What explains Americans’ ever-growing partisan polarization? The Rockefeller Republicans and Blue Dog Democrats of yesteryear have been largely purged, while Americans who strongly identify with a political party have grown increasingly hostile to the other side. A common explanation is that as the parties developed more ideologically consistent messages and staked clear positions on a wide variety of public policies, beginning in the waning decades of the twentieth century, Americans responded to these clear distinctions by sorting themselves into the party that best fit their own beliefs and values.
In this scenario, one of the main partisan sorting factors is religion. Scholars have documented a growing “God gap” in American politics: Republicans are far more likely to attend religious services than Democrats, regardless of denomination. (This relationship between religiosity and Republicanism does not extend to African Americans or minority faiths such as Judaism or Islam.) The rise of the religious right during the 1980s pushed the GOP rightward on social issues as the party sought to secure the votes of conservative Christians who were appalled by abortion, gay rights, and the shrinking of religion’s role in the public sphere. This synergy between conservative Christians and the Republican Party in turn made the Democratic Party more appealing for Americans who were alarmed by the merging of religion and politics.
While there are still plenty of religious Democrats and probably more than a few secular Republicans within the American electorate, the God gap hypothesis continues to dominate many analyses of the electorate due to its simplicity: it’s easy to see from exit polls and other analyses that religious conservatives dominate the voting coalitions of the Republican Party while secular Americans are becoming a bigger part of the Democratic Party. Given that the GOP agenda has extensively catered to the views of conservative white Christians for decades, it’s not surprising that studies find that they are the most likely to indicate that their religious beliefs play a role in their voting behavior and partisanship.
But what if the reverse is also true? What if partisanship itself impacts individual attitudes about religion? In other words, does being a Republican make some Americans more religious? Does being a Democrat make them less? In her excellent new book, From Politics to the Pews, Michele Margolis argues that the answer is yes. Her findings have important implications for our understanding of the relationship between religion and politics in the United States.
In developing her thesis, Margolis, a political scientist at the University of Pennsylvania, draws from the life cycle theory of American religion. Sociologists have long known that late adolescence and early adulthood are when Americans are most prone to stop attending church or even affiliate with a religious tradition. But, as they marry and have kids, many return to the religious fold. After that period, religious switching is relatively rare.
Margolis’s insight is that because partisan identity tends to become fixed earlier than religious identity, it should be possible to measure the impact the former has on the latter. She reasons that “when individuals must decide if and how to engage in the religious world, their partisanship identities—solidified earlier in young adulthood—may exert an important and lasting influence on their religious identification and practices.” Focusing on younger adults is key: once religious identities have solidified, “partisanship’s effect on religion should be relatively muted for those squarely in adulthood.”
More specifically, Margolis hypothesizes that Republican young adults with children will be more likely to return to a specific religious tradition, and resume higher rates of church attendance, than young Democratic parents. She tests this theory statistically using data from the Youth-Parent Socialization Panel Study, which interviewed the same set of individuals, and their parents, four times across four decades, beginning in 1965, when the children were seniors in high school. She finds that in 1982, when these young adults were the most likely to have gotten married and begun to have kids and when religion and politics were becoming linked in new ways, Republicans were significantly more likely than Democrats to return to the pews and significantly less likely to be secular. While 30 percent of Democrats in 1973 who had not yet become parents attended church regularly, compared with 32 percent of Republicans, “by 1982, 42% of Republicans reported being a regular attender compared to 33% of Democrats.” Members of both groups returned to the fold as they reached full adulthood, but Republicans did so in higher numbers. And, “[b]y 1997, Democrats’ religiosities had not ‘caught up’ with Republicans.” The parent cohort, meanwhile, did not show any change with respect to their religious behavior, consistent with the life cycle theory.
To illustrate that this phenomenon is not merely a vestige of growing up in the 1960s, Margolis analyzed more recent panel data collected by the National Annenberg Election Study and the Institute for the Study of Citizens and Politics covering the time period from 2007 to 2014. Comparing rates of religious change among Republicans who were new parents in 2007 with Democrats in the same category, Margolis found similar partisan-driven religious change. Church attendance grew among Republicans by 10 percentage points more than among Democrats, and “Republicans became 17% more likely to be religious identifiers relative to Democrats.”
In addition to analyzing those panel studies, Margolis conducted her own national survey experiments to try to figure out what’s driving these disparities. Democrats who read about the role of conservative evangelicals in politics, she found, became more likely to report a lower religious identification score; the reverse was true among Republicans. (In all cases, this effect was most pronounced among younger adults who were married with children living at home.) But when Democrats were exposed to a newspaper article linking religion to liberal immigration reform, their religiosity didn’t shift. Those results suggest that the increasing association between Republicans and conservative Christian theology is driving Democrats away from religion.
There are two major caveats to her findings. First, the life cycle effect is not apparent among Americans who have very low levels of political knowledge, roughly one-third of respondents in her data set. Second, the impact is largely constrained to white people. African Americans, in particular, are more devout overall and report higher levels of engagement with their churches. And since they are overwhelmingly likely to be Democrats, there’s no real partisan split to analyze.
Margolis’s work should make us more careful in assuming that people’s partisan affiliations are the product of their personal values, and not the other way around. So, for instance, if many conservative voters have become more religious because they’re Republicans, then a Democratic politician is unlikely to peel away their votes by adopting more conservative policy positions on an issue like abortion.
Although the name Donald Trump does not appear in these pages, it’s hard not to think about the 2016 presidential election while reading this book. Margolis’s data shows that conservative political cues matter most when it comes to activating religious behavior, particularly among parents with school-age children; it does not, however, explain why the most devout conservative Christians overwhelmingly supported Trump, and continue to do so, despite his obvious moral shortcomings.
To address this latter question, I would love to see another study that considers whether partisanship shapes the theological arguments that conservative Christian Americans use to defend Trump. One marvels at the contortions some religious leaders engage in to justify their politics, as when they liken Trump to King David, the heroic Old Testament king who also committed adultery. But is Trump changing what rank-and-file white evangelicals believe to be the teachings of the Bible? I’m guessing yes. In October 2016, a PRRI poll showed that 72 percent of white evangelicals, more than any other group, agreed that “an elected official who commits an immoral act in their private life can still behave ethically and fulfill their duties in their public and professional life.” By contrast, just 30 percent of evangelicals agreed with this notion in 2011. If partisanship can alter the moral lens of acceptable political behavior, then it seems likely that it will alter theological beliefs more generally. But we don’t have empirical evidence yet to back up this claim.
In the short term, Margolis’s work helps explain the emergence and staying power of the mutually reinforcing relationship between Republicanism and conservative Christianity. The long-term implications of her study, however, should give Republicans pause. While white Christian Americans make up the most reliable voting coalition of the GOP, their share of the population is shrinking dramatically. Moreover, From Politics to the Pews offers compelling evidence that the ties that bind one group of voters to the Republican Party is repelling a younger generation of Americans, who are far more supportive of gay rights and far more suspect of the mixing of religion and politics. Politics alone doesn’t explain why Millennials are opting out of religion at a record pace, but it is surely a factor. Given that the Millennial generation is currently the largest one by population size, the GOP may soon find itself preaching to an ever-shrinking choir.