Revenge of the Gay Voters

Conventional political wisdom has it that so-called security moms -- suburban mothers who called terrorism their most pressing political concern -- were the key voters in the 2004 presidential election. In this year's midterm elections, newly out gays and lesbians could cast the deciding votes.

Eleven states -- nine of which are home to at least one hotly contested congressional race -- saw the number of out gay, lesbian and bisexual couples increase by at least half between 2000 and 2005, according to a new study of census data released by the Williams Institute, a Los Angeles think tank focusing on sexual orientation and public policy. A second related study concludes that the impact of gay, lesbian and bisexual voters will be greatest in House races with Republican incumbents -- which include several of the most contested races -- and Senate races with Democratic incumbents, such as the New Jersey contest.

Ohio, which notoriously decided the 2004 election, ranks sixth among states whose gay, lesbian, and bisexual population mushroomed. The state will figure just as importantly in the midterm elections, with neck-in-neck races for three House seats, a Senate seat and the governor's mansion.

In Minnesota, the number of out gay, lesbian and bisexual couples increased by 76 percent. Next week, the state will decide two key House races and a Senate seat. Minnesota's gubernatorial election will also play an important role in establishing the political zeitgeist among the nation's governors, and Minnesota is one of eight states deciding on an anti-gay marriage amendment.

Although sexuality, unlike age or religion, is rarely identified in polling data, the newly out gays and lesbians identified by the Williams Institute study are already accounted for in general public opinion polls. However, Gary Gates, the author of the study, suggested that because "they're now being more open," there could be some shift in their voting behavior. "It's not unreasonable to think that perhaps gay populations are paying a bit more attention to the political environment. If that's true, they might be a little more likely to vote," he said.

Ken Sherrill, a political scientist at Hunter College, is one of just a handful of people studying the gay, lesbian and bisexual community as a voting bloc. When the national exit poll added a question about sexuality in 1990, it became the most significant source of information on the voting behavior of GLB people (there is no data on the voting habits of transsexuals, who add the final T to the more common abbreviation GLBT). Sherrill's analysis of exit polls from the 2004 election shows that those who identified themselves as gay, lesbian or bisexual were 50 percent more likely to call themselves Democrats than those who didn't, and were more than twice as likely to characterize themselves as liberals. Not surprisingly, GLB voters were also twice as likely as heterosexuals to support gay marriage.

GLB voters are most likely to affect the outcomes of the next week's election in the eight states voting on anti-gay marriage amendments. All of those states have existing legislation prohibiting same-sex marriage. The measures exist primarily to bring religious conservatives to the polls, where they will most likely vote Republican: Five of the eight states are home to congressional races which are too close to call.

Beginning in the mid-1990s, 38 states have passed laws prohibiting gay marriage. In the late '90s, states began amending their constitutions to banish the practice with no risk of running afoul of constitutional protections. Nineteen states have passed such amendments in addition to the eight states voting on them next week. (The numbers add up to more than 50 because many states have both a law and a constitutional amendment.)