Introduction
In September 2019, Donald Trump held a rally in Rio Rancho, NM. In the state with the largest share of Latinos in the nation (U.S. Census Bureau, 2017), the president posed the following question: “Who do you like more—the country or the Hispanics?” Although Trump’s question was sarcastic and intended for Steve Cortes, his Latino ex-advisor who was in the audience, his quip tapped a broader identity issue within the Latino community. In the elections the following year, Trump increased his support among this constituency (Agiesta et al., Reference Agiesta, Luhby, Sparks and Struyk2020; Muravchik and Shields, Reference Muravchik and Shields2020). In other words, it seemed that in 2020, many Latinos indeed chose being American over being Latino. Any negative perceptions of Trump related to the Latino constituency and to the issue of immigration were apparently of little electoral consequence as far as voting among this political minority was concerned. Compared to 2016, more Latinos, standardly known to vote for the Democratic Party, cast a ballot for the Republican seeking reelection. Our article identifies and describes in detail this unexpected trend of minority support for a minority-hostile candidate, as well as aims to shed new light on broader questions concerning political minorities, elections, and voting in America.
Latinos, we show, have diverse political preferences even around immigration. This is a topic not only widely associated with this ethnic group for years but also one Latinos are expected to be of one mind on. However, stories such as that of Ms. Nury Martinez, a Latina who served as president of the LA City Council, are indicative of a different reality. On council meeting records released through social media, Ms. Martinez is heard making derogatory racist remarks about Black, Jewish, and gay people. Such inter-minority tensions are not surprising. Yet, the fact that Oaxacans were also an object of her pejorative comments highlights intra-minority tensions, that are mostly unfamiliar (Wharton, Reference Wharton2022). Furthermore, anti-Latino sentiments are not limited to Latino elites. According to the Pew Research Center, Latinos experience discrimination from other Latinos about as much as from members of other groups (Noe-Bustamante, Reference Noe-Bustamante2022). Such antagonism within the Latino constituency, we argue, is highly consequential politically.
Standard theoretical accounts on this topic are typically based either on ideology or on economic conditions. Ideology-based theories would suggest that it was conservative Latinos that supported Trump in bigger numbers in 2020 (Churchill, Reference Churchill2020; Lauter, Reference Lauter2021). His Supreme Court appointments placing pro-life justices on the high court and his conservative rhetoric about religious and family values mobilized such voters, these theories would suggest. As for theories about the effects of economic conditions on voting behavior, in such frameworks it was positive perceptions of the economy that increased support for the incumbent among Latino voters (Churchill, Reference Churchill2020; Narea, Reference Narea2021; Ramírez de Arellano, Reference Ramírez de Arellano2020). Common to both ideological and economic theories is the notion that Latinos who supported Trump did so despite—and certainly not because of—his hostile immigration policy.
Our robust findings based on data from the 2020 Cooperative Congressional Election Study (CCES) and the Democracy Fund + UCLA Nationscape survey indicate that neither theory persuasively explains this trend in voting behavior among Latinos.Footnote 1 We find that anti-immigration sentiments had a surprisingly substantial effect on the Latino vote. Indeed, this effect was strong enough to warrant a reevaluation of extant theories on the topic. White Americans are commonly theorized as a politically diverse group, which stands in stark contrast with standard depictions of voting behavior among political minorities. The latter are perceived as politically homogenous, voting together as a bloc. Our findings cast doubt on this understanding of Latinos, even on issues where they are commonly perceived to be in one mind on, such as immigration. Indeed, Latinos are more heterogeneous than typically thought, including on issues where such variance is not expected, like anti-immigration sentiments. What is more, such sentiment is politically consequential. Utilizing multiple indicators to measure support for Trump’s restrictive immigration policy, while controlling for perceptions of the economy and ideology, we find that Latino voters who joined the Trump camp in 2020 did so not despite his anti-immigration positions. They did so because of those positions.
Our argument is not that those anti-immigration Latinos exhibit anti-Latino prejudice per se and antagonism toward their own group. Though we partially deal with identity-related issues, which in this case revolve around immigration, our analyses focus on the political level. Arguing that Latinos siding with Trump are simply “against Latinos” is a politically vague assertion. Rather, we are concerned with the political manifestation of this claim, within the context of the Latino racial and ethnic group and its perceived shared interests. Accordingly, of key explanatory value for us is not a general anti-Latino sentiment. Rather, it is support for anti-Latino measures, specifically in the form of restrictive immigration policy.
Activation of dormant political attitudes and its political ramifications
Theories about Latinos espousing anti-immigration sentiments are not new. Economic hardships have been linked to support for more restrictive immigration policy among minorities and Whites alike (e.g., Burns and Gimpel, Reference Burns and Gimpel2000; Olzak, Reference Olzak1992). Yet, our account is innovative on this point as well. First, we argue theoretically and then demonstrate empirically that ethnic solidarity among Latinos around the topic of immigration, especially following the 2006 immigration reform protests, is significantly lower. While they replaced accounts about anti-immigration Latinos in the 1990s (e.g., De la Garza, Reference De la Garza1998), theories developed after the immigration reform protests fall short in explaining the current political climate. Second, our analyses suggest a different explanation for this variation among Latinos. We find that anti-immigration sentiments are more prevalent among Latinos of better, rather than worse, economic standing. Latinos who are well off feel less solidarity with their fellow Latinos around the topic of immigration.
The explanatory structure we suggest builds on theories concerning the activation of political attitudes, where external triggers—such as political actors—could stimulate dormant (or latent) dispositions among voters. The less consciously articulated such dispositions are, the more likely they are to require activation (Hawkins et al., Reference Hawkins, Rovira Kaltwasser and Andreadis2020). Furthermore, to be activated these dispositions require an enabling context (Ardag et al., Reference Ardag, Silva, Thomeczek, Bandlow-Raffalski and Littvay2020). Extant theories argue for a main effect of the economy on Latinos voting for Trump. Instead, we argue that there is a conditional effect. The economy serves as an enabling context for the activation of less consciously articulated sentiments about immigration. The role of the economy as a predictor for Latinos voting for Trump is as a contingent effect wherein perceived economic conditions determine how anti-immigration sentiment influences voters. This latent mindset of anti-immigration sentiments was triggered by Trump, a political actor, and enabled by the context of better perceived economic conditions among a substantial group of Latino voters. Trump would not trigger such sentiments among Latinos who are not as well off.
As Hawkins et al. (Reference Hawkins, Rovira Kaltwasser and Andreadis2020) indicate, the idea that certain attitudes constitute dispositions whose effective presence depends on external triggers—which is growingly acknowledged in political psychology literature—has been used in explaining a range of political phenomena such as populism (Ardag et al., Reference Ardag, Silva, Thomeczek, Bandlow-Raffalski and Littvay2020), authoritarianism (Feldman, Reference Feldman2003), and framing or priming (Chong and Druckman, Reference Chong and Druckman2007; Jackson, Reference Jackson2011; Nelson et al., Reference Nelson, Oxley and Clawson1997; Valentino et al., Reference Valentino, Brader and Jardina2013). We contribute to this debate by employing the same theoretical framework to explain Trump’s increased support among Latinos, expanding the scope of political phenomena for which such a framework applies.
Apart from the theoretical innovations in our work, the findings we present here are also important for their real-world political implications. Because of their strategic location in key battleground states, Latino voters sharing anti-immigration sentiments have likely influenced the outcome of the 2020 elections. Florida, Arizona, and New Mexico are good cases. In line with our explanatory structure, we show that in Florida, which Trump carried impressively, there were high levels of anti-immigration sentiments among Latino voters. In addition, the perceived economic standing of Latinos in Florida was much more positive than of Latinos in Arizona. On the other hand, in the Grand Canyon State—where Trump lost to Biden after easily carrying the state four years earlier—the share of this anti-immigration segment among the Latino electorate was considerably smaller and was coupled with more negative perceived economic conditions compared to Latinos in the Sunshine State. Latinos in Florida thought themselves as better off economically than Latinos in Arizona, showed lower levels of ethnic solidarity, and overall, their behavior was more similar to that of White voters.
In New Mexico, a blue state with the largest share of Latinos in the nation (U.S. Census Bureau, 2017), not only did Trump increase his vote share from 40.04% in 2016 to 43.50% in 2020 but also managed to broaden his support substantially in three of its most heavily Latino-populated counties: Lea, Chaves, and Eddy. As we show, compared to other states and in particular to Arizona, New Mexico Latinos showed higher levels of anti-immigration sentiment and far more positive economic perceptions, lending further support to our argument.
Latinos with anti-immigration attitudes, especially in those states where Latino votes are crucial—like Florida, Texas, Arizona, New Mexico, Nevada, and Colorado—could determine election outcomes. Despite their significant political sway and the new light they shed on extant theoretical accounts and conventional wisdom about political minorities, anti-immigration Latinos are consistently overlooked in political science as well as in public discourse. In conclusion, we discuss the political and theoretical implications of this diversity within the Latino community and examine whether it warrants a reevaluation of our understanding of the role in American politics of ethnicity, anti-immigration sentiments, and ethnic and racial minorities more broadly.
Immigration and ethnic solidarity among Latinos
Immigration is a key issue in American politics. Since the 1990s, when the number of newcomers from Central and South America grew significantly, the issue has become increasingly contentious. In particular, it has been controversies around the rights and legal status of undocumented immigrants, many of whom are Latinos, which generated the most belligerent political debates (Chavez, Reference Chavez2001; Pérez, Reference Pérez2015; Valentino et al., Reference Valentino, Brader and Jardina2013). Of the approximately 1M Latino immigrants in the U.S. in 1992, an estimated 600K entered illegally. By 2018, that number increased almost twentyfold when an estimated 10.6M people living in the U.S. were undocumented, the overwhelming majority of whom were Latinos (Tucker, Reference Tucker2020). Illegal immigration generated controversies at the national level, such as H.R. 4437 in 2006 and about a decade earlier at the state level with Proposition 187. If passed, H.R. 4437 would make it a felony to be undocumented in the U.S. (Gutierrez et al., Reference Gutierrez, Ocampo, Barreto and Segura2019). At the state level, Proposition 187 sought to limit access for illegal immigrants in California to public services and benefits, including healthcare and educational services.
Both in political discourse and in scholarship, it has been assumed that such controversies form and then solidify a sense of ethnic solidarity among members of migrant communities (Chong and Rogers, Reference Chong and Rogers2005; Gutierrez et al., Reference Gutierrez, Ocampo, Barreto and Segura2019). This was expected to be particularly true for the major immigrant community in the U.S., Latinos (Barreto et al., Reference Barreto, Manzano, Ramírez and Rim2009; Martinez, Reference Martinez2008; Pantoja et al., Reference Pantoja, Menjívar and Magaña2008). Members of a group that is at the heart of a national debate, which may also involve increased hostility toward that group, are prone to feel threat, fear, and animosity. According to social identity theory, members of a group in an inferior class—such as Latinos and Blacks—may take part in behavior aimed to improve their class and status (Deaux et al., Reference Deaux, Reid, Martin and Bikmen2006; Gutierrez et al., Reference Gutierrez, Ocampo, Barreto and Segura2019; Levin et al., Reference Levin, Sidanius, Rabinowitz and Federico1998). Of pivotal importance here is the notion of a linked fate (Chong and Rogers, Reference Chong and Rogers2005; Cohen, Reference Cohen1999; Dawson, Reference Dawson1994), which is a sense shared by members of a minority group that increases their sense of solidarity.
Latinos, however, differ from other minority groups such as African Americans, as distinctions exist between linked fate and ethnic solidarity theories. Chong and Rogers (Reference Chong and Rogers2005) define solidarity as social identity and interpretive-prescriptive ideologies on a group basis, transmitted through, among others, shared cultural and historical exposure. Unlike Black Americans, Latinos do not share the understanding that their individual fate is invariably linked to that of other members of their community. Instead, they feel bound by the goals defined by their group. Latinos generate ideological and organizational mobilization for the implementation of such goals, which are distinct from those defined by other groups in society (Barreto et al., Reference Barreto, Manzano, Ramírez and Rim2009). Many viewed the 2006 immigration protests as a clear manifestation of such ethnic solidarity. A common understanding was that in the wake of anti-immigration initiatives, all Latinos—regardless of generation, citizenship status, origin, or social status—were being adversely affected (Martinez, Reference Martinez2008), forging a sense of solidarity within that community.
Those theoretical frameworks would suggest that exposure to Trump’s rhetoric in the 2016 and 2020 election campaigns would lead Latinos to respond to his attacks not as Mexican Americans or Puerto-Rican Americans, but as Latinos (Gutierrez et al., Reference Gutierrez, Ocampo, Barreto and Segura2019). They were expected to unite around the issue of immigration with a strong sense of solidarity. Such expectations aside, reality proved to be different. Previous work on the subject showed that as the national mood toward illegal immigration grew hostile, indications increased for Latinos who distanced themselves from immigrants (Michelson, Reference Michelson2001; Tucker, Reference Tucker2020). What is more, even among members of specific subgroups within the Latino community—such as Mexican American, which is the largest community—divergence, and even antagonism, exist around the topic of immigration (Gutierrez, Reference Gutierrez1995; Martinez, Reference Martinez2008). In sum, while expected prima facie, ethnic solidarity among Latinos was more fragmented than what would be anticipated by extant theories. Such fragmented solidarity has theoretical roots and political consequences hitherto overlooked.
Anti-immigration Latinos
Theories of ethnic solidarity offer some explanations for why and when ethnic solidarity might be weaker. In times of collective threat, those members whose levels of group identity are higher, tend to engage in political efforts aimed at reinforcing ingroup favoritism and pride. Conversely, low identifiers might instead dissociate from the ingroup (Branscombe et al., Reference Branscombe, Schmitt and Harvey1999; Leach et al., Reference Leach, Rodriguez Mosquera, Vliek and Hirt2010; Doosje et al., Reference Doosje, Spears and Ellemers2002; Pérez, Reference Pérez2015). Other explanations focus on structural positioning. Similarities in social standing among members of a certain group lead to a shared sense of social identity (Hechter, Reference Hechter1975; Martinez, Reference Martinez2008). Yet, in the absence of this precondition, ethnic solidarity is unlikely. Consequently, connections between political behavior and ethnicity are salient when individuals of the same ethnic group share identical interests and when the conditions are conducive for group formation (Leifer, Reference Leifer1981; Martinez, Reference Martinez2008).
There are few topics as closely linked to structural positioning and related interests as immigration. Likewise, immigration is closely linked to a sense of collective threat. The controversies around H.R. 4437, Proposition 187, and Trump’s White House bid are all good examples. In 2014, 67% of registered Latino voters personally knew someone who was undocumented (Gutierrez et al., Reference Gutierrez, Ocampo, Barreto and Segura2019). Ethnic communities heavily affected by immigration, such as the Latino community, deal with basic inner divisions: those between citizens and non-citizens and those between documented and undocumented immigrants. Such internal divisions are symbolically reflected in the views of officers of Mexican ancestry working at the U.S.–Mexico border. Those officers are often critical of undocumented immigrants and are generally skeptical even about recently arrived legal immigrants (Tucker, Reference Tucker2020).
Such sentiments, however, are not limited to Latino Border Patrol agents. Au contraire, they are present in the general Latino population as well, citizens and noncitizens alike. Noncitizens of Mexican origin were just as likely as Latino citizens to oppose high levels of immigration and the prioritizing of immigrants from Latin American countries, possibly due to employment competition between those who recently arrived and those who wish to come (De la Garza, Reference De la Garza1998). According to the Pew Research Center, Latino support varies around border security issues (Krogstad and Lopez, Reference Krogstad and Hugo Lopez2021). Where asylum seekers coming from Central America are concerned, for example, there is a significant share of Latinos who do not express an obligation to help them (Samuels, Reference Samuels2022). Indeed, many Latinos worry that if others of their ethnic group do not immigrate the legal way, it would have a lasting impact on the Latino population currently living legally in the U.S. (Al-Gharbi, Reference Al-Gharbi2020).
This reluctance among parts of the Latino community to side with illegal immigrants brings to the surface identity complexities related to their social positioning compared to that of Whites. Interviews with Mexican Americans about Proposition 187 in California indicated that, notwithstanding grave collective consequences for Latinos, many of them found the Proposition favorable because of a desire to improve social standing and to align themselves with White Americans (Basler, Reference Basler2008). Along the same lines, Alamillo (Reference Alamillo2019) found that Latino support for Trump in 2016, and to a lesser degree for Romney in 2012, was largely due to denial of racism (Alamillo, Reference Alamillo2019). Per this logic, for Latinos who seek to become White, Trump’s anti-Latino and anti-immigrant rhetoric served as a test for one’s allegiance. Latinos whose ambition was to be socially equal to Whites might adjust their behavior and preferences to resemble those of Whites. This would involve adopting White views on racism and society—like colorblind ideology—or voting for White candidates with similar views, such as Trump (Alamillo, Reference Alamillo2019; Gans, Reference Gans2012).
At the basic level of racial identification, a majority among Latinos (58%) see themselves as White (Pew Research Center, 2021). This trend among the Latino community corresponds with literature concerning Latino internalized racism around skin tone and phenotype differences. A preference for lighter skin, for instance, was detected among Latinos of both lighter and darker skin (Golash-Boza and Darity, Reference Golash-Boza and Darity2008; Haywood, Reference Haywood2017). According to the Pew Research Center, Latinos experience discrimination from other Latinos about as much as from members of other groups (Noe-Bustamante, Reference Noe-Bustamante2022). Similarly, there are intra-group class divisions with an overriding power of social class in Latin American culture. In accordance with Latin American social norms, a dark-skinned person can self-identify as White if they belong in a higher social class (Golash-Boza and Darity, Reference Golash-Boza and Darity2008).
Latinos’ racial formation around and in relation to Whiteness is not unique in American history. Individuals of European origin, who were not initially considered White by the standards of the early 19th century, had to upon arrival to America subordinate ethnic identities and country, religious, or national animosities to a new color-based solidarity (Ignatiev, Reference Ignatiev1995; Jacobson, Reference Jacobson1999; Jardina, Reference Jardina2019; O’Malley, Reference O’Malley2022; Roediger, Reference Roediger2006). Just like in the case of Latinos in the 21st century, the color was white. Albeit hailed as an ethnic achievement, this “Whitening” process of some European ethnic groups was involved with assimilating into the discriminatory ideologies of the mainstream (Ignatiev, Reference Ignatiev1995). This was done while “eliding the politics of racial power and privilege” (Ignatiev, Reference Ignatiev1995; Kennedy, Reference Kennedy2022, 425). Given these similarities between historical and present-day circumstances, literature on Americanization and Anglo-assimilation has also found these notions relevant to the case of Latinos (e.g., Golash-Boza, Reference Golash-Boza2006).
Combined with Pérez’s (Reference Pérez2015) argument on the tendency for group exit in response to group threat among low-identifying group members, these theories and numbers provide solid foundations for a surprising claim. Rather than a political impediment for him, Trump’s anti-immigration rhetoric and policies were key in garnering unexpectedly high levels of support among Latinos in 2016 and even more so in 2020. Latino dispositions toward contentious topics like immigration are not dependent solely on ideological orientations, but instead derive from personal, even psychological, reasons (Alamillo, Reference Alamillo2019). Debating legal and illegal immigration renders ethnic identity salient for many Latinos, as they have been the constant focus of such debates (Pérez, Reference Pérez2015). It is for this reason that at least part of our study on anti-immigration Latinos is distinct from current research in the European context rationalizing why minority group members might vote for anti-minority political parties there. Because of the predominant role of racial and ethnic politics in the U.S., and since that across this nexus of identities the Latino group identity is fixed upon immigration-related issues, mere ideological divergence is proven short of explaining such complex identity positions, and the set of political behaviors they entail.
Around the world, but particularly in Europe, populist radical rightwing parties and politicians, who advocate for anti-immigration policies, have gained momentum electorally in recent years. Studies show that minority voters in Europe—including from migrant groups—contributed to the success of minority-hostile and anti-immigration parties, candidates, and policies. In Switzerland, for instance, a large share of voters who were formerly immigrants voted in favor of curbing immigration in a referendum “against mass immigration” in February 2014 (Masseneinwanderungsinitative; MEI). Strijbis and Polavieja (Reference Strijbis and Polavieja2018) explain this surprising similarity in voting behavior between natives and immigrants using labor market competition and high levels of exposure to cross-border commuters. Alternative for Germany (AfD), a populist radical right party, did remarkably well in the 2017 German federal elections, winning a record 12.6% of support. Like in Switzerland, voters who were formerly immigrants proved pivotal in this surprising electoral success. Of all potential groups, it was Russian-Germans, Germany’s biggest group of voters with an immigrant background, that were one of AfD’s key sources of political support (Spies et al., Reference Spies, Jasmin Mayer, Elis and Goerres2023). Spies et al. (Reference Spies, Jasmin Mayer, Elis and Goerres2023) found that the more Russian-Germans were integrated economically and socially into the overall German society, and the better their German language skills were, the lower was the likelihood of them voting for the AfD. These are just two out of various examples from Europe demonstrating the distinctiveness of the Latino case we analyze. In Switzerland, economic factors, rather than identity, were at the center for anti-immigration migrants. In Germany, economic and social integration had a negative relationship with anti-immigration, which is the opposite of what we, as well as other authors, find for Latinos in the U.S. (Alamillo, Reference Alamillo2019; Pérez, Reference Pérez2015).
These differences largely derive from a key issue. Latinos in the U.S. are considered one ethnic group, with ample literature dealing with its shared group identity that largely revolves around the issue of immigration. Notwithstanding a certain level of significance for ethnic identity, like in the case of immigrants from Turkey who are referred to as “Turks” in the European public, yet fails to distinguish between Turks, Kurds, and Alevis (Kentmen-Cin and Erisen, Reference Kentmen-Cin and Erisen2017; Koopmans, Reference Koopmans2015; Özyürek, Reference Özyürek2009), scholarship in the European context has mainly highlighted different aspects. Besides the Swiss and German cases, other accounts show a positive correlation between national chauvinism and anti-immigrant sentiment (Gustavsson and Stendahl, Reference Gustavsson and Stendahl2020; Jeong, Reference Jeong2013; Huddy and del Ponte, Reference Huddy, del Ponte, Gustavsson and Miller2019). Those share similarities with the Americanization, Anglo-assimilation, or Whitening theory in the American context. Some dealt with “economic insecurity” and “cultural backlash” theories (Dennison and Geddes, Reference Dennison and Geddes2019), similarly to the American context with respect to White support for Trump (Buyuker et al., Reference Buyuker, Jadidi D’Urso, Filindra and Kaplan2021; Cepuran and Berry, Reference Cepuran2022; Hopkins, Reference Hopkins2018; Reny et al., Reference Reny, Collingwood and Valenzuela2020; Sides et al., Reference Sides, Tesler and Vavreck2016). But overall, these theories, like those in the American context, fall short in explaining the counterintuitive political outcome of Latino voters siding with Trump in growing numbers.
Like Alamillo (Reference Alamillo2019), we hold that the moment in time when Latinos face identity-related questions is the moment when a contentious candidate, such as Trump, enters the political arena. Latinos would make up their minds about the issue of immigration, which underlies their ethnic group identity, when immigration is in the headlines and attracts public attention. It is therefore, as Pérez (Reference Pérez2015) argues, times of social ferment, rather than tranquility, which determine who engages in political activity to advocate for their group, and who chooses the option to exit. De La Garza (Reference De la Garza1998), Basler (Reference Basler2008), and Tucker (Reference Tucker2020) show that anti-immigration sentiments did exist among Latinos in the past. Since the 2006 immigration protests, accounts on ethnic solidarity have taken the lead in literature concerning Latino group identity, with the idea of ethnic solidarity in itself serving as antithetical to these past accounts on anti-immigration. Indeed, after the 2006 protests, statistics by the Pew Research Center showed that in the 2008 and the 2012 presidential elections there was a decline in public attention to the topic of immigration (Pew Research Center, 2008; 2012). This trend dramatically changed in 2016 due to Trump’s aggressive campaign and presidency, when immigration assumed again its position as one of the most central topics occupying the attention of the public. Particularly among Latinos, CCES data show that in 2008, only 1.42% ranked immigration as the most important issue of the elections, whereas in 2016, 50.9% of Latinos deemed immigration a topic of “very high importance” (Ansolabehere, Reference Ansolabehere2011).
In line with Alamillo (Reference Alamillo2019) and Pérez (Reference Pérez2015), but distinctively as we rely on an innovative framework of political activation, we contend that these could be understood as the conditions needed to render Latino group identity salient and to surface latent dispositions on anti-immigration that had already been present among this constituency in the past. Trump’s anti-immigration rhetoric and policies placed immigration at the heart of public debate in 2016 and then again in 2020. This served as an external trigger according to theories of political activation. But whereas past accounts on anti-immigration Latinos suggested that it was economic hardships that explained anti-immigration attitudes among Latinos, we demonstrate that it is in fact economic contentment that is the context needed for such counterintuitive political outcome to take place. Accordingly, as in the case of White voters, we expect an effect for anti-immigration sentiments on the voting choices of Latinos. Our first hypothesis is:
HYPOTHESIS 1: Ceteris paribus, higher levels of anti-immigration sentiments among Latino voters will be correlated with voting for Trump.
Competing accounts
To accurately assess the effects of anti-immigration sentiments on voting behavior, we also control for the effects of two other variables, widely considered in the literature to be of key influence on voting: ideology and the perception of economic conditions. While the two may be partly related to anti-immigration, in addressing and evaluating each separately as well as side by side with anti-immigration, we aim to capture Latino political behavior more comprehensively. Put simply, we want to show that the effect of anti-immigration sentiments on Latino voting behavior remains significant and of meaningful size even when controlling for the effect of ideology and perceptions of the economy.
Ideology—republicans but just don’t know it yet
Despite traditional support among Latinos for the Democratic Party (De la Garza and Cortina, Reference De la Garza and Cortina2007), there are several unusual characteristics of the Latino Democratic voter. President Reagan was the first to identify the proximity between Latino “social conservatism” and the Republican Party. In his and other conservative leaders’ view, Latinos were defined by principles of faith and family (Aguilar, Reference Aguilar2010). The “Republicans but Just Don’t Know It Yet” argument derives from the fact that this group is more religious than the general population, an attribute usually associated with voting for the Republican Party. Indeed, 53% of Latinos identify as Catholic and 25% as Protestant, whereas only 12% are religiously unaffiliated (Arana, Reference Arana2013).
Along the same lines, on key social issues, such as abortion and homosexuality, Latinos hold more conservative beliefs than the general public (Donato and Perez, Reference Donato and Perez2016). According to the Pew Research Center, more than half (51%) of adult Latinos say abortion should be illegal in most or all cases, which is significantly more than the 41% of voters who are pro-life in the general public (Taylor et al., Reference Taylor, Hugo Lopez, Martínez and Velasco2012). Thus, although Latino families have been susceptible to the same cultural and economic forces that are disrupting other American families, such as the rising rates of divorce and the greater prevalence of blended or mixed households, they are still more likely than other groups in American society to embrace traditional gender roles (Cox, Reference Cox2016; Forman-Rabinovici and Sommer, Reference Forman-Rabinovici and Sommer2018a; Forman-Rabinovici and Sommer, Reference Forman-Rabinovici and Sommer2018b; Forman-Rabinovici and Sommer, Reference Forman-Rabinovici and Sommer2019; Sommer and Asal, Reference Sommer and Asal2020). Another key factor in ideological divergence among Latinos is national. Of all Latin American countries, those who came from Cuba are much more likely to identify as Republicans, whereas Mexicans, Puerto Ricans, and Central Americans are more likely to side with the Democrats (Alamillo, Reference Alamillo2019; Alvarez and García Bedolla, Reference Michael and García Bedolla2003; Uhlaner and Garcia, Reference Uhlaner, Chris Garcia, Segura and Bowler2005).
Given those various elements of conservative ideology among Latinos, we hypothesize that:
HYPOTHESIS 2: Conservatism will have a positive correlation with the probability of voting for Trump among Latino voters.
Importantly, if ideology in itself were a sufficiently strong predictor for Latino political behavior, our notion of Republicans but Just Don’t Know It Yet would probably be of little purchase. Accordingly, while in H2 we expect a positive relationship between conservatism and the probability of voting for Trump, we also take into account that a closer examination of ideology among Latinos and its relation to political choice may reveal more complex dynamics, as we further elaborate below.
Perceptions of the economy
Scholarship has extensively examined the effect of economic conditions on vote choice, political preferences, and policy evaluations (e.g., Brooks and Brady, Reference Brooks and Brady1999; Fiorina, Reference Fiorina1981; Lewis-Beck and Stegmaier, Reference Lewis-Beck and Stegmaier2000; Reny et al., Reference Reny, Collingwood and Valenzuela2019). The 1984 Reagan and 1992 Clinton campaigns relied heavily on the issue of the economy. Reagan did so positively, through his It’s morning again in America campaign ad. Clinton did so negatively, using the slogan It’s the economy, stupid against the incumbent President G. H. W. Bush.
There are two axes according to which economic voting could be measured. First, scholars draw a distinction between two different kinds of economic conditions: prospective and retrospective. Prospective economic conditions are evaluations of future economic performance, while retrospective economic conditions are evaluations of past economic performance (Brooks and Brady, Reference Brooks and Brady1999; Fiorina, Reference Fiorina1981). In this article, we refer to retrospective economic voting. A second distinction lies within the retrospective category: between egocentric and sociotropic voting. The former relates to those voters who focus on their own financial standing, favoring their personal or their family’s economic conditions compared to that of the nation. The second prioritizes the collective aspect of the economy, usually the nation’s economic standing as a whole (Lewis-Beck and Stegmaier, Reference Lewis-Beck and Stegmaier2000). In this article, we measure retrospective economic voting with an index variable that is a combination of egocentric and sociotropic voting, measuring respondents’ opinions regarding their personal economic conditions, as well as collective economic circumstances.
Since Donald Trump assumed office in 2016, and through his 2020 presidential bid, many scholars identified relationships between economic conditions and voting for Trump. Unsurprisingly, as the Reagan and Clinton campaigns illustrate, negative perceptions of the economy were significantly associated with voting for Trump in 2016 (Reny et al., Reference Reny, Collingwood and Valenzuela2019). This is in line with the notion that negative perceptions of economic conditions are associated with voting against the incumbent party (Brooks and Brady, Reference Brooks and Brady1999), in this case the Democrats. Along the same lines, positive perceptions of economic conditions were significantly associated with voting for Trump in 2020 (Buyuker et al., Reference Buyuker, Jadidi D’Urso, Filindra and Kaplan2021).
Economic conditions are also linked to attitudes about immigration. Opposition to immigrants often arises from natives due to economic hardships (Burns and Gimpel, Reference Burns and Gimpel2000; De la Garza, Reference De la Garza1998; Olzak, Reference Olzak1992). People whose perceived economic conditions are worse will be more likely to harbor anti-immigration sentiments, which applies to Latinos as well (Burns and Gimpel, Reference Burns and Gimpel2000). Likewise, Strijbis and Polavieja (Reference Strijbis and Polavieja2018) show that labor market competition largely explained anti-immigration advocacy among immigrants in Switzerland, relying on a vast literature that argues for the same relationship (Kunovich, Reference Kunovich2013; Ortega and Polavieja, Reference Ortega and Polavieja2012; Polavieja, Reference Polavieja2016; Strijbis and Polavieja, Reference Strijbis and Polavieja2018). However, we argue that in the case of Latinos voting for Trump, the reverse logic applies. Although Trump could not singlehandedly alter public opinion among Latinos, as mass opinion on topics such as immigration cannot shift due to the activities of a single politician, he was able to activate a latent anti-immigration mindset among this constituency and render it politically salient.
Activation of political attitudes points to the effect political actors have as external triggers in activating dormant attitudes among voters (Hawkins et al., Reference Hawkins, Rovira Kaltwasser and Andreadis2020). An enabling context would lead to activation and consequently to political salience (Ardag et al., Reference Ardag, Silva, Thomeczek, Bandlow-Raffalski and Littvay2020). Contrary to competing theories, we argue that the economy serves as an enabling context for activating an anti-immigration latent mindset among Latinos, and less as a predictor with a main effect. Given the importance of structural positioning in the formation of ethnic solidarity (Hechter, Reference Hechter1975; Martinez, Reference Martinez2008), as described above, we expect that those Latinos with more positive perceptions of the economy will be also more likely to oppose high levels of immigration. The smaller the economic gap between Latinos and the average White majority, the weaker is their ethnic solidarity toward fellow Latino newcomers. Trump’s anti-immigration rhetoric, combined with the enabling context of Latinos in good economic standing that are structurally positioned far from immigrants, renders the latent anti-immigration mindset salient, with influence on their voting choice. Accordingly, we hypothesize that:
HYPOTHESIS 3: Among Latino voters, the effect of anti-immigration sentiments will be more pronounced on the probability of voting for Trump, when perceptions of the economy are positive.
Methodology
Data from the Cooperative Congressional Election Study (CCES) for the 2020 presidential elections were used. Administered by YouGov, this dataset consists of more than 50K responses from potential voters, gathered through a national stratified sample survey. Half of the questionnaire, administered to the entire pool of respondents, is focused on common content, disregarding ethnic, racial, or any other group affiliation. The other half is designated to group content. This half comprises of questions relevant to certain respondents based on their group affiliation and was answered by a subset of 1K people. In election years, the survey is carried out in two waves. Prior to the elections, respondents are given questions concerning their general political attitudes, various demographic factors, assessment of roll call voting choices, political information, and voting intentions. This first wave constitutes two-thirds of the questionnaire and is performed from late September to late October. The post-election cycle makes the other third and centers mostly on items related to the elections respondents had just voted in. This post-election cycle is administered in November.
For our dependent variables, we are interested in measuring support for Trump. Accordingly, we focus on two variables gauging voters’ presidential choices. The first asks those respondents who indicated that they voted early: “For which candidate for President of the United States did you vote?” The second poses the following question to respondents who indicated they had voted on Election Day, making the vast majority of respondents: “For whom did you vote for President of the United States?” In order to refine results, observations analyzed include only those respondents who supported either one of the two leading candidates, Trump and Biden, omitting respondents that marked “other,” “I did not vote in this race,” and “I did not vote.”
As for our independent variables, the key predictor, Anti-Immigration Sentiment, is the sum of four questions, each measuring agreement with statements concerning legal and illegal immigration. These four statements are: “Increase the number of border patrols on the US-Mexican border,” “Withhold federal funds from any local police department that does not report to the federal government anyone they identify as an illegal immigrant,” “Reduce legal immigration by 50 percent over the next 10 years by eliminating the visa lottery and ending family-based migration,” and “Increase spending on border security by $25 billion, including building a wall between the U.S. and Mexico.” The variable has been recoded on a scale of 1 to 5 so that higher scores reflect higher levels of anti-immigration sentiments.
The distribution of this independent variable across the U.S. lends preliminary support to its importance. Figure 1 shows levels of anti-immigration sentiments among Latinos by state. It is evident that in places with substantial Latino populations and where Trump improved his electoral performance—even in states like Nevada and New Mexico where he performed better though eventually did not carry the state—higher levels of anti-immigration sentiment among Latinos correlated with greater electoral success for the Republican candidate. Conversely, the resemblance in anti-immigration levels among Latinos between Arizona and states like California and New York may account for some of the hurdles Trump faced in the Grand Canyon State.
The second independent variable, Perception of Economic Situation (PES), measures how respondents perceive the state of the economy. It is composed of the responses to two questions, asking respondents about their assessment of economic conditions in the nation as a whole, as well as their own, on a scale from 1 (gotten much better) to 5 (worse). The questions are “Would you say that OVER THE PAST YEAR the nation’s economy has …,” and “OVER THE PAST YEAR, has your household’s annual income…?” The variable has been recoded on a scale from 1 to 10 so that a higher score indicates more positive perceptions. The third predictor in the regression models is ideology. It estimates respondents’ ideology on a scale from 1 (very liberal) to 5 (very conservative).Footnote 2 , Footnote 3
To increase robustness, several coding schemes for both the predictors and the outcome variables were used. Models were estimated with two different variables measuring Latino ancestry: race and “Hispanic Yes/No.” In this way, we can test our hypotheses on two different groups of respondents (however partially overlapping). Additionally, two of the predictors were coded both as a continuous variable and then with a different version of the variable as dichotomous. PES was coded as a dummy that assigns a score of 1 (positive perceptions of economic conditions) to respondents with scores higher than 5 on the continuous version of the variable, and 0 otherwise.Footnote 4 Ideology was coded as a dummy that assigns a score of 1 (conservative) to respondents with scores higher than 3 on the continuous version of that variable, and 0 otherwise. In the body of this article, we report the results for models estimated with the continuous versions of the variables.Footnote 5
Nationwide multivariate analyses
We first provide some descriptive statistics and correlative figures for the country as a whole. We then estimate a series of multivariate models. In Figure 2, levels of anti-immigration sentiments nationally among Latino voters are on the x-axis and the probability of those voters voting for Trump is on the y-axis. Lending preliminary support to H1, this figure showcases a strong positive relationship among Latino voters nationally between anti-immigration sentiments and the probability of voting for Trump.
To examine the systematic effects of our key independent variables, let us now turn to multivariate models. Table 1 presents models estimated for Latino and non-Latino voters (early voters and Election Day voters). In all models, anti-immigration, perceptions of the economy, and ideology are highly significant and positive. Anti-immigration is statistically significant when controlling for ideology and perceptions of the economy. Ceteris paribus, for each additional level of pro-immigration sentiment, the likelihood of voting for Trump increases by 3.120 for Latino and non-Latino voters casting votes early (Model 1) and 3.169 for Latino and non-Latino voters voting on Election Day (Model 2). This effect is statistically significant, controlling for alternative hypotheses.
Standard error in parentheses.
***p < 0.01, **p < 0.05, *p < 0.1.
The results in Table 2, presenting models estimated for Latino voters only (early voters and Election Day voters), lend strong support to our hypotheses. We fail to reject H1: anti-immigration is statistically significant when controlling for ideology and perceptions of the economy also for Latino voters only. Opposition to immigration explains 44.96% and 44.79% of the variance in early Latino voting for Trump. Furthermore, opposition to immigration accounts for 40.31% and 41.78% of the variance in Election Day Latino voting for the Republican incumbent. A highly significant positive correlation exists between anti-immigration sentiments and voting for Trump among Latino voters in Models 5–10. Ceteris paribus, for each additional level of pro-immigration sentiment, the likelihood of voting for Trump increases by 2.791 and 3.540 for Latino voters casting a vote early and 2.517 and 2.537 for Latino voters voting on Election Day. This effect is statistically significant, controlling for alternative hypotheses. H2 finds support across the board, as in all models the effect of ideology is statistically significant. In addition, these Models respectively explain 62.33% and 55.83% of the variance in Latino voting for Trump. Although not statistically significant among Latino early voters (Model 6), the interaction between anti-immigration and perception of the economy is significant among Latino voters on election day (Model 9), lending support to H3. What is more, Figure 3 shows the average marginal effects of anti-immigration at different levels of the moderator (PES). In line with H3, among Latino voters, the effect of anti-immigration sentiments on the probability of voting for Trump becomes more pronounced as perceptions of the economy grow more positive. Lastly, although we find the level of correlation among Latinos of voting for Trump and anti-immigration sentiments (0.67) to be stronger than that with ideology (0.58) and perception of the economy (0.53), Tables 1 and 2 do show that ideology is still the strongest predictor of voting for Trump.
Standard error in parentheses.
***p < 0.01, **p < 0.05, *p < 0.1.
To highlight the power among Latino voters of anti-immigration sentiments, Figure 4 shows the combined effects of anti-immigration and ideology among Latino voters nationwide, lending preliminary support to the hypothesized correlation between conservatism and electing Trump. More importantly, it shows that at the lowest level of anti-immigration sentiments (value of 1 for this variable), the distance between the maximum level (“Very conservative,” red line) and minimum level (“Very liberal,” blue line) is visibly smaller than the distance between these two lines at the highest level of anti-immigration (5). This shows how different levels of the moderator, ideology, increase the likelihood of voting for Trump, but with clearly different slopes, contingent on the level of anti-immigration sentiments. The effect of ideology is most substantial in the middle of anti-immigration levels (3 and 4), it is intermediate when anti-immigration is low (2) or high (5), and it is the lowest at the lowest level of anti-immigration (1).
Figures 3 and 4 highlight the significant role of anti-immigration, showing that greater anti-immigrant sentiment is generally associated with greater likelihood of voting for Trump, but even more so, for those who have a positive view of the economic situation. Latino voters with a positive perception of the economy are unlikely to vote for Trump if they are not anti-immigration. Likewise, Latino voters who are extremely conservative but in favor of a permissive immigration policy are unlikely to cast a ballot for the incumbent in 2020. Anti-immigration sentiment has a significant effect on voting for Trump among Latinos, besides those of ideology and economic perceptions, which are well documented in the literature.
In the Online Appendix (Tables A.2 and A.3), we include results for the 2012 and 2016 presidential elections. In line with our explanatory structure concerning Trump’s ability to activate dormant political dispositions among Latino voters around immigration, in 2012, when Mitt Romney was the Republican nominee, opposition to immigration explained only 19.87% of the variance in Latino Election Day voting for the Republican. While the explanatory power of anti-immigration sentiments had increased to 26.88% of the variance in Latino Election Day voting for Trump in 2016, the dramatic change came four years later when it more than doubled, with R2 = 0.4031. In other words, in 2020, anti-immigration sentiment alone accounted for 40.31% of the variance in Latino support for Trump on Election Day (Table A.4). This is further corroborated by the shrinking explanatory power of perception of the economy and ideology combined. In 2012, a model including only PES and ideology accounted for 49.29% of Latino support for Romney on Election Day (Table A.3), while in 2016 it was merely 32.80% (Table A.2). These suggest that Trump’s anti-immigration rhetoric and policies were indeed key in activating latent anti-immigration tendencies among Latinos. Romney, regarded much more moderate in terms of attacks on Latinos and immigrants (Alamillo, Reference Alamillo2019), won less Latino support than Trump. The scant support that Romney did win among the Latino constituency was explained mostly by perceptions of the economy and ideology. Not only did Trump, who ran a campaign heavily infused with immigration issues, win more support among Latinos than Romney, but it was anti-immigration sentiments that account for this support, both in 2016 and 2020.Footnote 7
As some occasionally argue for a geographical effect on the political behavior of Latinos—like in 2016 when it was rural Latinos who, similar to Whites, shifted toward Trump (Cadava, Reference Cadava2016)—we also ran analyses using the CCES’s indicator for geographical divergence. This variable allows us to draw a distinction between cities, suburbs, towns, and rural areas. We also created a dummy version of that variable to explore an urban vs. nonurban dichotomy. Results are unequivocal. Geographical differences among Latinos have negligible effect on the probability of voting for Trump and almost zero correlation with anti-immigration sentiments. Both the original CCES’s indicator and its dummy version explain 0.56% of the variance in Election Day Latino voting for Trump.
Finally, we include in the Online Appendix regression results for a large group of samples taken from the Democracy Fund + UCLA Nationscape survey. These samples comprise Latino, White, and Black voters from the 2016 elections (Tables A.9–A.28) and those who voted or intended to vote in the 2020 elections (Tables A.29–A.38). All the samples were randomly selected. In each case, we include the first ten samples of each phase of the three conducted in the survey. These analyses include the original three-indicator regression model we use, as well as a version specified with five new variables measuring group unfavorability toward undocumented, Latinos, Whites, Asians, and Blacks. In all those variables, the coding ranges from 1 for most favorable to 4 for most unfavorable.
These additional results strongly corroborate our findings and demonstrate how robust our argument is. In almost every sample among the 30, we include in those robustness tests, anti-immigration is highly significant and with a substantial effect among Latinos. Like ideology, the anti-immigration scale ranges from 1 to 5. In many samples, anti-immigration has a stronger effect on Latinos than ideology, and usually stronger even than among Whites. With respect to a relatively large portion of the samples, unfavorability toward undocumented immigrants (who are strongly associated with Latinos) is highly significant and with a substantial effect among Latinos. This effect is often stronger, or at least similar, compared to the same effect among Whites. Among Blacks, anti-immigration and unfavorability toward the undocumented are negligible. The Democracy Fund + UCLA Nationscape survey further confirms our argument that Latinos do find immigration an issue of key importance. What is more, their stances on this matter, including their disposition toward the undocumented, greatly shape their electoral preferences.
Latinos who find immigration important are not only those who vote Democrat. Increasing numbers of such Latinos cast a Republican vote. What is more, the fact that unfavorability toward Latinos among Latinos is for the most part negligible, while unfavorability toward undocumented immigrants is significant and with a substantial effect, demonstrates that much of the effect does not stem from anti-Latino prejudice. Rather, it is about political opposition to certain policies surrounding immigration. The effect is so strong, that this is true even when this opposition is viewed as prima facie harming the interests of one’s own group.
Anti-immigration Latinos in Florida, Arizona, and New Mexico
After demonstrating how at the national level, anti-immigration sentiment was a key motivator for Latinos in casting a vote for Trump, we now move to state-level analyses aimed to show how these anti-immigration Latinos influenced election outcomes in several key states where their votes were crucial. Figures 5 and 6 show the average scores of perceptions of economic conditions and anti-immigration sentiments among Latinos and non-Latinos in the top 10 states by percent of Latino population. Numbers are also reported by gender. Whereas among Latinos and non-Latinos (Figure 5), PES seems to have little bearing on anti-immigration sentiments (a line of observations that is almost straight), regardless of gender, in the case of Latinos alone (Figure 6), a clearer pattern appears. This is particularly true for Latino men. In states where the average PES score is higher, there are also higher levels of anti-immigration sentiments among Latino males (a clear shift to the right compared to Figure 5). Although both figures show greater prevalence for higher levels of PES and anti-immigration among males, Figure 6 demonstrates that this distinction is more pronounced in the case of Latinos. In Figure 6, there is a visually larger divergence between male observations (squares) and female observations (circles). Such a divergence corresponds with reports that Trump did particularly well among Latino male voters (Muravchik and Shields, Reference Muravchik and Shields2020). This finding lends support to our explanatory structure, according to which Latinos higher on PES will also be higher on anti-immigration. The fact that this pattern is not discernible in the general public further demonstrates the uniqueness of the Latino case. Indeed, comparing Florida—where Trump won impressively—and Arizona—where he bitterly lost—it is evident that Floridian Latino males were much more similar in their PES and anti-immigration levels to the general population (upper right end in both figures), than Arizonian Latino males. In the case of Latino males from Arizona, the values in the figure shift from almost the highest point of both axes in Figure 5 to roughly the middle on anti-immigration in Figure 6.
Figures 7 and 8 show levels of anti-immigration sentiments among Latinos and Whites in Arizona and Florida respectively, according to their perceptions of economic conditions. In Arizona (Figure 7), Latinos (in the top panel) with negative perceptions of economic conditions (red line) held considerably less anti-immigration sentiments than Whites (bottom panel) of the same PES category (red line in bottom panel). Latinos with positive perceptions of economic conditions (blue lines) were more anti-immigration overall. There is some indication, however, that some degree of ethnic solidarity did exist among Latinos around the topic of immigration in Arizona, even among those with positive perceptions of the economy.
The juxtaposition of these two states also supports our anti-immigration thesis in the face of an alternative account based on ideology. In the case of Latinos, the argument goes, those who vote for Trump are just more conservative. Data from the Pew Research Center fail to support such an alternative account. Of the population identifying as conservative in Arizona, 22% are Latinos, while the number in Florida is 19% (Pew 2022). Hence, based solely on ideology and given the relatively large share of conservative Latinos there, we would expect Trump in 2020 to carry the state of Arizona rather than the state of Florida. What happened was the opposite.Footnote 9 , Footnote 10
To further highlight the consequentiality of anti-immigration sentiments among Latino voters, let us also examine the state of New Mexico. After two heated campaigns and a presidency filled with controversies surrounding the topic of immigration (Reny et al., Reference Reny, Collingwood and Valenzuela2020)—including the family separation policy at the border, which led to the caging of child immigrants—very few expected that Trump would gain ground in the Land of Enchantment, a blue state with the largest percentage of Latinos in the country (U.S. Census Bureau, 2017). However, not only did Trump increase his overall vote share in the state from 40.04% in 2016 to 43.50% in 2020, but he also managed to substantially broaden his support specifically in three of its most heavily Latino-populated counties: Lea, Chaves, and Eddy.
Figure 9 shows the distribution of perceptions of economic conditions among Latinos (blue line) and Whites (orange line) in New Mexico (top panel) and Arizona (bottom panel). In New Mexico, the overall perceptions of the economy among Latinos were better than that of Latinos in Arizona. Indeed, within the positive range of 6–8 on the PES scale, in New Mexico Latinos relatively outnumbered Whites, while the opposite was true in Arizona. In Arizona, there was also a greater prevalence of Latinos concentrated at the most negative end of PES. This means a greater potential for anti-immigration sentiments among Latinos in New Mexico relative to Arizona. Furthermore, as Figure A.5 in the Online Appendix indicates, all the New Mexico Latinos who espouse anti-immigration sentiments have positive perceptions of economic conditions. This stands in contrast with typical accounts for anti-immigration sentiment among Latinos, linking it to economic hardships instead (Burns and Gimpel, Reference Burns and Gimpel2000; De la Garza, Reference De la Garza1998; Olzak, Reference Olzak1992).
In Figure 8 in Florida, on the other hand, indications for ethnic solidarity are weaker as the overall patterns of Latinos and Whites are quite similar. Those Latinos with a positive perception of the economy are much more like their White counterparts in their levels of anti-immigration sentiments. In other words, patterns of Latinos in Florida are much closer to those of Whites than Latinos in Arizona. This suggests lower levels of ethnic solidarity around the topic of immigration among Latinos in Florida.
The concentration of Latinos in those swing states, and the markedly different distribution among Latinos in each state on anti-immigration sentiments is important for its political ramifications. Such battleground states are won or lost with shifts even in a small number of votes. In Florida, where the presence of anti-immigration Latino voters was more significant compared to other swing states, in 2020 Trump did well electorally and carried the state with an impressive 3.36% gap, even more decisively than four years earlier. The same goes for New Mexico, where anti-immigration Latinos are numerous despite the state being a blue state overall. Indeed, the Trump campaign was aware of the potential among Latino voters in New Mexico. The on-the-ground operation of the campaign aimed to rally Latino voters in the state, especially across the borderline (Pettypiece and Alba, Reference Pettypiece and Alba2019; Rucker and Thebault, Reference Rucker and Thebault2019). Conversely, in Arizona, where Trump lost (after carrying the state in 2016 by 3.5%), the share of the anti-immigration segment among Latinos was considerably smaller. Compared to other states, including Florida and New Mexico, in Arizona Latinos opposing restrictive immigration policies far outnumbered those holding anti-immigration positions.
Discussion
Our results offer a plausible explanation for why a racially and ethnically controversial president gained electoral success among minorities, and in particular among Latinos. Frameworks highlighting their ethnic solidarity around the topic of immigration notwithstanding, Latino voters are not as homogenous as commonly thought. There is variance among Latinos even in the extent to which they identify with anti-immigration sentiments. Anti-immigration is present among this constituency and is highly correlated with supporting Trump on Election Day. A significant correlation between negative attitudes toward immigration and voting for Trump, which has been detected by a large body of work tracing the effect of anti-immigration sentiments on White voters in 2016 and 2020, is present also among Latinos. Although the prevalence of anti-immigration among Latinos is smaller relative to Whites, its effect on the probability of voting for Trump is substantial. What is more, given the geographically strategic location of concentrations of Latino voters in swing states, anti-immigration sentiments among Latinos may also be more electorally consequential than what the absolute numbers would suggest.
Conclusions
Throughout his four years in office, and even before he was elected, Donald Trump was perceived by many as a candidate and then a president hostile to minority groups. Moreover, many (Bseiso, Reference Bseiso2019; Cole, Reference Cole2019; Rosenfeld, Reference Rosenfeld2020) attributed to Trump White supremacist positions, when he championed White identity seeking to reestablish social hierarchal structures where Whites are at the top (Jardina, Reference Jardina2019).
Following his 2016 upset victory, a large body of research found a significant link among White voters between voting for Trump and anti-minorities sentiments, and in particular anti-immigration sentiment (Buyuker et al., Reference Buyuker, Jadidi D’Urso, Filindra and Kaplan2021; Cepuran and Berry, Reference Cepuran2022; Jardina, Reference Jardina2019; Schaffner et al., Reference Schaffner, Macwilliams and Nteta2018; Schaffner and Clark, Reference Schaffner and Clark2017). This effect predated the Trump years, as classical work identified the effects of anti-minority sentiments on political behavior of different groups in American society and among Whites in particular (Ostfeld, Reference Ostfeld2017; Sides et al., Reference Sides, Tesler and Vavreck2016).
Yet, Trump’s performance in the 2020 presidential elections offered a conundrum for political science: compared with 2016, the incumbent president gained increasing support among Latinos. How was it possible that such a controversial candidate in terms of race and ethnic relations managed to broaden his support among this racial and ethnic minority? Our argument may be surprising, but quite straightforward. Just as White voters who held anti-immigration positions supported Trump in bigger numbers, Latino voters who held similar positions also supported the president. Similar to the heterogeneity among Whites, there was divergence among Latino voters. On the one hand, there is a segment among Latinos which was appalled by his anti-immigration rhetoric and policy. On the other, among some members of the same constituency, Trump managed to activate a latent mindset of anti-immigration sentiments, while priming their low levels of ethnic solidarity.
Like in the case of populist politicians in other countries (Hawkins et al., Reference Hawkins, Rovira Kaltwasser and Andreadis2020), Trump’s harsh rhetoric with regard to illegal immigrants fed into an anti-immigration mindset familiar to researchers for several decades (Burns and Gimpel, Reference Burns and Gimpel2000; De la Garza, Reference De la Garza1998). But his rhetoric alone could not be effective without the accompanying structural disparity within the Latino constituency—between those who are better off economically and those who struggle—serving as the enabling context necessary for such activation to occur (Ardag et al., Reference Ardag, Silva, Thomeczek, Bandlow-Raffalski and Littvay2020).
Our study questions Latinos’ cohesiveness as an electoral bloc. We demonstrate that there are certain factors—the activation of dormant anti-immigration sentiment key among them—that weaken what is considered ethnic solidarity with loyalty to the Democrats. This is an important contribution to research on the activation of latent political attitudes in various political contexts (Ardag et al., Reference Ardag, Silva, Thomeczek, Bandlow-Raffalski and Littvay2020; Chong and Druckman, 2007; Feldman, Reference Feldman2003; Hawkins et al., Reference Hawkins, Rovira Kaltwasser and Andreadis2020; Jackson, Reference Jackson2011; Nelson et al., Reference Nelson, Oxley and Clawson1997; Valentino et al., Reference Valentino, Brader and Jardina2013). Such latent attitudes and their activation by a common stimulus in the form of Donald Trump suggest that the Latino constituency is more diverse than commonly thought and on a range of issues. Such diversity is true even on such issues where we would expect near unanimity such as positions toward their own community and on the issue of immigration. This diversity is consequential because it changes the way we think about and theorize Latinos as a pan-ethnic group. Secondly, the strategic geographical location of Latinos renders this diversity a fulcrum of decisive influence on election outcomes.
Our project helps to cast doubt on two key premises. First, our findings question the validity of what has been coined in literature as ethnic solidarity among Latinos around the topic of immigration. Second, instead of linking economic hardships and opposition to immigration, we find that anti-immigration sentiments are more prevalent among Latinos with positive perceptions of their and the nation’s economic conditions. Due to the activation of latent political attitudes, Trump was successful in priming low ethnic solidarity among Latinos with more positive perceptions of economic conditions because of their structural remoteness from fellow newcomers. Latinos with positive economic perceptions harbored lower levels of ethnic solidarity which constituted the enabling context needed for the effect that Trump had on the activation of their dormant anti-immigration sentiments.
All three of our hypotheses find strong support in the data. Latino voters higher on anti-immigration sentiments, with a more conservative ideology, and a more positive perception of economic conditions, were more likely to support Trump. Furthermore, the effect of anti-immigration sentiments is similar to those of either ideology or perceptions of economic conditions, which are well documented in the literature. Conservative Latinos, who are not anti-immigration, will overwhelmingly stay away from Trump. Likewise, Latinos who believe the economy is doing well would nonetheless be unlikely to vote for the incumbent if they are not anti-immigration. Indeed, the interaction between anti-immigration and perception of the economy is significant among Latino voters on election day.
This anti-immigration segment in the Latino constituency has important electoral consequences because of its relative concentration in swing states. Compared to other states with a substantial Latino vote, including Florida and New Mexico, in Arizona Latinos with low levels of anti-immigration sentiments far outnumbered those high on this scale. In terms of campaign strategy, the GOP and conservative leaders may draw optimism from our findings. Efforts by the Trump campaign to rally Latino voters in population centers across the nation, allocating campaign resources, signal the Republican establishment’s growing recognition of this electoral potential among Latinos (Pettypiece and Alba, Reference Pettypiece and Alba2019; Rucker and Thebault, Reference Rucker and Thebault2019). Moreover, our findings question occasional calls inside and outside the Republican Party to espouse more permissive immigration policies in an attempt to lure Latino voters. Our findings show that Republican candidates are able to increase Latino support while touting an anti-immigration agenda. Although not sufficiently successful to lend him a victory in 2020, the Trump campaign astutely recognized this potential. In battleground states, several thousand votes cast by anti-immigration Latinos could have tilted the outcome one way or the other, either in Trump’s favor or against him. This group of voters is largely overlooked in political science, despite its significant political sway (Tucker, Reference Tucker2020).
At least three venues for future work are of interest. First, our findings call for further exploration of the effects of anti-immigration on political behavior among Latinos. The premise that the Latino constituency is characterized by a unanimous perspective on immigration, or on politics more broadly, proved to be invalid. Second, the connection between structural, socio-economic positioning, and opposition to immigration among Latinos requires further exploration. Students of Latino politics should delve into the mechanisms leading to the effect we identify for anti-immigration sentiments on voting behavior among Latinos. Finally, with this research focusing on the Latino community and another analyzing the case of African American support for Trump (Sommer and Franco, Reference Sommer and Franco2023), future work could utilize our theoretical framework with regard to other minority groups, such as Asian Americans, where pro-Trump sentiments also grew between 2016–2020. While the context may be different, the counterintuitive trend is important if only for the fact that it casts political minorities in America from a different angle, including their political positions, their voting preference, and their impact on politics writ large.
Minorities may not be as uniform politically as they are often portrayed in the literature, their preferences may not lie exclusively with the Democratic Party, and they may espouse positions that on the face of things are directed against their very own political group. Given the relative concentration of members of minority groups in battleground states, what we reveal here is of interest not only from a scholarly perspective but also for the potentially surprising political implications it may spell.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/S1755773923000371.
Acknowledgments
The authors would like to acknowledge the support of the Center for the Study of the United States with the Fulbright Program at Tel Aviv University. This work received support from the Institute for Humane Studies under grant IHS016892 and from the Richard and Rhoda Goldman Fund. The authors would also like to thank participants, chairs, and discussants at the Midwest Political Science Association Annual Meeting in Chicago, IL, April 2023, in the panel: Trump’s America, and at the International Political Science Association Annual Meeting in Buenos Aires, July 2023, in the panel: Presidential Politics. Udi Sommer would like to thank Amy Adamczyk and Andrew Sidman as well as the Political Science and Sociology Departments for their hospitality at John Jay College of Criminal Justice at the City University of New York and their generosity while serving as John Jay Fall Research Fellow.
Competing interests
The authors declare no conflict of interests.