Winning numbers: Religion in the 2004 election

In winning reelection George W. Bush expanded his 2000 coalition primarily by increasing the turnout and his support among key constituencies, including religious communities. The Kerry campaign tried to do the same, but it had less success, especially on the religious front.

Bush was reelected with 51 percent of the popular vote (59.6 million at this writing), with John F. Kerry receiving 48 percent (56.1 million). Thus, Bush turned his 500,000 vote deficit of four years ago into a 3.5 million advantage. These gains occurred while turnout soared from 52 percent of the eligible electorate to at least 57 percent. The result was a narrow win over Kerry, who nonetheless received 6 million more votes than Al Gore did in 2000.

White evangelical Protestants are the largest group of Bush’s religious supporters. In 2004, 78 percent of evangelicals voted for Bush, providing about 21 million votes (35 percent of the total Bush ballots). Changes in the exit poll wording in 2004 make a direct comparison to 2000 problematic, but Bush received at least 3.5 million more evangelical votes this year—equal to his overall popular vote margin. If the claims of Bush strategist Karl Rove about the lack of evangelical turnout in 2000 have merit, then the increase may have been larger.

Bush also made gains among Catholics, becoming the first Republican presidential candidate to win a majority of that faith. He bested Kerry, a Catholic, by 52 to 48 percent. The Republicans did especially well among white Catholics (56 to 44 percent) but also improved their tally among Hispanics (42 percent this year, 31 percent in 2000). Overall, Catholics provided Bush with 16 million votes in 2004 (27 percent of all Bush ballots), for an increase of about 3 million.

Bush also won white mainline Protestants by 52 percent, down a bit from 56 percent in 2000. Mainline Protestants accounted for 13 million votes (22 percent of all Bush ballots), up by about 1 million. The president also made some gains among black Protestants (doubling his 2000 support to 16 percent) and did a bit better among Jews (25 as opposed to 19 percent). The combination of these groups, people of other faiths (measured with great imprecision in exit polls), and the unaffiliated provided Bush with another 9 million votes (16 percent of the total Bush ballots).

So the increase in evangelical votes was critical to Bush (it equals his margin of victory), but the increase in votes from Catholics and other groups was nearly as important.

What about religion and the Kerry coalition? Kerry’s biggest source of votes was Catholics. Although he lost the vote among Catholics, they provided 15 million votes (27 percent of all Kerry ballots). The second largest source was white mainline Protestants with 12 million votes (21 percent of all Kerry ballots), followed by black Protestants with 9 million (16 percent of all Kerry ballots). In all three cases, Kerry received roughly 2 million more votes than Gore did in 2000. But if Kerry had repeated Gore’s performance among Catholics, he would have had another million votes. The combination of Jews, people of other faiths and the unaffiliated provided another 14 million votes (25 percent of the total Kerry ballots). Although the small numbers make it hard to judge, it appears that Democrats received nearly all the Muslim vote in 2004. Finally, Kerry received 6 million evangelical votes (11 percent of all Kerry ballots), down about 2 million from Gore’s total in 2000.

If Kerry could have replicated Gore’s performance among Catholics and evangelicals, he could have substantially closed the gap with Bush.

Worship attendance, a crude measure of religious traditionalism, was a strong predictor of the vote, as it was in 2000. Overall, individuals reporting weekly attendance supported Bush by 61 percent, while nonattenders went 62 percent for Kerry. Bush edged out Kerry among those who attended a few times a month, 50 to 49 percent, erasing Gore’s 51 to 46 percent advantage, but Kerry maintained the Democratic lead among those who attended a few times a year (54 to 45 percent). Thus, in most religious communities Bush did best among more traditional adherents and Kerry among the less traditional.

One reason for these patterns was the mix of issues in the campaign. Opposition to same-sex marriage was a galvanizing issue for evangelicals and other traditionalists. It and other “moral” questions were adroitly exploited by the Republicans and their allies.

On the other side, the Democrats and their allies talked about the economy, social programs and the Iraq war, issues that resonated with many religious people’s sense of social justice.

The exit polls reveal the impact of these rival agendas. More than one-fifth of the respondents said “moral values” most influenced their vote and 80 percent backed Bush. Although this category is vague, additional questions showed that 16 percent of the electorate favored banning all abortions (77 percent of this group went for Bush) and 37 percent opposed giving any legal status to same-sex couples (70 percent for Bush).

Meanwhile, 20 percent said that the economy and jobs most influenced their vote and 80 percent of them backed Kerry. In addition, 15 percent listed the Iraq war (73 percent for Kerry) and 12 percent listed education or health care (75 percent for Kerry). Indeed, “social justice” priorities may well have exceeded “moral” priorities in the electorate as a whole. (Another 19 percent gave priority to terrorism and backed Bush with 86 percent.)

In sum, it is important not to overstate the role of any particular religious group in the 2004 election. In a close contest, most religious groups matter—and so do all manner of other demographic and political factors. Religious constituencies were important parts of both the Kerry and Bush coalitions. The Bush coalition prevailed—by a nose.