In the United States, as in most of the world, lesbians, gay men, and bisexuals are outnumbered and despised. Unlike most other potentially political groups, gay people are further disempowered by virtue of being born as if into a diaspora—probably randomly distributed about the population at birth. Their quest for political power is disadvantaged by barriers to the formation of political community as well as by lack of access to significant power resources. The relative political powerlessness of gay people stands in a contradistinction to their depiction by advocates of “traditional values” as a powerful movement advancing a “gay agenda” in American politics.

Similarly, the attention paid to the occasional electoral victories of openly lesbian and gay candidates distorts the reality that fewer than one tenth of 1% of all elected officials in the United States are openly lesbian, gay, or bisexual. With only two exceptions—Wisconsin and Minnesota—every state that has passed legislation banning discrimination on the basis of sexual orientation is either on the Atlantic Coast or on the Pacific Coast (including Hawaii). Political scientists long have understood that rare local victories—whether they be the enactment of nondiscrimination legislation or the election to office of a member of a group—do not translate into national political power for the group.

In the pages that follow, I will attempt to measure the access gay people have to meaningful political resources and to discuss how differential access to these resources disadvantages gay people in the political process.

During the past thirty-five years many commentators have expressed concern about declining support for the American political system, noting familiar evidence of the steady erosion in electoral turnout (Stanley and Niemi 1995, 78; Teixeira 1992), falling participation in political parties (Rosenstone and Hansen 1993), plummeting levels of political trust (Lipset and Schneider 1987), and weakening civic engagement (Putnam 1995a). Evidence for declining confidence in American government is well-established. The standard National Election Study measures show that in 1954 three-quarters of the American public trusted government in Washington to do what was right ‘just about always or most of the time’. By 1994, a quarter of the public proved as trusting. Moreover how far Americans trust each other—or social trust—has also fallen by more than a third since the early sixties (Uslaner 1995; Putnam 1995a).

Not all the evidence points in the same direction, and some alternative forms of political activity may have risen over time (Verba et al. 1995, 70–71). Moreover comparative research (Klingemann and Fuchs 1995) provides no evidence for a uniform secular decline in electoral turnout and confidence in government across advanced democracies, as sometimes assumed by observers. Nevertheless it is widely believed that American democracy has been experiencing a crisis of legitimacy, with angry voters disillusioned by Washington politics as usual.

An examination of presidential voting patterns between 1828 and 1992 for all counties and most large cities in the continental U.S. (approximately 135,000 cases altogether) confirms that there has been a decline in turnout rates since 1960, as most commentators have suggested. For the nation as a whole, turnout in presidential elections dropped 20% from 64% in 1960 to 51% in 1988. Turnout rebounded slightly in 1992 to 55%, due perhaps in large part to the interest generated by Ross Perot.

As striking as these data appear, they must be put in historical perspective to be properly understood. Graph 1 displays the turnout rate for every presidential election between 1828 and 1992 (See Table 1 for a listing of the actual rates). It shows that while turnout in the presidential election of 1988 was the third lowest since 1828, surpassed only in 1920 and 1924 (with turnout rates of 44% in each election), the 1960–1988 decline is not unprecedented in U.S. electoral history. In an earlier 28-year period (1896–1924), turnout rates in presidential elections declined from 72% to 44%. This is a 39% decline in turnout almost double that experienced between 1960 and 1988.

Graph 1 also shows that while most commentators use the 1960 election as a basis for gauging changes in turnout, it is a somewhat misleading baseline. Turnout in 1960 was 64%, which was the highest turnout rate for U.S. presidential elections since 1900. The 1960 election capped a 36-year rise in turnout rates, and represented a 45% increase over the low points registered in 1920 and 1924. Indeed, with the exception of 1944 and 1948, when the nation was preoccupied with World War II and its aftermath, the period between 1928 and 1968 shows a steady increase in presidential turnout rates.

For at least two decades, it has been conventional wisdom among political guras that our system of choosing presidents through electoral votes, accumulated state by state, offers Republicans a substantial edge. Representative is a comment by a Democratic strategist to Newsweek (7/27) in the summer of 1992: “It's the same map we've had to deal with for years—and the results haven't been good. You need all the breaks with a map like that….”

The notion that Republicans have an “electoral lock” on the White House has been received wisdom among pundits ever since Horace Busby, a former aide to President Lyndon Johnson, coined the phrase. The idea is that even if Democrats fare well in the popular vote, the electoral map is skewed against them.

There is only one problem with the idea. The election numbers just don't support it.

It is obvious that, in presidential contests of the 1970s and 1980s, major regions went overwhelmingly Republican. From 1972 through 1988, the Democrats did not once carry a single state among the eight of the Mountain West. They did little better in the South: Jimmy Carter dominated the region in 1976, but Democrats carried a total of just one old Confederacy state in the other four elections. Clearly these have been regions of deep Democratic weakness at the presidential level.