Everything else being equal, the more religious the individual in the U.S. today, the higher the probability that the individual identifies with or leans toward the Republican party. I called this the “R and R rule” in my 2012 book on religion, found the phenomenon alive and well in my 2014 review of Gallup data, and now, nine years later, Gallup’s data confirm that this religiosity gap is more evident than ever.
Americans’ political identity is a powerful correlate of a wide range of Americans’ attitudes and behaviors, including, in particular, a wide range of attitudes about hot-button political and social issues. And we know that political identity is related to views of the national economy, views of the nation’s institutions, happiness, perceptions of the nation’s most important problems, and a variety of other measures. It is thus not surprising that political identity would also be related to religion.
One key measure we use in analyzing this politics and religion relationship is the absence of religious identity — those who, when asked by a survey researcher about their religious identity, reply that they have none. “Nones” have risen from essentially zero in some Gallup surveys in the 1950s to above 20% in recent Gallup surveys (and higher than that in some other firms’ surveys). Not surprisingly, researchers have zeroed in on this phenomenon, and the term “nones” has entered the popular vocabulary (as exemplified by books with titles like The Nones: Where They Came, From Who They Are, and Where They Are Going (by Ryan P. Burge), None of the Above: Nonreligious Identity in the U.S. and Canada (by Joel Thiessen and Sarah Wilkins-Laflamme) and The Rise of the Nones: Understanding and Reaching the Religiously Unaffiliated (by James Emery White).
An aggregate of five recent Gallup surveys conducted from May 2021 through May 2023 confirms the extent to which Democrats are much more likely than Republicans to be nones. Twenty-six percent of nones identify with or lean toward the Democratic party, compared with just 11% of nones who identify with or lean toward the Republican Party.
Looked at differently, across the combined 2021-2023 data, equal proportions of Americans identified as Democratic or Democratic-leaning (46%) or Republican or Republican-leaning (46%). Yet, the group of Americans who are religious nones split 63% Democratic versus 26% Republican — far different from the population at large.
Other measures of religiosity Gallup tracks show the same general R and R relationship. Republicans are more likely than Democrats to report that religion is important in their daily lives, and Republicans are more likely than Democrats to report attending religious services.
Partisan Gap in Nones Increasing Over Time
The magnitude of this religion gap has increased over the years, to the point where I think it is fair to say that a significant part of the explanation for the “rise of the nones” lies with changes among Democrats, not Republicans.
I looked at this relationship on a year-by-year basis using our annual May GPSS Values surveys, a valuable exercise because the same questions have been asked in basically the same survey context every May since 2001. The percentage of Republican nones has edged up over the last 10 years or so, but the percentage of Democratic nones has increased significantly more. In short, the relatively small partisan gaps in none identification seen two decades ago have increased substantially over the years.
Republicans Less Likely to Be Nones, but Much More Likely to Be Protestants
Mathematically speaking, if Democrats are more likely to be nones than Republicans, Democrats have to be less likely to identify with one or more other religious groups to make up for it. The two largest religious groups where this could play out are Protestants (about 46% of all national adults) and Catholics (about 24% of all national adults). The data are clear that it is the Protestants who take the brunt of the rise of the nones among Democrats, not Catholics.
The differences are striking. Fifty-six percent of Republicans identify as Protestants, compared with just 38% of Democrats. That compares to the 11% of Republicans and 26% of Democrats who, as noted above, identify as nones. By contrast, the percentage of each party identifying as Catholic is roughly equal. Unlike Protestants, Catholics have maintained a roughly stable percentage of the American population over time and, as seen here, have remained largely immune from the partisan divide evident among Protestants and nones. Explanations for this phenomenon are tied up with — among other factors — the impact of immigration on the composition of the Catholic population.
Searching for Explanations
Figuring out why the basic R and R relationship exists provides a fascinating and important challenge for social scientists. Not surprisingly, we have seen a cascade of books, papers and other publications focused on theories and explanations for this aspect of American social and political life.
As is true with any correlation between two variables, the first question becomes one of causality (as taught in beginning statistics classes, correlation is no proof of causation). It could be that those Americans who — for whatever reason — are personally religious pick and choose a political identification that best fits their underlying religious beliefs. Or, conversely, it could be that Americans who — for whatever reason — are either Republican or Democratic in their political orientation tailor their religious beliefs to their politics. Or it could be that a third underlying factor — say, one’s ethnic or racial background, or the geographic location of one’s residence, or simply one’s family heritage — is the underlying causal factor that causes one to both be religious and to be Republican, or to be less religious and to be Democratic.
University of Pennsylvania political scientist Michele Margolis, in her 2018 book, From Politics to the Pews, makes the case that political identity is the primary causal factor in determining Americans’ religious identity, more so than the other way around. Her analyses, and our own Gallup discoveries on the extraordinary power of political identity in shaping Americans’ worldview, make this a quite plausible explanation.
This leads to the idea of an antireligion “backlash” — the hypothesis that Democrats have moved away from religion in reaction to the increasing visibility given to the association between religion and Republican and conservative politics. Sociologists Michael Hout and Claude Fischer in 2002 concluded a review of the evidence by saying, “This political part of the increase in ‘nones’ can be viewed as a symbolic statement against the Religious Right.” University of Connecticut sociologist Ruth Braunstein has more recently reviewed in great detail various manifestations of how this backlash theory could be playing out in the U.S. today.
The key takeaway is the possibility of a spiral or self-fulfilling prophecy effect — the idea that as religion becomes identified with Republican and conservative politicians and positions, it will continue to drive Democrats and liberals away from religion, amplifying the pattern by which Republicans increasingly dominate the group of those who remain religious. This, in turn, may mean that we are seeing a general pullback from religion in part because religion (like so much else in American life today) has become politicized.
This could work the other way around, as Braunstein points out in her review. Democrats could, in theory, react to what they see among those on the Religious Right by organizing their political beliefs around progressive values (the Religious Left) — one of three “narrow backlashes” Braunstein describes. There is not a lot of evidence supporting the idea that this is occurring. But this line of thinking raises the possibility of future patterns different than the increasing defection of Democrats from religious identity we have been seeing up to this point.
One explanation for the connection between Protestant religion and Republican politics is that it benefits both sides of the equation. Protestant religious leaders can gain fame, influence and authority when they are connected to political positions and political leaders. (Dr. Robert Jeffress, senior pastor of the very large First Baptist Church of Dallas, begins his online biography with an explicit reference to his political outreach: “Dr. Robert Jeffress is Senior Pastor of the 16,000-member First Baptist Church, Dallas, Texas, and a Fox News Contributor.”) Politicians, for their part, covet the support of highly religious Protestants, and Republican political candidates make repeated pilgrimages to evangelical churches and to religious events in efforts to gain the support of these highly religious voters. So, there are compelling reasons why the R and R connection may continue in the months and years ahead.
But this connection between religion and politics carries within it a singular paradox. Most religions, including Christianity, the dominant religion in the U.S., argue for social cohesion and love for one’s neighbor, while politics carries within it the fundamental structure of disagreement, conflict, argument, and castigation of one’s opponents. How this plays out in the future is a matter of fascinating theological and practical concern.
All of this matters, because religion matters in society. The evidence is clear that more religious people have higher levels of wellbeing and happiness. And religion has a number of other positive functions in society, including its influence on morality and pro-social behavior, its influence on charity and giving back to the community, and its contribution to social cohesion and solidarity. A continuing decrease in overall religiosity in American society, like the continuing decrease in faith in other American institutions, can have significant consequences for the health and viability of the country going forward.