The “Simple Demography” of Race and Ethnicity

It is a long-standing convention in comparative politics that parties and coalitions of the Right are at root a project of economic and social elites. Being statistically minoritarian, they compete effectively in democratic politics only by organizing cross-class alliances, often by deploying noneconomic appeals.^{Footnote 3} This may be an unspoken basis of the interpretation that conservatism decisively expanded its electorate in and after the 1960s only by taking advantage of an opportunity to help reorganize American politics away from the class cleavage characteristic of the New Deal era and toward a racial cleavage.^{Footnote 4} When national Democrats realigned their party with civil rights in the early 1960s, they alienated many white Southerners. National conservatives associated with Barry Goldwater and Richard Nixon developed a “Southern strategy” to attract these votes with proposals to resist school desegregation and busing. When it became clear that this strategy could attract whites in all regions of the country, Nixon and later Republicans like Ronald Reagan eventually crafted a core conservative agenda that incorporated items with explicit racial connotations (such as opposition to busing and affirmative action) but was otherwise ostensibly race-neutral, focusing on law and order, smaller government, tax relief, and welfare reform. These agenda items were nonetheless very much about race, and served to brand the GOP as the de facto party of whites and white-ness. With time, American images of minorities expanded to include Hispanics and eventually Asian Americans and Muslims as well. Organizing white identities and interests against all of these “others” explains why Republicans were able to develop and sustain a powerful cross-class coalition, reaching not only into the white middle class in every region but also deep into the ranks of white lower-middle class voters and working-class “Reagan Democrats” as well.^{Footnote 5}

This interpretation of conservative politics has been used to generate a prominent prediction. Conservatism may be able to mobilize a majority, maybe even a growing majority, of white voters. But despite that fact (more precisely, because of it), conservatism faces a crisis over the medium and long term. Racial prejudice may or may not have declined, but to the extent that whites develop more inclusive attitudes, conservatives may not be able to rely on ever-larger majorities of them.^{Footnote 6} At least as important, America’s nonwhite population is steadily growing. Many observers claim that ethnic and racial diversification will doom the GOP. Judis and Teixeira, for example, say this diversification “will constitute a formidable advantage for any Democratic candidate.” Chait remarks that “[e]very year, the nonwhite proportion of the electorate grows by about half a percentage point—meaning that in every presidential election, the minority share of the vote increases by about 2 percent, a huge amount in a closely divided country,” and one highly unfavorable to Republicans.^{Footnote 7} Such analyses were common around the 2012 election. A New York Times headline put it succinctly shortly after the vote: “Demographic Shift Brings New Worry for Republicans.”^{Footnote 8}

This ethno-racial interpretation of conservatism suggests three reasons why conservatism will have difficulty adapting effectively. The first is that many minority voters perceive Republicans as highly unattractive.^{Footnote 9} The second and third reasons concern, perhaps counterintuitively, the GOP’s existing and heavily white base. If some conservatives reach out to minority voters, we might expect resistance from many white conservatives, who would logically interpret the policy shifts required for such outreach as signals that the GOP will no longer defend their values and interests. A debate in the mid-2000s could illustrate the point. Party strategists like George W. Bush, Karl Rove, and John McCain proposed immigration reform as a large-scale policy gesture to Hispanics. But rank-and-file conservatives played a crucial role in blocking the legislation. In this interpretation, Republican elites have encouraged whites to adopt a form of identity politics; threatening that identity would logically inspire a backlash, in this case against accommodation of resented out-groups. Adam Przeworski and John Sprague’s analysis of historic European socialism offers another reason why Republican outreach to minorities might be difficult. They argue that European socialist parties needed to attract middle-class voters in order to win elections, and could do so only by downplaying the salience of class as a source of political identity. But this very act invited working-class voters to adopt non-class-based identities as well, such as religious ones, taking from the socialist electorate with one hand what it might give with the other.^{Footnote 10} Today’s American conservatives might face a similar dilemma. If Republican elites execute the policy shifts needed to attract substantial numbers of minorities, they will have downplayed race and ethnicity as a basis of political identity. In the process, they could run the serious risk that working-class whites, for example, would recast their own identities along class lines, something Democrats might be eager to help them do. For all these reasons, conservatives might find it very difficult to maintain a winning coalition in a demographically changing America.

This analysis of conservatism is based on several assumptions. First, it assumes a consistent, significant pace of demographic change in the electorate.^{Footnote 11} But Sean Trende and others argue that change is not occurring nearly as quickly as forecast. Because African Americans are broadly stable as a share of the population, Hispanics are at the heart of this discussion. Trende points out that Hispanics have consistently failed to grow as a share of the electorate at the rates many have predicted. That was even before rates of immigration from Mexico began to fall in the mid-2000s, including for reasons that may prove durable, such as declining fertility rates in Mexico.^{Footnote 12} This suggests white voters will retain substantial electoral sway for many years to come. There is even reason to believe that the notion of what constitutes white-ness may expand over time. Warren and Twine point out that the claim that whites are becoming a minority rests on “the essentialist premise that Whiteness is a fixed racial category . . . that racial categories are static across time and place.” They insist that the once-narrow American category of “white” was expanded “so that Slavs, Celts, and Mediterraneans could fit snugly” into it. As a result, “one cannot safely assume that Whites will become a numerical minority simply because the current immigrant population consists largely of people who are today defined as non-White.”^{Footnote 13}

Second, the demographic prediction assumes that minority voters will reliably prefer liberal political options at very high rates. This may well be true of African Americans for the foreseeable future. But Hispanics are famously not so emphatically Democratic. Republicans have proven capable of attracting substantial numbers of Hispanic voters, perhaps 40 percent in 2004 and 2010.^{Footnote 14} This is in part because Hispanic voting preferences are much more influenced by income than are African American votes. Trende notes that “wealthy Latinos actually vote more heavily Republican than poor non-Hispanic whites.”^{Footnote 15} Of course, Hispanics generally have lower incomes than whites, which structures their overall vote in more Democratic directions. But if incomes gradually rise, extrapolations of past Hispanic voting for Democrats could be misleading.^{Footnote 16} Winning one-third of Hispanics is no basis for conservative victories. But the sharp divergence between black and Hispanic voting habits and the sensitivity of Hispanic votes to factors like income suggest that the support of America’s largest minority group will be politically constructed rather than demographically predetermined. That is also suggested by significant shifts in Hispanic voting for the GOP over time, down through the 1980s and early 1990s, then up through 2004, and down and up again since then.

Third, the demographic prediction assumes that conservatives will be unable effectively to diversify their coalition ethnically because large numbers of whites identify with conservatism precisely because they understand it as a project of (implicit) whiteness. If this were true, conservatives might well be caught in a self-limiting and ultimately self-defeating loop. But this interpretation is also questionable. Modern conservatism sometimes seems reducible to the traditional, and overwhelmingly white, small-town heartland of which Sarah Palin seems emblematic. This is encapsulated in Allan Lichtman’s labeling of conservatism as the politics of the “white Protestant nation.”^{Footnote 17} Earlier in the twentieth century, such a politics might have seemed unsustainable in an increasingly non-WASP country. But historic Protestant-dominated conservatism was motivated to reach out across the salient ethnic lines of the day, with the result that today’s conservatism is already the product of waves of assimilation of minorities. The Republican National Committee had a Special Activities Division and a Nationalities Division in the 1950s precisely to reach out to Catholics in general and Italians, Poles, Ukrainians, Slovaks, Balts, and others in particular, especially on the basis of a shared anticommunism.^{Footnote 18} Under President Eisenhower, GOP leaders voiced ongoing determination to attract non-WASPs: “We never seem to have Italian-Americans, Polish-Americans, Negroes or many Catholics or Jews.”^{Footnote 19} The campaigns and administrations of Eisenhower, Nixon, and Reagan made considerable efforts to attract Catholic voters, among others.

These efforts achieved considerable success. The GOP gradually and jerkily absorbed millions of Catholic voters—ultimately half of them, constituting one of the great changes in postwar U.S. politics.^{Footnote 20} Catholics were not relegated to supporting roles. They were absorbed into modern conservatism’s leadership and provided much of its intellectual firepower, in the forms of public intellectuals, the leaders of National Review magazine, and much of the conservative judiciary.^{Footnote 21} Texas Republicans made major drives to attract Hispanics starting as early as the 1960s.^{Footnote 22} In more recent decades, conservatism assimilated conservative Jews and Cuban immigrants and their descendants, also not merely in supporting roles but as leading intellectuals, policy staffers, and officeholders. Over the last decade, conservative rank and file have done more than merely tolerate members of minority groups ascending to positions of leadership. These voters appear to have actively preferred minority candidates like Kenneth Blackwell, Marco Rubio, Nikki Haley, Tim Scott, and Ted Cruz over more centrist white primary rivals. In these and other cases, a demographically complex conservative identity was openly articulated and sought.^{Footnote 23} This suggests that many white conservatives do not understand their political identity in ethno-racial terms, and instead understand it in terms of values, policy prescriptions, and political orientations that can operate across religious, racial, and ethnic lines. This may not mean that substantial numbers of minority voters will reidentify as Republicans. But it suggests that the conservative project would not be at risk of fragmenting if such a shift took place.^{Footnote 24}

Secularization and the Role of Religion

A similar framework has been used to analyze the role that religiosity has played in conservative politics. The same underlying assumption may be at work: conservative socioeconomic elites deploy noneconomic appeals in order to develop needed cross-class electoral coalitions. Thomas Frank’s analysis of Kansas, for example, argued that economic elites controlling the Republican Party used appeals to traditionalist religious beliefs to attract the less-affluent voters required to make conservatism electorally viable.^{Footnote 25} In this interpretation, national Republicans encouraged the realignment of traditionally religious voters to the GOP, beginning as early as Eisenhower and primarily by engaging in symbolic politics. Gestures were as superficial as Nixon’s publicly close relationship with Billy Graham or as profound as shifts on policy issues, such as the GOP’s move to its current opposition to abortion rights.^{Footnote 26} Religiously inspired support enabled conservative success in Kansas and well beyond it, producing “the historical shift that has made it all possible.”^{Footnote 27} This suggests an appropriate strategy for liberals: reestablishing the salience of class. Howard Dean echoed this analysis when he was Democratic National Committee chair, proposing to fight the 2004 election on “our turf, which is going to be jobs, education, and health care” instead of “guns, God, and gays.”^{Footnote 28}

This interpretation, too, generates a prediction about conservatism, via long-standing forecasts of secularization. Sociological research on religious practice is complex. Common predictions about U.S. conservatism draw on the research tradition that identifies long-term declines in rates of individual religious belief and practice in increasingly affluent and educated societies. American secularization may lag Western European trends, but is widely assumed to be both under way and an inevitable feature of modernity.^{Footnote 29} This trend is predictably concentrated among the young, which may help explain why notable changes in attitudes toward homosexuality and other sexual mores have occurred in just the last generation.^{Footnote 30} Commentators have invoked these expectations to predict diminishing prospects for political conservatism. This figured in the “emerging Democratic majority” thesis of Judis and Teixeira, who cite data suggesting that American levels of religious practice may be lower than often assumed, and who argue that broad secularization will be unfavorable to Republicans. Americans who do not attend religious services tend “to vigorously support Democrats rather than Republicans,” including because the GOP’s occasionally religious stances and candidates can alienate less religious voters.^{Footnote 31} A 2012 Pew report found that a growing share of Americans report “none” as their religious affiliation, and that these “nones” lean heavily liberal and Democratic.^{Footnote 32} The Pew data inspired commentary, from both liberals and conservatives, proposing that Republicans face a secularization problem paralleling their ethnic-racial one.^{Footnote 33}

This interpretation presents conservatives with the same problems. Conservatives have grounded their electoral coalition in large part on religious voters, but in the process garnered a large share of a declining market. Secular voters are put off by the religious component of the Republican brand. And conservative strategists who may try to adapt to changing rates of religious practice by accommodating more secular voters could well face the same two dilemmas considered in the previous section. That is, their existing rank and file would presumably strongly oppose policy accommodations that parallel, say, the support for gay rights championed by the British Conservatives under David Cameron. And even if GOP leaders accomplished such policy shifts, they might face Przeworksi and Sprague’s dilemma of declining support on their other flank; by eroding the salience of religiosity as a basis for voting preferences, they would invite many of their existing voters to consider forming preferences on nonreligious bases, such as class. The predicted decline of religiosity thus seems to portend a diminished future for conservatism.

Surveys have certainly identified changing views on homosexuality, premarital sex, reported levels of religious belief and practice, and other issues related to religion. But there are at least two reasons to question the secularization thesis about conservatism. First, straight-line projections of past trends may be even more problematic here than in the matter of America’s ethnic-racial makeup; if anything, we are not even certain what the past trends of religious belief and practice are. Both facts and theory surrounding past secularization remain strikingly unclear. Our stereotypes about relatively high contemporary levels of practice in the modern United States may be exaggerated, which would strengthen the secularization thesis by suggesting that we underestimate how much secularization has already taken place in America.^{Footnote 34} But our stereotypes about levels of practice in the European past may also be exaggerated, which would undermine the thesis by suggesting that we overstate how much secularization has taken place there. The history of religious practice in Western Europe appears more to suggest fluctuation across time, including sometimes rising from previous lows, rather than a steady decline from a historic “age of faith” that may never have existed.^{Footnote 35} In modern America, Putnam and Campbell summarize a main finding of their ambitious study: religious practice appears to have declined only gently over five decades. Openly professed nonbelief has risen, but analyzing one major source of survey data, they conclude that if a process of secularization is taking place in America, “it will take a couple of centuries to reduce American religious observance to current European levels.” As a result, “while recognizing the slow generational decline in American religiosity over the last half century, we are skeptical about bold assertions of secularization in America.”^{Footnote 36}

The causal mechanisms that underpin the secularization thesis are at least as problematic. The version of the thesis concerned with individual beliefs has always been long on suggestive evidence—correlating growing affluence and education with declining practice in Western Europe—but short on demonstrated causal mechanisms. If we do not know why secularization has occurred where it has, what are the grounds for predicting that it will continue rather than stabilizing or even reversing over the next few decades? Evidence of dramatic recent cultural change has been offered to buttress the secularization thesis, such as rapid generational shifts toward greater tolerance of homosexuality. But this is a double-edged sword. If attitudes on these subjects are culturally constructed and can (apparently) change relatively rapidly in a given direction, what precludes rapid change in other directions? To justify an expectation of movement in only one direction, we might invoke Enlightenment platitudes about the incompatibility of modernity and religiosity, or assume that modernity induces secularism through liberal and scientific education or exposure to competing moral codes. But until the influence of these possible mechanisms has been established more effectively, these hypotheses remain what they have always been: speculative rather than strongly supported.^{Footnote 37} History, including twentieth-century history, has witnessed enough significant cultural swings over multidecade periods to suggest that insisting on unidirectionality seems analytically unjustified. Note that this is not to predict another Great Awakening or a reversal of the very gradual decline in American religious practice discernible across the postwar period. Rates of belief and practice could also fall more quickly over the next few decades. But there have been enough reversals and Great Awakenings, including ones that appear impacted by political processes, that it seems foolhardy to engage confidently in straight-line projections of continued decline.^{Footnote 38} The secularization thesis offers relatively little guidance as to what forms cultural construction will take in the next few decades.

A second reason to question the secularization thesis about conservatism is that the thesis offers even less guidance as to the even more contingent matter of how the role of religiosity will be politically constructed in the American party system. It is not clear how heavily Republican Party success relies on religiosity. Certainly religiosity has emerged as a significant contributor to party preferences in America.^{Footnote 39} But that does not mean that Republicans could not adapt to a change in levels of practice. Religious ranks have withered in Western European countries over the past half century without an evident corresponding weakening of center-right parties or strengthening of left parties. This includes Christian Democrats, who today downplay issues like divorce and abortion, but remain organizationally vibrant, electorally competitive, and quite possibly more conservative on economic issues than they were when levels of religiosity were higher.^{Footnote 40} This suggests at least that parties with religious electorates, and economic conservatives, can adapt successfully to declines in belief and practice, in no small part because the politics of religiosity are themselves a contingent product of contested political processes.^{Footnote 41}

Postmaterial Values and Postindustrial Society

Modernization is cited as a third factor undermining conservatism. The notion that increased affluence and education are broadly liberalizing has a substantial intellectual lineage. In modern American politics, Democratic activist Frederick Dutton in the 1970s emphasized the broadly liberal nature of the growing ranks of college-educated baby boomers.^{Footnote 42} Chait includes “whites with graduate degrees” in the core of the new Democratic coalition.^{Footnote 43} Judis and Teixeira identify a liberalizing trend among postindustrialism’s “highly skilled professionals.” Judis and Teixeira claim that whereas “old” professionals such as corporate managers, accountants, and financial professionals assess their performance in terms of profit, “new” professionals like teachers, architects, software designers, and doctors and nurses take pride in jobs well done, and tend to support the social safety net and public education. This worldview inclines them to support “progressive centrism” and puts them at odds with conservatism. This explains the liberalism, for example, of university towns and other research centers.^{Footnote 44} David Callahan invokes similar nonprofit motivations to explain the many progressives among America’s rich.^{Footnote 45} In political science, Ronald Inglehart makes the parallel claim that rising incomes and levels of economic security are inducing a shift in rich societies toward “postmaterial values” such as justice, aesthetics, and self-expression. In Inglehart’s interpretation, these values have broadly progressive content, such as liberalism on social issues.^{Footnote 46}

This has political implications because advanced industrial countries produce ever higher levels of education and per capita income. This means worsening prospects for conservatism, which in this view is associated with markers of low education such as racial and ethnic prejudice, anti-intellectualism, “anti-science” views, and social conservatism. In Chait’s words, the GOP has “increasingly found itself confined to white voters, especially those lacking a college degree and rural whites who, as Obama awkwardly put it in 2008, tend to ‘cling to guns or religion.’”^{Footnote 47} Inglehart has consistently proposed that the “emergence of a relatively large Post-Materialist section of the middle class offers the Left a new opportunity” and finds that postmaterialist values correlate with self-identification on the political left.^{Footnote 48} Morley Winograd and Michael Hais suggest that educated and technologically sophisticated young “millennials” will stack the electoral deck heavily for Democrats in coming years.^{Footnote 49} Judis and Teixeira predict that conservatism will not be safe even in rural areas; as they develop and attract more educated and cosmopolitan transplants, they too “will tend to become more Democratic and less Republican.”^{Footnote 50}

This line of argument is problematic for two main reasons. First, as with the racial and religious theses, the modernization thesis typically assumes the continuation of past trends. Many strands of the modernization thesis focus on education and assume continued growth in levels of formal education. Those levels have steadily grown for decades. But extrapolation may not be in order. Ranks of full-time journalists have been decimated by changes in technology and business practices. The legal profession is now being challenged by reduced demand for services and the spread of law software.^{Footnote 51} One apparent result has been sharp drops in matriculation at law schools, with likely knock-on effects for the size of law school faculties and staffs. More important, the cost of higher education has spurred movement toward competency-based certification and online education, which could substantially reduce the size of universitywide faculties and staffs. These professions, some of them the quintessentially politically progressive ones, could shrink rather than grow as shares of the workforce. Americans will almost certainly continue to increase their skills levels. But this may happen in the more diverse forms and settings permitted by competency-based credentialing, and the result might not mimic the effects of the years of formal education that produced educated professionals in past decades.

Second and more important, the modernization thesis assumes that it is natural and inevitable that higher levels of education in particular result in liberal political worldviews. In affluent countries since the 1960s, liberal-progressive views have been more fully articulated and embedded—than conservative views—in many quasi-official discourses, legislation, and the practices of diverse institutions, and became particularly pervasive in universities and mass media. This might be an inevitable feature of modernity. It might also reflect the historically particularistic intellectual climate in a specific set of countries in a specific time period. To decide which of these descriptions is more persuasive, we need to be able to distinguish inherent features of the human future from historically particularistic and contingent features of rich countries in the postwar period. It is not clear that available versions of the modernization thesis, including theorizing about postmaterialism, provide us with tools for making that distinction. This makes it difficult to disaggregate the particularistic from the permanent.

Consider policy views on economics. It appears to have been relatively conventional wisdom among highly educated Western Europeans and Americans in the 1940s–60s that desirable economic outcomes, in rich and less-developed countries alike, would result from extensive government investment, state regulation of entry and pricing in diverse sectors, and close working relationships between corporations, unions, and governments. But those views did not prove durable. Skepticism about the efficacy of those policies started to rise in the 1970s, as attitudes shifted in more market-friendly directions.^{Footnote 52} In the wake of the financial crisis of 2008–9, views on these issues may be shifting again in more state-friendly directions. Should we consider any one set of these attitudes to constitute “modern” thinking? Or does the fact that elite attitudes have mutated across time make it more plausible that attitudes, which have broadly favored the left, then the right, now perhaps the left again, are constructed through ongoing and contested political processes, and hence are contingent on an array of factors whose values are likely to change across time?^{Footnote 53}

Or consider abortion. Inglehart codes support for abortion rights as a “postmaterialist” view. It is easy to see why. Abortion rights have expanded choices for women, helping unshackle them from the reproductive roles to which they were often assigned historically. But there is an equally accessible argument to be made against abortion rights, focusing on rights deriving from the emancipatory and expressive potentialities of the unborn. Debate continues over whose rights are more salient, and whether a fetus is a person to begin with. To some, one set of answers to these questions is obviously modern. But it does not seem controversial to suggest that people’s answers to them are influenced by family and community views and teaching carried out in particular settings. Inglehart would be right to find that pro-choice views predominate in the political culture of educated elites in rich countries, particularly in the second half of the postwar period. But that predominance was constructed. It is because such views are constructed and contingent that opposition to abortion in the United States has stabilized and sometimes risen across recent decades, including among young people.^{Footnote 54} To note that pro-choice attitudes have predominated for a specific period of time is very different than demonstrating that pro-choice views represent the intrinsically obvious solution to this question for educated and affluent people and hence that such views will predictably dominate the future. If certain attitudes on these and other issues are fixed features of modernity, liberal politics could well dominate the future. If they are contested, then the future is up for grabs.

Conclusion: Contesting Future Policy Debates

These three narratives converge on the prediction that U.S. conservatism faces bleaker prospects over the next few decades than it enjoyed over the last few. Its bases are being eroded by ethno-racial diversification, secularization, or expanding numbers of educated professionals, and perhaps all three. These claims share two main assumptions: that past societal trends are likely to continue, and that party preferences and ideational orientations are to a substantial degree exogenous to politics, or at least are now fixed and exogenous to ongoing political contestation. Ethnic and racial minorities in America are assumed to be relentlessly growing in numbers and are assumed to favor liberal options given the politics of race that were created in the past. Students of secularization consider irreligiosity as matter-of-course in modern, rational, and educated societies, and assume secularism is antagonistic to modern conservatism. And theorists of postmaterialism treat left-liberal orientations as natural to the growing numbers of educated affluent. Politics may have contributed to the politics of race, religiosity, and educated affluence, but those politics are now basically set, leaving conservatism durably disadvantaged in its attempt to fight these historical tides.

This article has argued that these lines of argument are not persuasive. These demographic trends cannot safely be extrapolated. And even if they could, the politics of each issue is more plausibly the result of ongoing political contestation and construction. This matters because conservatives became more influential in American politics in no small part by mobilizing significant organizational, intellectual, and communicative resources, admittedly from the low baseline of their weakness in the shadow of the New Deal.^{Footnote 55} And it is these resources that have helped them compete in the formation of partisan and ideological identities. Conservatives are now better positioned to do so than at any earlier time. William F. Buckley Jr.—who considered it possible enough to contest the future that he founded a magazine dedicated to “standing athwart history yelling stop”—in 1999 noted the contrast between conservative capabilities earlier and later in the postwar period: “With distressing frequency, those who upheld the [conservative] heritage left the stage or studio exposed to humiliation by the poverty of their resources [for debate]. It’s different now.”^{Footnote 56} More succinctly, Trende notes that “[p]ermanent majorities simply do not exist in American politics.”^{Footnote 57} This is in large part because individuals and political sectors are free to adapt and compete. In this case, conservatives are free to adapt to changing demographics and to contest the politics of ethnicity, religiosity, and other factors.