IN THE 2004 presidential election, sixty percent of the voters said that they supported same-sex marriage or civil unions and 37 percent opposed any form of legal recognition for same-sex relationships. Under normal circumstances in American politics, a sixty to 37 percent margin would be considered to be a stunning victory. Instead, the notion that the issue of same-sex marriage cost the election to the Democratic Party has been uncritically accepted as the common wisdom. In fact, majority opposition to legalized same-sex relationships was limited to members of a relatively small number of overlapping analytic categories, and virtually all of these are among the core supporters of the Republican Party—people who are not likely to vote Democratic under almost any condition.
Here are the fourteen groups in which a majority opposes legal recognition of same-sex relationships:
1. Those who believe that abortion should be illegal in all cases (sixteen percent of all voters): 74 percent oppose legal recognition of same-sex relationships.
2. Those who believe that abortion should be illegal in most cases (26 percent of all voters): 52 percent oppose.
3. Those who attend religious services more than once a week (sixteen percent of all voters): 68 percent oppose.
4. Protestants who attend services weekly (sixteen percent of all voters): 62 percent oppose.
5. Born-again or evangelical Christians (36 percent of all voters): 61 percent oppose.
6. Those who say they are “enthusiastic” about the Bush administration (22 percent of all voters): 58 percent oppose.
7. Voters who were contacted by phone or in person by the Bush campaign but not by the Kerry campaign (ten percent of all voters): 53 percent oppose.
8. Those who are “not at all concerned” by the availability and cost of health care (two percent of all voters): 68 percent oppose.
9. Those who usually think of themselves as Republicans (37 percent of all voters): 51 percent oppose.
10. Those who think of themselves as conservatives (31 percent of all voters): sixty percent oppose.
11. Those who voted for George Bush in 2004: (51 percent of all voters): 51 percent oppose.
12. People who are Mormons or members of the Church of Latter Day Saints: (two percent of all voters): 61 percent oppose.
13. Those who voted for George Bush in 2000 (44 percent): fifty percent oppose.
14. White Protestant conservatives (sixteen percent of all voters): 64 percent oppose.
These data show that even among these groups of core Republican voters, opposition to legal recognition of same-sex relationships rarely is overwhelming. Only among those who oppose abortion in all cases do close to three quarters of the voters oppose, and only among those who attend religious services more than once a week do two-thirds oppose.
Some will argue that while the issue of same-sex marriage did not affect the direction of the vote, it did affect turnout: those who opposed it were more likely to vote, particularly in “battleground” states. This might be the case, but the election returns indicate that President Bush did less well in these states than in states that did not have referenda on same-sex marriage; it also begs the question of why the Democratic Party, the Kerry campaign, and their affiliated interest groups were unable to find a way to mobilize their core constituents. We cannot eliminate organizational failure on the part of the Democratic campaign to generate comparable turnout among their core constituents as a more likely explanation of the Republican victory.
Why should same-sex marriage motivate turnout more effectively than unemployment in a state like Ohio? Is it possible that Republicans were more likely to get voters out because they relied on in-state volunteers, while Democrats relied on paid workers and out-of-state volunteers? Any politician can tell you that local organization matters in a get-out-the vote effort. The Democratic leadership’s placing the blame for their loss on same-sex marriage may reflect nothing more than an inability to face the party’s organizational failures—or it may reflect a felt need for a scapegoat. Our data provide scant, if any, evidence that the issue of same-sex marriage had a significant effect on the presidential election. Far from it, our data indicate that under forty percent of the American voters supported denying legal recognition to same-sex couples.
The interesting story about the increased turnout in 2004 was that a rising tide lifted all boats. While all groups increased in the number of voters, the relative percentages of the groups remained constant, so there was no net effect. Analyzing turnout data and county-level data, Simon Jackman, a Stanford political scientist, found that the same-sex marriage initiatives in eleven states boosted turnout by about 3.3 percent in those states. Looking across the electorate, all groups remained roughly constant since 2000, including GLB voters at four percent. Jackman found that the marriage initiative seems to have boosted turnout and the intensity of opinion, but there was no relationship between the change in Bush support and the change in turnout. In fact, the six states in which Bush had the greatest increase in his share of the vote did not have a same-sex marriage initiative on the ballot, while two of the states that did have an initiative were among the five in which Bush’s percentage changed the least. On average, Bush went up 2.8 percent in states that had no marriage initiative and 2.5 percent in those that did. Jackman looked at Ohio in detail and found no relationship between an increase in turnout and support for a same-sex marriage initiative across the state’s 88 counties.
The data indicate that same-sex marriage had little net effect on the outcome of the election. Those in favor of civil unions split 52 to 47 for Bush over Kerry, mirroring the national outcome. Among the 25 percent who favored same-sex marriage, Bush got 22 percent of the vote—or about 5.5 percent of the total. Among the 37 percent who opposed all legal recognition of same sex relationships, Kerry got 29 percent of the vote—or about eleven percent of the total. Kerry, in fact, got more votes from opponents of legalized recognition than Bush got from supporters of same-sex marriage, so the issue’s net effect may have been to benefit Kerry rather than hurt him.
The growth in support for civil unions is astounding. Just a few years ago, legislators in Vermont feared that their careers were over as a result of voting to enact a system of civil unions. Today, the public clearly views civil unions to be a viable alternative. On average, 35 percent of all voters in the 2004 supported civil unions, and the support is broad-based. Thirty-four percent of Kerry voters and 36 percent of Bush voters supported civil unions, while forty percent of Kerry voters supported same-sex marriage and 51 percent of Bush voters opposed any legal recognition of same-sex relationships.
Four groups offer majority support for same-sex marriage. Jewish voters are the ones most likely to support same-sex marriage, with 64 percent in favor, 28 percent supporting civil unions, and only four percent opposed to any legal recognition of same sex marriages. They’re followed by liberals (54 percent), those who believe abortion should be legal at all times (52 percent), and lesbian, gay, and bisexual voters (51 percent).
The fact remains that, on average, only a quarter of the voters supported same-sex marriage. This accounts for the success of the referenda defining marriage as being between one man and one woman in thirteen states this year. When the issue is framed as a yes-or-no question, same-sex marriage is defeated. Surveys can provide voters with a greater range of choices and thus produce more nuanced results. Only three voters in eight opposed providing legal recognition to same-sex relationships when provided with the alternative of civil unions.
Kenneth Sherrill is professor and chair, Department of Political Science, Hunter College, City University of New York.
