Trumpism and its Discontents

The following is a chapter from Trumpism and its Discontents. Click to download a PDF of the book here.

By Cybelle Fox 

 

Image

The face of America is changing. In 1965, the United States was 84 percent white. Today, non-Hispanic whites represent closer to 60 percent of the population, and if current trends continue, they will make up less than half of the total population by 2050. This growing diversity has come about almost entirely through immigration from Asia and Latin America. The Asian population was just 1 percent in 1965 and could reach 14 percent by 2065. And the Latino population was just 4 percent in 1965 and could hit 24 percent by 2065. These demographic changes are largely the result of the Hart–Celler Act of 1965, which eliminated the discriminatory national origin quotas that had been in place since the 1920s.1 The United States, of course, is not the only country undergoing profound demographic shifts. In Canada and Europe migration is also reshaping the population in significant ways. 2

Ever since the late 1990s, when these demographic shifts were becoming more apparent, scholars and pundits alike have blamed diversity or changing racial and ethnic demographics for a host of societal ills. These include the decline of trust, altruism, civic engagement, social capital, and social solidarity.3   Robert Putnam famously said that, at least in the short run, “people living in ethnically diverse settings appear to ‘hunker down’—that is, to pull in like a turtle.”4 Scholars have also blamed increasing diversity for the resurgence of white nationalism,5 growing white support for conservative policies,6 the defection of whites to the Republican Party,7   the rise of right-wing populism or Far Right parties,8 the United Kingdom’s passage of Brexit,9 and the election of Donald Trump.10  

Variations on this scholarly narrative appear in the media, too. Writing for The Atlantic, Derek Thompson called this phenomenon “the doom loop of modern liberalism.” 11 In his version of this story, low birthrates lead to calls to increase immigration, and that growth increases xenophobia, raising support for right-wing parties, which in turn decreases support for the welfare state. Thompson concludes that “pluralist social democracy is stuck in a finger trap of math and bigotry, where to pull on one end (support for diversity) seems to naturally strain the other (support for equality).”12  

A strikingly different narrative, however, sees changing racial and ethnic demographics as our country’s salvation. In this narrative, expanding diversity breaks down social barriers, leading to higher rates of interracial friendships and intermarriage, and eventually blurring racial and ethnic boundaries.13 The political version of this narrative was popularized in 2002 when John Judis and Ruy Teixeira published their book The Emerging Democratic Majority.14 In it, they argue that “the country’s shifting demographics were giving rise to a strong new Democratic-voting population base.”15

Shifting demographics will enable this new Democratic coalition because over the last few decades, Asians and Latinos have joined blacks and some progressive whites to form a stable Democratic coalition.16   Estimates of the Latino and Asian American votes in the 2016 presidential election vary, but according to some estimates, 89 percent of blacks, 79 percent of Latinos, and 69 percent of Asians cast their votes for Clinton as compared with just 37 percent of whites who did.17 As the proportion of Asians and Latinos increases, the Democratic coalition will continue to expand. In the long run, this increase will spell doom for the Republican Party, especially if it continues to promote a Far-Right nativist agenda, which has pushed Asians and Latinos toward the Democrats.

Both viewpoints—diversity as responsible for all that ails us and diversity as what will save us from ourselves—treat demography as destiny. And they lead to opposing policy solutions. The first view suggests that all immigration (or perhaps especially that of low-skilled workers) should be stopped or slowed in order to save modern liberalism and liberal democracy. The second suggests that immigration should be increased or kept at the same rate while we await the inevitable demographic transition.

Demography, however, is not destiny. Neither of these perspectives—one that blames diversity for the rise of white nationalism or the other that sees diversity as the answer to all of our problems—is appropriate. The first belief, sometimes stated explicitly and at other times far more subtly, places the blame for what ails us on immigrants or people of color. Doing so, however, obscures the real sources of the problem: whites’ racism.18 Meanwhile, the second belief—the one that frames people of color as our salvation—too often obscures the importance of power: the power that white people have to either maintain the status quo or reverse the gains of the last fifty years through such means as redistricting and voter suppression—and the power of people of color (and progressive whites) to resist these shifts and to alter the poisonous political dynamics at play. Indeed, a survey of the literature in social psychology, political science, and sociology, suggests that there is ample evidence that white racial attitudes pose a significant threat to liberal democracy. There is also ample evidence—much of it based on experiments—that exposure to information about changing demographics leads whites to experience fear and a sense of a growing threat that can increase their support for the Republican Party and right-wing nationalism. But there is far less conclusive evidence that actual diversity has all the negative effects that some attribute to it. And there is also increasing evidence that changing demographics alone may not bring about a permanent Democratic majority.

Various studies have now demonstrated quite conclusively that out-group antagonism, prejudice, and ethnocentrism among white Americans pose serious problems for modern liberalism. These factors predict support for immigration restriction and anti-immigrant policies.¹⁹ They also predict opposition to welfare,20  income redistribution,21  and government-sponsored health insurance.22 And they predict support for or defections to the Republican Party,23  voting for Republican candidates,24  support for the Tea Party,25  and support for Donald Trump. Specifically, scholars have demonstrated that anti-black, anti-immigrant, and anti-Muslim attitudes were among the strongest predictors of support for Trump during the 2016 election.26  John Sides, Michael Tesler, and Lynn Vavreck argue that xenophobia helps to explain the small number of whites who voted for Barack Obama in 2012 and for Trump in 2016. The authors show, for example, that “Clinton retained almost all of Obama’s white voters with positive views of immigration. But she lost about a third of white Obama voters whose attitudes on immigration were at the negative end” of the scale.27

There is also a growing body of experimental research that examines the effects of people’s exposure to information about the nation’s changing demographics. Compared to a control group, those whites who are exposed to information about their impending minority status express greater anxiety,28 more anger toward and fear about racial minorities,29 more sympathy for whites, greater preference for racial homophily in social settings and interpersonal interactions,90 and more negative evaluations of racial minority groups.31   They also express lower support for race-related policies like affirmative action and immigration and more support for (race-neutral) conservative policies, including those related to defense spending and health-care reform.32 And they express greater opposition to government spending on welfare33 and for taxes to fund K–12 education.34 Exposure to information about changing demographics also predicts people’s greater support for the Tea Party35 and for Donald Trump.36

Although racial attitudes do matter and exposure to information about changing racial demographics can lead some whites to perceive these changes as threats to their standing and status, the effects of actual diversity or demographic change on attitudes about racial or immigrant issues, policy preferences, and support for political figures or parties are far more mixed.37   After surveying the vast literature on this topic, one finds that there is no consistent effect of diversity on social trust38 and no consistent relationship between diversity or demographic change and attitudes toward immigrants, immigration, or Latinos.39 One study, for example, found that Americans living in US counties with more foreign-born residents than others have had more positive feelings toward immigrants.40   Another found that rapid growth in the size of minority populations may fuel feelings of being threatened, but only when national rhetoric is politicizing immigration.41 There is no consistent relationship between diversity or demographic change and the adoption of anti-immigrant policies.42   There is also no consistent relationship between diversity and support for or spending on social welfare programs.43 And there is no consistent relationship between diversity or demographic change and support for Trump. Some studies suggest that it is whites who live in racially isolated communities who have the highest levels of support for Trump44 while other studies show that Latino population growth is correlated with vote-switching for Trump, especially among white working-class Democrats and Independents.45

Why do we see such mixed results in the literature on the effects of demographic context or demographic change when the literature is far less ambiguous about the importance of out-group antagonism and the effect of information about changing demographics? One reason for these mixed findings has to do with divergent interpretations of the data. Maria Abascal and Delia Baldassarri, for example, revisited Putnam’s theory about the trade-off between diversity and social trust. They argue that the negative relationship between diversity and trust that Putnam has observed is the result of a compositional effect. It is not that diversity lowers social trust, as Putnam argues. Rather, nonwhites and immigrants report being less trusting than whites are, and they live in more heterogeneous communities. In other words, the negative relationship between diversity and trust is “an artifact of nonwhites’ lower levels of trust combined with their overrepresentation in diverse communities.”46

A second reason is that proximity or exposure to racial and ethnic out-groups can stimulate a threat response among whites.47   Some whites who live near large numbers of blacks, for example, may feel a greater sense of physical, political, or cultural threat. But living among diverse neighbors can also stimulate a contact response in which attitudes toward out-groups become more positive under the right circumstances.48 Whereas the threat hypothesis assumes that prejudice is the result of real or perceived conflict over resources or status, the contact hypothesis assumes that prejudice is irrational and learned at a young age and that it thrives on ignorance. The more whites get to know nonwhites, the more they will discover that their stereotypes have been misguided, and they will adjust their beliefs accordingly.49 In other words, diversity can stimulate opposing reactions among whites. Larger group size or a sudden increase in the number of nonwhites can feel threatening to whites. But the increase in size can also increase the probability that whites will have meaningful intergroup contact, which can decrease their feelings of being threatened.50 Adjudicating between the threat and the contact response is difficult because most studies that examine the consequences of diversity rely on cross-sectional data, making causal inference difficult. There is also reason to believe that some people may choose to live or remain in more or less diverse neighborhoods based on their levels of prejudice. Experiments are useful to get around these sorts of problems, but for understandable reasons, few people would consent to being randomly assigned to live in a neighborhood just so that social scientists could properly test the effects of neighborhood racial composition. 

To get around this problem, political scientist Ryan Enos conducted an experiment. He sent two Spanish-speaking Latino confederates to sit on commuter trains coming from racially homogenous—or predominantly white—communities. He wanted to see whether white commuters would feel threatened by the presence of nonwhites on those trains. To measure the effect, he used pretreatment and posttreatment surveys to estimate the effect of contact with nonwhites on whites’ attitudes toward immigrants and other issues. He found that exposure to Spanish-speaking Latinos increased anti-immigrant attitudes among these white commuters. “Treated subjects were far more likely to advocate [for] a reduction in immigration from Mexico and were far less likely to indicate that” undocumented “immigrants should be allowed to remain in” the United States. This response was obtained after just three days of the commuters having ridden on slightly more diverse commuter trains. Interestingly, Enos also found that these effects declined somewhat over time—within just ten days, in fact—a finding that suggests that initial exposure can produce threat effects, but repeated exposure over time can “mitigate initial negative reactions.”51

A third reason is that in any given context, some out-groups appear to be more threatening to whites than others, and different studies use different measures of diversity. Combined, these two factors help to explain some of the mixed results. Some studies measure diversity by looking at the percentage of Latinos, nonwhites, or immigrants who live in a given community. Many other studies, however, use a measure of ethno-racial fractionalization. Fractionalization measures the probability that any two randomly chosen individuals belong to the same ethno-racial group. This kind of a measure treats all ethno-racial groups as functionally equivalent.52 According to this measure, a community that is 70 percent black and 30 percent white is just as fractionalized as a community that is 70 percent white and 30 percent Asian. A measure that treats all groups as functionally equivalent is problematic, however, if not all diversity is actually functionally equivalent in terms of its effects or consequences. And there is good reason to believe that it is not.53 The content of white stereotypes about Asians and blacks, for example, differs significantly. Abrajano and Hajnal note that “Asian Americans are often viewed as an intelligent, hardworking, law-abiding[,] and successful model minority.” 54 Blacks, by contrast, are often stereotyped by whites as being less intelligent, prone to violence or crime, dependent on welfare, and poor.55   

In the US context, there is very clear and consistent evidence that whites feel threatened by large black populations.56 However, the presence of large Asian, Latino, or immigrant populations—those groups that are fueling demographic change— has never produced those same consistent responses. In my own research, I found that anti-Latino stereotypes were stronger in communities with fewer, not more, Latinos.57 I also found that not all diversity has the same effect on redistribution. In research that I conducted with Irene Bloemraad and Christel Kesler, we examined social welfare spending across American states. While there was a negative relationship between diversity and social welfare spending, the results were driven entirely by black racial context.58 Some scholars have even found that larger Asian American populations are associated with more liberal rather than conservative views.59  

A fourth reason is that there is evidence that the effect of diversity may not be the same for all ethno-racial groups. In their study on diversity and social trust, Abascal and Baldassarri found that it was “only for whites” that “living among outgroup members predicts lower levels of trust.” 60 Blacks and Latinos did not have the same negative reaction to living in more diverse communities. The authors speculate that this situation is true because living in more diverse communities usually means living alongside more coethnics for racial minorities in the United States while the reverse situation is usually true for whites.

Finally, we often see mixed results for actual diversity because out-group political power is often greatest in areas where out-groups are more numerous. Abrajano and Hajnal found that “in states with larger Latino populations, public goods provision drop[s] significantly, and funds for welfare, health[,] and education all decline.” But “once the Latino population passes a threshold[,] . . . policy outcomes become more pro-Latino.”61   Political power matters because it gives out-groups a voice in the policy-making process and in the selection of political leaders. They can elect their own to public office, thereby increasing descriptive representation. They are also more likely to have their voices heard, especially in competitive electoral districts where every vote counts.

Instead of fretting over immigration levels, then, we would be better off by first addressing white bias. This is a point that Abascal and Baldassarri make: if we think that diversity is the problem, then we should aim for social policies that preserve or promote homogeneity. But if the real problem is white bias, not diversity, “then policy makers should contemplate group-specific interventions.” 62 To be sure, reducing or combating white racial bias will not be easy,63 but it is necessary work. “The only way to make progress on racial issues,” explains noted racial attitudes scholar Lawrence Bobo, “is to face them directly and honestly.”64 One way to do so is to call out racism where it exists. We know that white racial attitudes can be primed by news stories or campaign advertisements that contain implicit racial appeals.65 But when the subtext of coded racial appeals is made plain, white voters who adhere to a norm of racial equality tend to reject such appeals.66

Second, the literature suggests that information and framing matter and that we can alter the way information about immigration and demographic change is framed to “reduce its most divisive effects.”67 For example, if one tells whites that, despite demographic change, their status in the racial hierarchy is likely to remain unchanged, then the effects of information about changing demographics disappear.68 Such framing devices may be less effective if and when real efforts are made to dismantle racial hierarchies and inequalities.

Third, elites matter. Although context and social geography may play a role, they are not deterministic.69   Elites, including politicians and the media, help make some group differences salient in certain times and in certain places.70 Abrajano and Hajnal find that the tone of media coverage on immigration—including that in liberal outlets like the New York Times—is largely negative and focused on Latinos, helping to create and sustain a narrative of immigrant threat.71 This media-generated threat narrative affects partisan beliefs among the US public, lowering the percentage of Democratic identifiers. Similarly, René Flores’s work suggests that anti-immigrant laws can have important symbolic effects, increasing anti-immigrant behaviors and hardening group boundaries.72

Fourth, there is some evidence that promoting integration and intergroup contact can be helpful.73 Enos argues that segregation is an important dimension of social geography that affects intergroup relations. Segregation makes intergroup attitudes “more negative and their political consequences . . . more severe.”74 Promoting integration and intergroup contact, then, may help mitigate some of the tensions that can arise in diverse societies.

Last, increasing the political power of people of color is critical for demographic shifts to have any real effect on the political situation.75   This is especially important if Republicans are able to effectively suppress the minority vote through voter ID and felon disenfranchisement laws, voter roll purges, or redistricting.76   Asians and Latinos are much less likely to vote than whites and blacks are, however.77 Part of this voting gap is due to either lack of citizenship or the age distribution of the Latino population in particular—that is, many Latinos are still too young to vote. Consequently, although the United States might become a majority–minority nation by 2043, “the share of the eligible electorate that is minority lags behind the estimates for the entire population” by at least a decade and a half.78   But even controlling for citizenship and age, Asians and Latinos are still less likely to vote.79  

As a result, progressives cannot wait for demographic change to generate political change. They need to build movement power, sponsor naturalization drives, support automatic voter registration laws, and promote voter mobilization efforts to speed things along. Officials need to do what they can to reduce language barriers to political participation. Political parties and civic organizations need to do more targeted outreach to minority communities to increase their voter participation rates.80   Parties will also need to run candidates that speak to the concerns of people of color if they want them to turn out to vote or to vote for their candidates.81   Minorities are more likely to turn out if they believe that their vote matters. Absent changes that decrease the “minority–white turnout gap,” “non-Hispanic whites will continue to be a dominant force across the country for decades to come.”82   California may be instructive here. As Allan Colbern and Karthick Ramakrishnan explain, for most of its history, California was a leader in anti-immigrant mobilization and policy adoption.83 Anti-immigrant mobilization in California helped produce the Chinese Exclusion Act and other immigrant exclusions.84 Local communities in California led the way in repatriating more than a hundred thousand Mexicans and Mexican Americans during the Great Depression.85 In the 1990s, voters passed Proposition 187, which barred undocumented immigrants from most nonemergency services and required cooperation between service providers and immigration officials. Although the measure was blocked by a federal district court before it was ever implemented, it is widely credited with helping to inspire anti-immigrant legislation in other states and at the federal level in the decades that followed. And when Bill Clinton first ran for president in 1992, he could not count on California to vote for the Democratic candidate. Indeed, between 1952 and 1988, California had only once voted for a Democrat for president, choosing Lyndon Johnson over Barry Goldwater in 1964.86 Fast-forward to 2018, and California has styled itself as a leader of the Trump resistance. It is now a safely blue state, and one of the most progressive states in the country for immigrant rights.87 Republicans, the Sacramento Bee reported in 2018, now rank third in the state among registered voters, behind both Democrats and those who have no party preference.88

This change in California did not come about simply through a demographic transition. A couple of factors appear to have been critical for the shift in politics in the state. The first is the increasing political power of the Latino community.89 Latinos naturalized in large numbers, registered to vote, and elected their own to public office, actions that helped to give Democrats a supermajority in the state legislature. Latinos were motivated to do so partly because of their sense of being threatened as a result of a series of ballot initiatives that targeted immigrants and racial minorities more generally,90 as well as by the Republican Party’s hard-right turn on immigration-related issues.91  

Second, advocates were able to capitalize on that sense of threat because they organized. Although advocacy groups were not able to thwart the passage of these ballot propositions, many of these groups chose to play the long game and to prioritize increasing immigrant political power instead.92 With the support of philanthropic  organizations, labor groups, multi-racial coalitions, and immigrant rights organizations built their organizational capacity, honed their protest and lobbying skills, and pushed state officials to pass more pro-immigrant policies.93   They sponsored naturalization campaigns, voter registration efforts, and get-out-the-vote drives.94 By doing so, they helped to change the politics of the state, altering the incentives of politicians who ran for office there.

Third, some white voters also changed their views. Shaun Bowler, Stephen Nicholson, and Gary Segura argue that in addition to alienating Latinos, Propositions 187 and 227, which limited bilingual education, seemed to have alienated some Anglo voters, too, who came to regard the propositions as racist. This development in turn decreased their desire to identify with the Republican Party. “The cumulative effect” of the initiatives was “a 7.4 percent decline in Republican identification among whites.” 95 This shift among whites, the authors argue, was strongest for “new voters” under age thirty, who were “forming partisan attachments during this period” and who “perceived the Republicans as antiminority and thus moved toward the Democrats.”96

Ultimately, change came to California not from limiting immigration or simply waiting for a demographic transition. Change came from broad-based organizing and from the increasing political power of the Latino community. It was spurred along, too, by those young white voters who chose to reject racism and the politics of hate.