The success of Donald Trump's anti-immigrant campaign surprised many. But I show that it was actually a continuation of a long-standing Republican strategy that has targeted immigrants and minorities for over five decades. It is not only a long-term strategy but also a widely successful one. Analysis of the vote over time shows clearly that White Americans with anti-immigrant views have been shifting steadily toward the Republican Party for decades. The end result is a nation divided by race and outcomes that often favor Whites over immigrants and minorities.

“They're sending people that have lots of problems, and they're bringing those problems with us. They're bringing drugs. They're bringing crime. They're rapists. And some, I assume, are good people.”

With these now infamous lines about Mexican immigrants, President Trump appeared to set in motion his meteoric rise in the 2016 presidential campaign. Before giving that speech, Trump was floundering. Polls placed him near the bottom of the sixteen-candidate Republican field. But just a month later – after almost nonstop coverage of his immigration remarks – Trump had skyrocketed to first place in the polls. In the primary, Trump won over Republican voters who wanted to deport unauthorized immigrants, and he lost decisively among those who favored a pathway to citizenship. Indeed, immigration appeared to fuel his candidacy all the way through the general election. Three-quarters of Trump voters felt that illegal immigrants were “mostly a drain” on American society. Only 11 percent of Clinton supporters agreed.1

Trump's focus on immigration and the tight link between immigration views and the vote in 2016 raise a series of important questions. First, where did the immigration threat strategy come from? Was Trump's strategy unique and the course of 2016 exceptional, as many media accounts seem to suggest, or was 2016 simply an extension – albeit a more explicit and more extreme one – of a longer-term Republican project? Second, is there evidence that an immigrant threat narrative has actually propelled voters into the welcoming hands of the Republican Party? Although the close correlation between how Americans think about immigration and how they voted in 2016 suggests that immigration matters, and perhaps even that it is central to the partisan politics of this nation, we know that correlation is not causation. Finally, what are some of the major consequences of the increasingly central role of immigration in American politics?

Although many pundits and prognosticators were surprised by Donald Trump's tactics and his triumphs, it was all quite predictable. Trump's use of the immigrant threat narrative is a tried and true strategy. Well before Donald Trump arrived on the presidential scene, Marisa Abrajano and I wrote a book documenting the long-standing Republican tactic of scapegoating immigrants. In White Backlash, we argue that Republican elites had been able to garner more and more of the White vote by blaming immigrants for much of what ails America and by promising to stem the tide of immigration.2

The evidence of that long-term Republican strategy is extensive. It begins most conspicuously in California in 1994 when Pete Wilson, the Republican governor, campaigned on Proposition 187 to help counter his low approval ratings and sagging poll numbers. The proposition, which was nicknamed the “save our state” measure, sought to bar all undocumented immigrants from receiving public services. Campaign ads featuring grainy footage of immigrant hordes crossing the border, while a narrator intoned “They keep coming…” would become a model for subsequent Republican campaigns.

When Wilson won reelection using that strategy, Republicans around the country slowly took heed. In the ensuing decades, elites in both parties have expressed a variety of views on immigration, but the growing distance between the two parties on immigration is clear. Much of the early activity occurred at the state level with Republican-led state legislatures around the country passing thousands of laws that explicitly limited immigrants' rights or services.3 Perhaps the best-known example of these anti-immigrant laws is Arizona's sb1070, passed in 2010, which allowed police officers to target individuals suspected of being undocumented, prohibited unauthorized immigrants from applying for work, required individuals to carry their alien registration cards, and permitted warrantless arrest in cases involving probable cause of a deportable offense.

Many local Republican officials also clearly moved to the right on immigration. That movement was epitomized by Joe Arpaio, the former Republican Sheriff of Maricopa County, Arizona, who proudly proclaimed that “Nothing is going to stop me from cracking down on illegal immigration.” By 2008, the issue was receiving more prominent attention at the national level and even Mitt Romney, a member of the more moderate wing of the Republican Party, was including self-deportation and opposition to the Dream Act as part of his presidential platform. As one of his ads announced, “As President, I'll oppose amnesty, cut funding for sanctuary cities, and secure our borders.”

These increasingly divergent stances on immigration are borne out by votes in Congress. As political science researchers Gary Miller and Norman Schofield have demonstrated, Republican support for immigrants' rights was reasonably strong during the Reagan era and, as late as 1990, immigration-related legislation generated little noticeable partisan division. In fact, Ronald Reagan signed a law that granted amnesty to almost three million undocumented immigrants. But since that time, votes in Congress have revealed an increasingly stark contrast, with Republican legislators repeatedly supporting tougher laws against immigrants and Democrats favoring more admission and greater immigrants' rights.4

Political scientist Tom Wong has found that between 2006 and 2012, Republican House and Senate members favored restrictive policies 98.4 percent of the time, while Democrats supported those measures only 66.4 percent of the time.5 On any number of different immigration-related issues – erecting border fences, English as the official language, amnesty, government workers reporting undocumented immigrants, and so-called anchor babies (the U.S.-born children of immigrants) – Republicans and Democrats are increasingly on opposite sides of the immigration debate.

The strategy may have reached its apex in 2018 with Donald Trump's explicit comments about Mexican immigrants – “These aren't people. These are animals” – but decades of Republican campaigns have developed and proliferated the strategy; Trump is only continuing it. Despite the ubiquitous talk of Trump being extraordinary, the truth is, the patterns in 2016 mirror decades of American campaigns and elections.

All of this reprises a very old and quintessentially American story on immigration. America may be a nation of immigrants, but it has not always welcomed immigrants with open arms. Immigration has often sparked widespread fear and mobilization, especially when the number of new arrivals has been large, or when the makeup of new Americans has differed from the native born in obvious racial or ethnic ways.6 Indeed, the history of the nation can be told through a series of challenging immigrant-nativist confrontations. The rising tide of German and French migrants at the end of the eighteenth century sparked one of the first large-scale nativist movements. Numerous episodes followed: anti-Irish discrimination in the 1850s, a populist backlash against Chinese immigrants in the 1880s, prevalent anti-Southern and Eastern European sentiment in the early twentieth century, the internment of Japanese Americans during World War ii, and a backlash against Muslim Americans following September 11.7

Critically, with each wave of immigration to American shores, savvy politicians have attempted to use anxiety about immigration to garner votes. As a result, many of these nativist episodes were shaped by and had a real impact on the partisan politics of the day. The electoral advantage of immigration often accrued to the party – new or old – that most vociferously opposed immigration. In the 1850s, for example, a nativist backlash against Irish Catholic immigrants helped spur the Know Nothings and the American Party to electoral success.8

Ultimately, neither concerns about immigrants nor political parties seeking to gain from those concerns are new. Trump and the Republican Party of today are just one example of a recurring, longer-term phenomenon. It is, of course, also important to note that this is not a uniquely American phenomenon. Trump's rise is analogous to the rise of the extreme right in Europe and mirrors the success of the uk Independence Party in Britain, the Freedom Party in Austria, and the National Front in France, among many others.

But the Trump phenomenon and the larger Republican campaign are not just about immigration. The anti-immigrant story is only part of the White backlash story. Race, more broadly speaking, has been part of the Republican playbook for quite some time.9 A little over five decades ago, the Republican Party implemented its infamous Southern Strategy. Personified by George Wallace's segregationist rhetoric, the Republican strategy was to dismiss Black demands for justice as requests for ever-greater government handouts and to highlight the failings of the Black community in order to attract racially conservative White Southerners who had up to that point faithfully supported the Democratic Party. Through Gold-water, Nixon, Reagan, and onto George H. W. Bush, the campaign tactics were sometimes subtle and sometimes not so subtle. While Wallace would proclaim “Segregation now, segregation forever,” a conservative political action committee supporting George H. W. Bush more delicately ran an ad about Willie Horton, an African American felon, to stoke fear of Black crime. Almost always there was a hint of race in the air and at least an implicit denigration of African Americans.

For White Southerners, it was all too attractive. White Southerners who overwhelmingly sided with the Democratic Party in 1960 overwhelmingly voted for Republican candidates in 1990. And it was not just White Southerners. Since 1990, racial views and partisanship have only become more intertwined at the national level. For much of this recent period, racial resentment has been one of the strongest predictors of party affiliation.10 Barack Obama's presidency only increased the importance of racial views. As political scientist Michael Tesler has so aptly demonstrated, how people think about health care and a host of other ostensibly nonracial issues is now highly correlated with their racial views.11

All of this has fed back into the Trump phenomenon. Research during the 2016 primary campaign showed that White independents and Republicans whose racial identity was important to them were more than thirty points more likely to support Trump than those who did not think their racial identity was important.12 Another study found that racial resentment, more than populism or authoritarianism, determined who supported Trump and who did not in the general election.13 In short, Trump's rise is neither surprising nor unusual. It is a logical outgrowth of decades of a Republican strategy on immigration and race.

Superficially, at least, the strategy seems to be incredibly successful. Study after study has now demonstrated a close relationship between how Americans think about immigration and how they vote. My own research shows that this relationship was already firmly in place in 2008 when Barack Obama ran for president for the first time. Americans with the most positive views of undocumented immigrants tended at that time not to vote Republican for president or Congress. Only 18 percent did so for president and only 23 percent did so for Congress. By contrast, a clear majority of those with negative views of undocumented immigrants favored Republican candidates in 2008: 68 percent in congressional elections and 77 percent in the presidential election.

And the relationship only becomes tighter over time. In 2016, as I have already noted, 76 percent of Americans who thought the government should identify and deport undocumented immigrants supported Trump, whereas 77 percent of those who disagreed voted for Clinton.14 Further analysis shows that views on immigration were equally closely linked to the congressional vote, the gubernatorial vote, and the state legislative vote that year.15 By the 2018 midterms, there was an almost perfect correlation between immigration and the vote. Almost everyone (91 percent) who opposed granting legal status to people brought into the country as children voted for Republican candidates for Congress, while almost everyone (92 percent) who supported granting legal status to the same immigrants voted Democratic.16

But before making causal claims about that relationship, we need to consider the possibility that the link between views on immigration and partisan choice is spurious: a by-product of a connection with one or more other factors such as attitudes on war, the economy, terrorism, gay rights, or race. Any number of the issues on which the two parties have squared off could be driving the link between immigration and party.

One empirical strategy to interrogate the independent effect of immigration on the vote is to control for other factors that might impact it. For that analysis, Michael Rivera and I considered a wide range of positions on other issues, such as attitudes toward racial and ethnic minorities, perceptions of the economy, partisanship, ideological position, demographic characteristics, and just about everything else we think matters in presidential or congressional elections.17 Performing that analysis on the 2008 presidential election, we found that even after taking into account all of these different factors, how White Americans think about immigrants is still strongly related to how they vote. In the 2008 presidential contest, Whites with negative views of immigrants were – all else equal – 24 percent more likely to vote for John McCain than for Barack Obama. Views on immigration mattered to a striking degree, eclipsing other issues in an election taking place against the backdrop of one of the nation's sharpest recessions, ongoing wars in Afghanistan and Iraq, and with the nation's first Black presidential nominee on the ballot. The relationship between immigration and partisanship is not just a spurious one.

That analysis revealed one other important aspect of America's views on immigration, and in particular how those attitudes are and are not related to views on race. The 2008 data show that, in many ways, race and immigration are connected. How Americans feel about Blacks and how Americans feel about immigrants are related. Those who are anxious about immigration also often resent demands made by African Americans. I suspect that a lot of these feelings toward both groups have the same roots. Indeed, studies have shown that both attitudes on race and attitudes on immigration are closely linked to deep-seated psychological predispositions such as authoritarianism, intolerance, and ethnocentrism.18

However, immigration and race represent distinct dimensions, as the data make clear. Attitudes on race and immigration are correlated, but the correlation is not all that strong. In this particular case, the correlation between racial resentment and anti-immigrant attitudes is just 0.28, meaning that relatively little of one attitude can be explained by the other. Even more important, the fact that immigration predicts the vote even after taking into account racial views indicates that immigration has an impact beyond race.19 Further, the fact that we found that the size of the anti-immigration effects is roughly on par with the effects of racial attitudes suggests that immigration represents not only a distinct dimension of American politics but an important one as well.

But before one can be absolutely confident that attitudes on immigration are actually driving party identification and the vote, one more test is needed to rule out the possibility that party identification is itself the main driver of change. Individual Americans could be taking cues from partisan leaders, adjusting their stances on immigration to match those of a party that they know, trust, and believe in. Party identification, then, could be driving immigration attitudes, rather than the reverse.

We can examine this issue of “what causes what” by analyzing the same individual's views at different points over time. The key test is whether an individual American's position on immigration at one point in time shapes future changes in that individual's partisanship. That is, can we accurately predict who will shift to the Republican Party in the future based only on how those people think about immigration today? For these causality tests, we focused on panel data from the American National Election Study, which repeatedly asked the same respondents for their views and partisanship. Based on these tests, it is clear that how an individual thinks about immigration at one point in time predicts how their partisanship will change in the future. To be sure, the effect is not large. Over the course of a single year (in one case from 2008 to 2009), those with more negative views on immigration shift about one-quarter of a point more to the right on a seven-point party identification scale.20 But if these small shifts accumulate over time, they could help to account for large-scale partisan changes. Others have likewise found that attitudes on immigration help to predict which Americans would ultimately shift their votes from Obama in 2012 to Trump in 2016.21

We also find that views on immigration predict future shifts in partisanship at the aggregate or national level. Analyzing national patterns in aggregate White partisanship – the relative share of Democrats and Republicans in the White population – over the last two decades, we found that aggregate views on immigration at one point in time predict changes in aggregate White partisanship in subsequent periods.22 The size of the effect is again far from massive, but it is meaningful. A standard deviation shift in support for immigration is associated with a little less than one-tenth of a point shift on the five-point partisanship scale.

There are other signs as well. When we look at the media, we see yet more evidence of the power of immigration. Specifically, we find that increases in negative coverage of Latino immigration by the media are correlated with shifts in aggregate White partisanship toward the Republican Party. Analyzing three decades of New York Times coverage of immigration, we find that the tone of that coverage is overwhelmingly negative: there were four times as many negative news stories on immigration as there were positive news stories. Even more important, we find that the more negative stories focused on Latino immigrants in one quarter, the more Whites identified with the Republican Party in the next quarter. In this case, the effects were substantial. In the analysis, shifts in media frames on immigration had just as much impact on future partisanship as perceptions of the state of the economy and presidential performance.

Over time, all of this is likely to add up to major changes in the partisan leanings of the nation – or at least of the White population. As Figure 1 shows, during this period of Republican anti-immigrant tactics, there has been a slow, sometimes uneven, but also very clear movement of Whites toward the Republican Party and its candidates. In 1990, before Republican candidates had embarked on the immigrant threat narrative, White voters were almost evenly divided in their support of Democratic and Republican congressional candidates, and there was almost no correlation between attitudes on immigration and White partisanship. In 2016, after years of Republican campaigning against immigrants, views on immigration were tightly linked to the vote and Whites had become decidedly Republican in their congressional choices. In 2016, only 38 percent of White voters favored Democratic candidates in congressional contests. In 2020, the number was only 41 percent. Trump simply represents the apex of a long-term anti-immigrant backlash strategy.

Figure 1

Declining White Support for House Democrats

Figure 1

Declining White Support for House Democrats

Close modal

There is little doubt that many factors are contributing to the defection of White America from the Democratic Party. But one can make a plausible case that a backlash to immigration is helping drive this most significant development in American party politics in the twenty-first century. The striking feature of the empirical patterns here is not that immigration matters. U.S. history amply demonstrates that many White Americans have felt threatened by different racial/ethnic groups at various times. The arresting feature is, instead, just how wide-ranging those effects remain today. In a political era in which many claim that the significance of race has faded, immigrant-related views impact the political orientation of many members of the White population. Party identification – the most influential variable in U.S. politics – is at least in part a function of the way individual White Americans see immigrants. So, too, is the vote in national contests for president and Congress. In short, the immigration backlash is real and it is powerful.

The successful efforts of the Republican Party and in particular Donald Trump to bring the issue of immigration to the center of American politics have had major consequences for immigrants, for our polity, and for our nation as a whole.

The most obvious consequence of the increasing centrality of immigration in our politics is the lack of progress on immigration policy. In spite of the fact that large segments of the American public hold fairly positive views of immigrants, and the fact that a majority of Americans seem to be sympathetic to at least subsets of the immigrant population (such as Dreamers), there has been almost no movement forward on immigration policy in the last few decades. Prior to Donald Trump's presidency, the federal government basically experienced a multidecade stalemate on immigration, passing few or no major policy initiatives. In fact, one could argue that the Republican focus on the costs of immigration and the heightened anxiety that the immigrant threat narrative has produced has led to significant regression in terms of immigration policy. In particular, the Trump administration repealed the Deferred Action for Childhood Arrivals (daca) program,23 eliminated Temporary Protected Status for tens of thousands of Central Americans, severely limited the number of refugees allowed into the country, increased the number of apprehensions at the border (particularly by targeting families), widened the scope of deportations, spent considerable funds to extend the border wall with Mexico, prevented asylum seekers from entering the country, and, at one point, banned almost all legal immigration to the United States.

Critically, shifts in policy have not only occurred at the federal level. As I already noted, in the last two decades, state legislatures have passed over three thousand laws that dealt explicitly with immigration or immigrants. Moreover, the clear majority of these substantively significant laws have served to limit rather than expand immigrants' rights or interests.24 During this period, states have done everything from reducing or eliminating immigrants' access to public services in education, health, and welfare, to allowing the police to target individuals suspected of being undocumented. Unfortunately, for immigrants themselves, the states with the largest Latino populations have been the most active and the most aggressive. Texas, a state with one of the highest shares of Latino and undocumented residents, passed seven anti-immigrant laws between 2007 and 2009, including measures to detect and deter undocumented immigrant use of state Medicaid, to reduce eligibility for the state's Children's Health Insurance Program, and to require private companies that work with the state to demonstrate that they do not employ unauthorized workers.

But Texas is not alone. Arizona, a state in which Latinos make up 30 percent of the population, passed twelve anti-immigrant measures over that same time period. Arizona's 2010 efforts included the passage of the well-known sb1070, one of the strictest anti-immigrant measures ever passed. Over the same time frame, Colorado, likewise, ushered eleven anti-immigrant bills into law, including one requiring that employers be notified of the prohibition against hiring an unauthorized alien, and another that tied unemployment insurance benefits to citizenship.

Further analysis in White Backlash indicates that the backlash is not confined to measures that explicitly mention immigrants or immigration. The immigrant threat narrative has been so pervasive that it has crept into debates about policy issues that are ostensibly not about immigration. Public discussions related to welfare, health, education, criminal justice, taxes, and many other subjects have been infused with images and stories of the undocumented and the heavy economic, cultural, and criminal costs that these immigrants put on American society.

The net result is that state policy across a host of different arenas has become intricately connected to the immigrant population. In particular, how states raise and spend their money is closely linked to the size of the immigrant population. Table 1 demonstrates this relationship at its simplest level. I compare basic state policy in heavily Latino states to policy in states with smaller Latino populations. I focus on the size of the Latino population because I believe that the broader Latino population is the most visible shortcut for the immigrant population in the minds of many White Americans.25 Indeed, there is plenty of evidence that when White Americans think about immigration, the image they have in their head is an undocumented Latino.26

Table 1

Government Policy in Heavily Latino States Is More Regressive (Share of All State Spending)

States with a Small Latino PopulationStates with a Large Latino PopulationProportional Difference in Spending
Health Care 3.7% 2.5% −32% 
Corrections 3.9% 4.7% + 21% 
Education 25.8% 24.8% −4% 
Sales Tax 27.5% 36.4% + 32% 
Property Tax 5.8% 1.3% −78% 
States with a Small Latino PopulationStates with a Large Latino PopulationProportional Difference in Spending
Health Care 3.7% 2.5% −32% 
Corrections 3.9% 4.7% + 21% 
Education 25.8% 24.8% −4% 
Sales Tax 27.5% 36.4% + 32% 
Property Tax 5.8% 1.3% −78% 

Source: Annual Survey of State and Local Government Finances, U.S. Census.

As Table 1 reveals, larger concentrations of Latinos tend to be associated with state-level policies that are more regressive, more punitive, and less generous. Redistributive spending – money for health care and education – is lower in states where the beneficiaries of the policies are more likely to be Latinos.27 In the case of health care funding, the gap is sizeable. Medicaid spending drops 32 percent in heavily Latino states. The pattern is the opposite for punitive criminal justice spending. In states where Latinos represent a large share of the population and could be the target of tougher laws and harsher sentences, spending on prisons is substantially higher. Again, the absolute difference in the share of the budget going to prisons is small, but the gap represents a 21 percent increase in the share of the budget going to prisons. Critically, these relationships persist in regressions after controlling for a range of other factors that could be driving spending patterns. As the Latino population grows, Americans become less willing to invest in public services like education, health, and welfare, and are more willing to fund prisons. In other words, when the policy is more apt to impact Latinos, benefits decline and punishment increases.28 All of this indicates that America's increasingly diverse population is generating a real, wide-ranging backlash. It also means that as immigration has become more central to our politics, immigrants have been the ones who lose most.

Another obvious outgrowth of the Republican Party's heavy focus on immigration is electoral success. The immigrant threat narrative may not win extra votes in every election or in every context, but the Republican Party's shift to the right on immigration has almost certainly contributed to their largely winning the electoral battle against Democrats. Since the mid-1990s, when Republicans began their immigrant threat narrative campaign, the Republican Party has gone from being essentially shut out of government to having a relatively dominant position. In the mid-1990s, Democrats controlled the presidency, House, Senate, and the majority of state governments. As I write at the end of 2020, Republicans control the presidency and the Senate, they have solidified a conservative majority in the Supreme Court and recast the federal and appeals court judiciary, they occupy the majority of the governor's mansions, they control a majority of state legislatures, and they are the majority party in a majority of states. Many different factors have contributed to this decades-long partisan shift, but the willingness of Republican leaders increasingly and vocally to embrace an anti-immigrant narrative has rewarded the Party with a larger and larger slice of the White vote and widespread electoral victories from the local to the national level. Donald Trump, perhaps more than anyone else, knows this. As he told The New York Times Editorial Board: “I just say, ‘We will build the wall!’ and they go nuts.”

Whether the immigrant backlash strategy will continue to produce political victories in the future, as the racial and ethnic minority population continues to grow, is another question altogether. But it is worth noting that there are still many White Democrats with relatively racially conservative or anti-immigrant views. And when informed that the United States is poised to become a majority-minority nation in the middle part of this century, experiments show that White views tend to shift even further to the right.29 Given that White Americans still represent over 70 percent of all voters, and still more White voters could defect from the Democratic Party, Republicans could potentially reap the benefits of an anti-immigrant narrative long into the future.

A less obvious but equally important consequence of the immigrant threat campaign being waged by the Republican Party is an increasingly racially divided electorate. Just as many Whites have been attracted by the anti-immigrant rhetoric of the Republican Party, many racial and ethnic minorities have been repelled. Over the last few decades, more and more racial and ethnic minorities have entered the country, more and more have become engaged in the political arena, and perhaps most important, they have spoken with an increasingly clear political voice. In particular, the last three decades have witnessed a dramatic shift in Asian American partisanship. In the early 1990s, the Republican Party held a slight edge among Asian Americans, but by 2018, the number of Asian Americans who identify as Democrats outnumbers the number of Asian American Republicans two-to-one.30 Movement among the Latino is less obvious, but what is clear is that both Latinx and African American voters remain firmly entrenched in the Democratic Party. Two-thirds or more of the Latinx vote typically sides with the Democratic Party. For African Americans, the figure is generally closer to 85 percent.

All of this is readily apparent in Figure 2, which illustrates the changing racial composition of the two major parties over time. The White share of Democratic Party votes has declined sharply since the 1960s. As the population has become more diverse, and as more minorities have shifted to the Democratic Party, the Democratic base has become more diverse. Today, a little fewer than half of Democratic voters are non-White. By contrast, Republicans have remained steadfastly White despite the increasing diversity of the nation. Almost all of the votes that Republican candidates receive now come from White voters. Nearly 90 percent of the vote that McCain won in 2008, that Romney won in 2012, and that Trump garnered in 2016 and 2020 came from White Americans. The Republican Party is for almost all intents and purposes a White party. Politics in America is not perfectly correlated with race, but it seems to be deeply and increasingly intertwined with it.

Figure 2

The Changing Racial Composition of the Democratic and Republican Parties: Non-White Share of the Presidential Vote by Party and Year

Figure 2

The Changing Racial Composition of the Democratic and Republican Parties: Non-White Share of the Presidential Vote by Party and Year

Close modal

The end result is that American democracy is now divided more by race than any other demographic factor. The centrality of race for the vote was evident in 2016 and is illustrated in Figure 3, which provides a snapshot of the roles of race, class, religion, and other factors in American democracy. Specifically, the figure shows the gap between different kinds of groups in the likelihood of voting for the Democratic candidate, Hillary Clinton. The gap between Whites who gave only 37 percent of their votes to the Democratic candidate and African Americans who gave 89 percent of their votes to the same candidate was a whopping 51 points. That is more like a racial chasm than a racial gap. The gap between Whites and Asian Americans was a robust 42 points; between Whites and Latinos it was a substantial 38 points.

Figure 3

Race, Class, Religion, and Other Demographic Divides in the 2016 Presidential Contest

Figure 3

Race, Class, Religion, and Other Demographic Divides in the 2016 Presidential Contest

Close modal

Those racial gaps far outweighed any of the class divides that the exit polls recorded in 2016. The gap by income (seven points), education (four points), and union membership (five points) all fall far short of the racial divides. Critically, it is important to note that the effects of different measures of class work in different directions in 2016. Wealthier Americans are more likely than poorer Americans to vote Republican – a pattern that aligns with traditional class-based theories of American politics. But Americans with postgraduate degrees are actually less likely than Americans with a high school diploma to favor the Republican candidate. In other words, increased class status is sometimes associated with the political left, and sometimes associated with the political right; this pattern repeats across elections and time, not just in 2016. If I try to add all of the effects of class together by comparing the votes of wealthy, well-educated, full-time workers to the votes of lower-income, unemployed, high school dropouts, I find that class plays almost no net role in the vote.31 Because some class-based factors pushed toward Trump, and others led toward Clinton, the net effect of being high class was only four points. Not only does race have a larger impact than class when it comes to Americans' political diversity, it also has a much clearer and more consistent impact. Of course, as the media has repeatedly highlighted, there are growing class divisions within the White population, but that does not negate the fact that race more than class shapes the overall electorate.

The 2016 racial divides also dwarf divisions by gender (thirteen points), age (eleven points), marital status (eleven points), or military status (sixteen points). The only factor that begins to rival race is religion. The gaps between Protestants and atheists (thirty-one points) and between Protestants and Jews (twenty-five points) are both quite substantial, but fall somewhat below all of the White-non-White divides.

The electoral story in 2016 is one in which race was central; but the story in 2016 is also not unique. The numbers for 2020, which are still coming in, reveal a strikingly similar story. Looking across an array of other recent elections, I found remarkably similar electoral patterns from the national to the local level. In most electoral contests, American politics today pits the White majority against the bulk of the racial and ethnic minority population.

Politics is bound to create division, but when those divisions so closely mirror racial and ethnic identity, the situation is troubling. With race and party so closely matching each other, it is perhaps not surprising that hostility between Democrats and Republicans is increasing. Today, Americans tend to view fellow partisans as patriotic, well-informed, and altruistic, while they tend to attribute the opposite characteristics to members of the opposite party. Experiments reveal that partisan division has become so heated that Democrats and Republicans now regularly and openly discriminate against each other.32 When our political dividing lines begin to look a lot like a racial census, larger concerns about inequality, conflict, and discrimination emerge, and we are in danger of being driven apart.


The Democracy Fund Voter Study Group, 2016 VOTER Survey Full Data Set, June 2016,


See Marisa Abrajano and Zoltan Hajnal, White Backlash: Immigration, Race, and American Politics (Princeton, N.J.: Princeton University Press, 2015).


See Michael Rivera, The Determinants of State Immigration Policy (Ph.D. diss., University of California, San Diego, 2015).


See Gary Miller and Norman Schofield, “The Transformation of the Republican and Democratic Party Coalitions in the U.S.,” Perspectives on Politics 6 (3) (2008): 433-450.


See Tom Wong, The Politics of Immigration: Partisanship, Demographic Change, and American National Identity (New York: Oxford University Press, 2017).


See Peter Schrag, Not Fit for Our Society: Immigration and Nativism in America (Berkeley: University of California Press, 2011).


See John Higham, Strangers in the Land: Patterns of American Nativism: 1860-1925, 2nd ed. (New York: Atheneum, 1985).


See Tyler Anbinder, Nativism and Slavery: The Northern Know Nothings and the Politics of the 1850s (New York: Oxford University Press, 1992).


See Thomas Byrne Edsall and Mary D. Edsall, Chain Reaction: The Impact of Race, Rights, and Taxes on American Politics (New York: W. W. Norton and Company, 1991). See also Edward Carmines and James A. Stimson, Issue Evolution: Race and the Transformation of American Politics (Princeton, N.J.: Princeton University Press, 1989).


Nicholas A. Valentino and David O. Sears, “Old Times There Are Not Forgotten: Race and Partisan Realignment in the Contemporary South,” American Journal of Political Science 43 (9) (2005).


See Michael Tesler, “The Spillover of Racialization into Health Care: How President Obama Polarized Public Opinion by Race and Racial Attitudes,” American Journal of Political Science 56 (3) (2012): 690-704.


See Michael Tesler and John Sides, “How Political Science Helps Explain the Rise of Trump: The Role of White Identity and Grievances,” The Washington Post, March 3, 2016.


See Adam Enders and Steven Small, “Racial Prejudice Not Populism or Authoritarianism Predicts Support for Trump over Clinton,” The Washington Post, May 26, 2016.


A range of different surveys from the American National Election Study to the Cooperative Congressional Election Study across different election years and different levels of office, from the national to the state level, all show a tight relationship between immigration views and the vote. See Abrajano and Hajnal, White Backlash.


Author's analysis of the 2016 Cooperative Congressional Election Study,


Author's analysis of the 2018 Cooperative Congressional Election Study,


See Zoltan Hajnal and Michael Rivera, “Immigration, Latinos, and White Partisan Politics: The New Democratic Defection,” American Journal of Political Science 58 (4) (2014): 773-789.


See Donald Kinder and Wendy Kam, Us Against Them: Ethnocentric Foundations of American Opinion (Chicago: University of Chicago Press, 2012).


Immigration has mattered beyond race in more recent elections as well. See Marc Hoogeh and Ruth Dassonneville, “Explaining the Trump Vote: The Effect of Racial Resentment and Anti-Immigrant Sentiments,” PS: Political Science and Politics 51 (3) (2018): 528-535. They find, for example, that immigration attitudes rivaled racial considerations in explaining the vote in 2016.


The effect of immigration views on party identification is also apparent in two earlier panel studies from 2000-2004 and 1992-1993.


See John Sides, Race, Religion, and Immigration in 2016 (Washington, D.C.: The Democracy Fund Voter Study Group, 2017).


That analysis combines data from the two different data sets that most regularly ask about attitudes on immigration (The Gallup Poll) and partisanship (The CBS News/New York Times Poll). For attitudes on immigration, we focused on answers to the question: “Should immigration be kept at its present level, increased, or decreased?”


The Supreme Court has since struck down Trump's repeal of DACA. See Department of Homeland Security v. Regents of the University of California 591 US _ (2020).


See Rivera, The Determinants of State Immigration Policy.


Interestingly, whereas the size of the Latino population has a consistent, robust effect on policy, alternate analysis reveals relatively few connections between the size of the undocumented population or the size of the foreign-born population and state policy.


See Ted Brader, Elizabeth Suhay, and Nicholas Valentino, “What Triggers Public Opposition to Immigration? Anxiety, Group Cues, and Immigration Threat,” American Journal of Political Science 52 (4) (2008): 959-978.


Matthew C. Fellowes and Gretchen Rowe found similar patterns for welfare in “Politics and the New American Welfare States,” American Journal of Political Science 48 (2) (2004): 362-373.


There is, however, some good news for the Latinx and immigrant population. Our research in White Backlash also finds that once the Latino population share crosses a certain threshold, policy begins to shift back to the left. California's recent shift toward more welcoming immigration policy (including everything from offering undocumented immigrants in-state tuition, drivers' licenses, and the opportunity to practice law to multibillion dollar increases in education funding and significantly more lenient criminal justice policies) provides a clear illustration of this phenomenon. A state that was once the leader in anti-immigrant policy-making has transformed itself into a leader in providing creative, forward-thinking policies on immigration, largely thanks to its growing immigrant population and its recent shift to majority-minority status.


Maureen A. Craig and Jennifer A. Richeson, “On the Precipice of a ‘Majority-Minority’ America: Perceived Status Threat From the Racial Demographic Shift Affects White Americans' Political Ideology,” Psychological Science 25 (6) (2014).


See Zoltan Hajnal, Dangerously Divided: How Race and Class Shape Winning and Losing in American Politics (Cambridge: Cambridge University Press, 2020).


To do this analysis and to combine these categories, I used data from the 2016 Cooperative Congressional Election Study.


See Shanto Iyengar and Sean Westwood, “Fear and Loathing across Party Lines: New Evidence on Group Polarization,” American Journal of Political Science 59 (3) (2015): 690-707.

Published under a Creative Commons Attribution-NonCommercial 4.0 International (CC BY-NC 4.0) license. For a full description of the license, please visit