implications and Significance Analysis of the data from the 1992 presidential election exit poll points to a few clear conclusions: "the total gay, lesbian, and bisexual vote is unquestionably larger than the percentage of voters willing to self-identify; "the vote is concentrated in urban areas, and therefore can play a critical role in key states in a presidential election as well as in state and local elections; "the vote is more heavily represented among younger age groups, and therefore can be expected to grow over time; "the gay, lesbian, and bisexual vote displays a clear political profile that warrants the cliberal d label, and that cares deeply not only about cgay rights d issues, but about a wide range of domestic issues as well.<br><br> However, the numbers from 1992 only tell part of the story. They leave unanswered questions about whether a gay, lesbian, and bisexual vote can be effectively mobilized in 1996, what difference it can make, and which candidate or party will win it. Can the Gay, Lesbian, Bisexual Vote Be Mobilized in 1996?<br><br> Blocs of voters will go to the polls to the degree they perceive the political process as relevant to their needs. Since the last presidential election, when gay issues figured prominently in the rhetoric, platforms, and campaigns of the major candidates, issues affecting the gay, lesbian and bisexual communities continue to be a focus of attention in national, state, and local politics. During the first six months of the Clinton presidency, the debate over gays in the military was headline-making news, and constituted the first time that a gay issue dominated national politics for any length of time.<br><br> The Supreme Court hearing of the Colorado Amendment 2 case [the anti-gay civil rights ballot initiative] in October 1995, and the awaited decision on the case, has highlighted the importance that the Court, as well as the power to appoint new justices, holds for the gay community. In addition, the years since 1992 have seen other statewide and municipal ballot initiatives on gay rights, along with a growing 4 and unprecedented 4volume of legislative activity at the state level. Finally, the debate over same-gender marriage, which has erupted in the last few months and promises to grow more intense as the year goes on, is keeping gay issues in the public eye.<br><br> 12 Meanwhile, the early rounds of Republican caucuses and primaries have injected gay, lesbian, and bisexual issues into this year 9s electioneering. Pat Buchanan, who has promised to continue his campaign all the way to the Republican convention in San Diego, has frequently targeted the gay community as symptomatic of a decline in values in the United States. Republicans who are rooted in the extremist religious right have made opposition to gay rights and to same-gender marriage an important litmus test of political acceptability; virtually all Republican presidential candidates, including Senator Robert Dole, supported the cMarriage Protection Resolution d which the religious right began circulating in Iowa.<br><br> And, Senator Dole, who now has enough delegates to become the Republican nominee, has himself stumbled over the gay issue, as the controversy over the campaign donation from the gay Log Cabin Republicans last fall demonstrated. In other words, based on the high visibility of gay issues in recent years, there are good reasons to believe that gay, lesbian, and bisexual voters will be attuned to this year 9s political campaigns, concerned about where the parties and the candidates stand, and invested in the outcome of the elections. Can the Gay, Lesbian, Bisexual Vote Make a Difference?<br><br> Because self-identified gay, lesbian, and bisexual voters are not randomly distributed throughout the population, but instead are more heavily represented in highly urbanized states, they can make a significant difference to the outcome 13 of a presidential election, as well as state and local contests. States like California, New York, Pennsylvania, Ohio, Illinois, Florida, Texas, Michigan, New Jersey, Washington, and Maryland all have major urban areas whose voting patterns can decide which Presidential candidate will win in that state. They also have well-organized, visible, and politically motivated gay, lesbian and bisexual communities, many of whom have already had to mobilize around ballot initiatives, state legislation, and municipal civil rights campaigns.<br><br> No presidential candidate can hope to win without capturing at least several of these states. Together, they account for 49 percent of the total electoral vote. The urban concentration of the gay vote puts it in a strategically important position in a presidential election where the outcome is determined not by the total number of votes received, but by the number of electoral votes that are won.<br><br> High population urbanized states play a special role in a presidential election; a relatively small group of voters concentrated in a few areas can exert an influence out of proportion to their numbers. The gay vote can likewise play a major role in the outcome of local and statewide elections in many of these states. Significantly, key institutions within the gay, lesbian and bisexual community 4 particularly the more than five dozen community services centers that have sprung up in major cities around the United States 4have committed themselves to voter registration and mobilization as a central priority for 1996.<br><br> To the degree that these efforts succeed, they will be expanding the pool of registered voters, strengthening the political self-consciousness of their communities, and increasing the likelihood that a self-identified gay, lesbian, and bisexual vote will grow in 1996. Who Will They Vote For? And Will They Vote?<br><br> The tally of how gay, lesbian, and bisexual voters cast their ballots in 1992, and the political profile on issues that emerged from the 1992 exit poll, seems to tell a very simple tale of political loyalties. Not only did they vote for Clinton by a large margin in 1992, but their position on issues like taxes, government services, health care, social values, and abortion points toward a group of voters hostile to the direction charted by the cRepublican Revolution d of 1994. If the only two options facing these voters in the 1996 presidential campaign were a Democratic and a Republican candidate, the story would be over.<br><br> But gay, lesbian, and bisexual voters have a third option. They can stay home. Evidence from the 1994 Voter News Service exit poll points to this as a reasonable alternative.<br><br> In the 1994 midterm elections, in which gay, lesbian and bisexual issues received less attention, and at a time when disappointment over the Clinton Administration 9s support of the cdon 9t ask, don 9t tell d compromise in the debate over the military exclusion policy was still fresh, the self-identified gay, lesbian and bisexual vote shrunk to approximately two percent, a one-third decline from its percentage in 1992. The urban concentration of the gay vote puts it in a strategically important position in a presidential election 14 Why might many gay, lesbian and bisexual voters choose to sit out the 1996 election? President Clinton has been friendlier to the gay, lesbian and bisexual community than any previous president.<br><br> The Clinton Administration has adopted a rhetoric of inclusion, has been open toward gays, lesbians, and bisexuals in appointments within the federal government, has appointed a liaison to the community, has endorsed policies of non-discrimination in employment, and has supported government action in the fight against AIDS. Yet it is also true that a community 9s expectations, and its standard of what constitutes support, can change. Thus, along with the above support, the Clinton Administration has also made decisions that have deeply disappointed the gay community.<br><br> In particular, in the two most highly publicized gay, lesbian and bisexual issues of the last three years, the Administration was found wanting. First, it was unable to deliver on its promise to repeal the military 9s ban on gay, lesbian, and bisexual servicemembers, and instead supported a compromise, the so-called cdon 9t ask, don 9t tell d policy, that hasn 9t reduced the number of servicemembers discharged, and has deeply angered the gay community. Second, in what is without question the most important gay-related civil rights case of the last decade to come before the Supreme Court, the Administration declined to submit an amicus brief challenging the constitutionality of Colorado 9s Amendment 2.<br><br> Given these disappointments, low voter turnout among gay, lesbian, and bisexual voters is a third option. The likelihood of this outcome depends on several factors: whether the Democratic Party actively courts the gay, lesbian and bisexual vote in an attempt to counteract the disappointments in Clinton 9s record or whether it simply takes this community 9s vote for granted; whether the Republican party, through its hostile rhetoric, antagonizes the gay community in ways similar to what happened in 1992 or whether it chooses to avoid a focus on polarizing social issues. In other words, while it is not true that the gay, lesbian and bisexual vote is up for grabs, its loyalties and turnout level are uncertain.<br><br> The Republican presidential candidate is unlikely to capture this vote, but the Democratic presidential candidate can lose it by inaction. The Republican party can antagonize this vote, or neutralize it; the Democratic party can effectively mobilize it, or lose it. In light of the evidence detailing the size, concentration, and potential for growth of the gay, lesbian, bisexual vote, how it votes in 1996 can make a difference in the outcome of the Presidential election and in numerous local and statewide elections.<br><br> The difference the gay vote makes will depend on how each of the major parties responds to the needs and aspirations of this community. In the 1994 midterm elections, at a time when disappointment over the Clinton Administration 9s support of the cdon 9t ask, don 9t tell d compromise in the debate over the military exclusion policy was still fresh, the self- identified gay, lesbian and bisexual vote shrunk to approximately two percent 15 References: Bailey, Robert. cThe Lesbian and Gay Vote: A Background Report. d Unpublished Paper, 1996.<br><br> Edelman, Murray. cThe Gay and Lesbian Vote and Estimates of Population Size. d Paper presented at the annual meeting of the American Statistical Association, Toronto, Canada, 1994. Edelman, Murray.<br><br> cUnderstanding the Gay and Lesbian Vote in 892. d The Public P er spective, March/April 1993, pp. 32-33. Hertzog, Mark.<br><br> The Lavender V ote. New York: New York University Press, 1996. Forthcoming.<br><br> New York Times. November 5, 1992; November 9, 1994; November 10, 1994; November 13, 1994. 1 The VNS exit poll did not include ctransgender d as a self-identifying option.<br><br> Hence, the data generated by the exit poll does not permit any conclusions about transgendered voters. Additional copies of this report may be purchased at a cost of $5.00 (includes shipping and handling) Please contact our publications department at: NGLTFPolicy Institute Attn: Publications 2320 Seventeenth Street, NW Washington, DC 20009-2702 TEL 202-332-6483 ext 3327 EMAIL publications@ngltf.org

less