INTRODUCTION
Immigration policy has gone local. Despite significant political attention to federal policy, and perhaps because of it, sweeping national immigration reform has been elusive. The federal government has used some of its muscle to devolve the responsibility for immigration enforcement to sub-federal governments (Varsanyi et al. Reference Varsanyi, Lewis, Provine and Decker2012). These governments have also made their own policies (Jiménez et al. Reference Jiménez, Schildkraut, Huo and Dovidio2021; Pham and Van Reference Pham and Van2022; Williamson Reference Williamson2018). The first wave of sub-federal measures in the early 2000s restricted access to rights, resources, and institutions. However, over the past decade, municipal and state governments have expanded immigrants’ access (Pham and Van Reference Pham and Van2022). A vital component of this new wave of local policies has been the Offices of Immigrant Affairs (OIAs)Footnote 1, which implement and manage programs that help integrate immigrants (De Graauw Reference De Graauw, Caponio, Scholten and Zapata-Barrero2018). Despite immigration being a politically polarizing issue, the proliferation of OIAs has attracted little attention. Perceptions of municipalities that have established these new bureaucracies have implications for their political feasibility. These perceptions may also portend the viability of similar federal immigrant integration efforts proposed by three of the last four presidential administrations (Exec. Order No. 13404, 2006; No. 14012, 2021; Memorandum 2014). How do individuals respond to the establishment of a local immigration bureaucracy aimed at serving the needs of the local immigrant population? To what extent does the presence of an OIA shape public attitudes about the cities that establish an office? How does the public perceive OIAs?
We use a set of pre-registered survey experiments to understand how individuals perceive a city’s integrative and exclusionary local immigration policies in response to a growing immigrant population.Footnote 2 Overall, we find that respondents have a more favorable view of a city that creates an OIA than they do of a city that cooperates with federal immigration enforcement. OIAs also receive broad-based support regardless of the legal status of beneficiaries. Partisan and ethnoracial differences attain. Republican respondents have more favorable views of cities that cooperate with federal immigration enforcement than those that establish an OIA; they are also more inclined to favor an OIA that serves only documented immigrants than those that serve immigrants regardless of their legal status. Democrats, especially white Democrats, are more apt to favor a city that establishes an OIA than one that cooperates with federal immigration enforcement; they have favorable views of OIAs regardless of the legal status of the clients. Republicans, especially white Republicans, see cities cooperating with federal immigration enforcement more favorably than those establishing an OIA. They have the most favorable view of an OIA that serves only documented immigrants. Although there is a high favorability, the balance of favorability tips in favor of cities that establish OIAs compared to those that cooperate with immigration enforcement. Republicans are more likely to favor cities with an OIA (64pp) than Democrats are to favor one that cooperates with federal immigration enforcement (52pp). Democrat favorability is above 80pp regardless of the legal status of beneficiaries; for Republicans, it is 54pp even when an OIA includes immigrants regardless of their legal status.
The results add to a growing literature suggesting public attitudes about immigration are less polarized than those of political elites (Thompson Reference Thompson2020). Our findings indicate that OIAs could attract broad-based public support. Integration policies, which include OIAs, appear to achieve key policy goals of creating equity, improving attitudes, and reducing social distance between immigrants and established populations (Kende et al. Reference Kende, Sarrasin, Manatschal, Phalet and Green2022). OIAs could be a rare immigration policy that may be effective and supported.
GOING LOCAL: RESPONDING TO SUB-FEDERAL IMMIGRATION POLICIES
Since the early 2000s, states, counties, and cities have implemented immigration policies in what initially seemed like a precursor to federal action (Varsanyi et al. Reference Varsanyi, Lewis, Provine and Decker2012). Between 2005 and 2019, there were 3216 immigration laws enacted across local and state governments (Pham and Van Reference Pham and Van2022). The first wave was dominated by exclusionary measures that limited immigrants’, especially undocumented immigrants’ access to rights, resources, and institutions. This wave of sub-federal policies included, among other measures, formal pacts with Immigration and Customs Enforcement,Footnote 3 ordinances forbidding immigrants from renting, denying undocumented immigrants in-state tuition at state colleges and universities, prohibiting undocumented immigrants from obtaining driver’s licenses, allowing law enforcement to ask for proof of legal residency, and invoking trespassing laws to remove undocumented immigrants (see Pham and Van Reference Pham and Van2022).
Since 2012, sub-federal immigration policymaking has taken a more welcoming turn (Pham and Van Reference Pham and Van2022). Among the measures passed during this second wave have been the establishment of OIAs: county and city offices dedicated to providing information, resources, and services to local immigrant communities. OIAs are now present in at least 73 localities, including San Francisco, New York, Los Angeles, Tulsa, Dallas, and Dayton, among others (see Supplementary Materials). The functions of OIAs include ensuring that local government services are accessible in multiple languages, disseminating information about the legal rights of immigrants regardless of legal status, civic leadership training, naturalization assistance, grants for immigrant-serving organizations, employment opportunities for Deferred Action for Childhood Arrival (DACA)Footnote 4 program recipients, and sponsoring public relation campaigns highlighting the contributions of immigrants (see De Graauw Reference De Graauw, Caponio, Scholten and Zapata-Barrero2018).
Although national politics has become more localized, with immigration as a flash point of political divisions, the politics of OIAs have flown under the radar (Hopkins Reference Hopkins2018). Even with growing partisan polarization, OIAs have been established in both Democratic and Republican strongholds, suggesting that the public may not be entirely opposed or perhaps aware of their presence (Williamson Reference Williamson2018). The proliferation of OIAs is amidst federal inaction on immigration reform (Bloemraad, Korteweg, and Yurdakul Reference Bloemraad, Korteweg and Yurdakul2008). Given the presence of OIAs in a range of locales, we hypothesize that:
H1 – Respondents will view a city that establishes an OIA more favorably than those that enact anti-immigrant policies aimed at increasing deportations.
Partisanship and political orientation moderate immigration preferences. Republican strongholds are more likely to implement restrictive immigration policies (Ramakrishnan and Wong 2010). While rising anti-immigrant rhetoric among political elites has mobilized voters with anti-immigrant views (Sides, Tesler, and Vavreck Reference Sides, Tesler and Vavreck2018), the partisan divide on immigration appears stronger among elites than voters (Thompson Reference Thompson2020). Indeed, overall American public opinion has been moving in an accommodating direction (Schildkraut Reference Schildkraut and Berinsky2019). Partisan polarization on immigration is driven by Democrats’ attitudes moving in the accommodating direction more rapidly than Republicans’ (Wright and Levy Reference Wright and Levy2020). What remains unclear is how partisanship shapes attitudes about local immigration bureaucracies. We hypothesize:
H1a – Favorability toward cities that establish an OIA will be moderated by partisanship. Democrats and independents will show the most support, while Republicans will be the least supportive.
Attitudes about immigration policy also refract through an ethnoracial prism. Existing research shows that whites hold less-accommodating views than non-whites, especially Latinos and Asians (Sides, Tesler, and Vavreck Reference Sides, Tesler and Vavreck2018). Therefore, we hypothesize:
H1b – Favorability toward cities that establish an OIA will be moderated by ethnoracial background. Compared to whites, non-white respondents will view cities that establish an OIA more favorably than cities that cooperate with immigration enforcement.
The interaction of ethnoracial identity and political orientation shapes immigration attitudes (Jiménez et al. Reference Jiménez, Schildkraut, Huo and Dovidio2021; Kaufmann Reference Kaufmann2018). Intra-ethnoracial partisan divisions have become especially pronounced among whites. On some measures of racial attitudes and immigration, white Democrats have shifted to the left of all other ethnoracial group/partisan combinations (Jiménez et al. Reference Jiménez, Schildkraut, Huo and Dovidio2021; Kaufmann Reference Kaufmann2018).Footnote 5 These insights lead us to hypothesize that:
H1c – Compared to all other ethnoracial/partisan combinations, white Democrats will have the most favorable views of a city that establishes an OIA.
Attitudes about bureaucracies assisting immigrants depend on the subpopulations of immigrants who benefit. Though 77% of immigrants in the United States are citizens, lawful permanent residents, or visa holders, the estimated 23% who are undocumented attract enormous political and policy attention (Budiman Reference Budiman2020). Americans can be sympathetic to undocumented immigrants, believing that they should have an opportunity to adjust to their status while also viewing legal status as a bright line demarcating who should have access to government assistance (Jiménez Reference Jiménez2017; Jiménez et al. Reference Jiménez, Schildkraut, Huo and Dovidio2021). Americans are also more likely to associate undocumented status with criminal behavior and non-whiteness, potentially diminishing support for undocumented immigrants’ access to the welfare state (Flores and Schachter Reference Flores and Schachter2018). We therefore hypothesize:
H2 – Respondents will be less supportive of an OIA when they serve undocumented immigrants than when they focus on authorized immigrants.
Together, these hypotheses capture an extensive set of attitudes about OIAs and their potential contribution to making cities desirable places to live for immigrants and established communities.
DATA AND METHODS
Using the Lucid Theorem platform, we collected a nationally representative sample of 2,107 respondents in October 2021. We collected a national sample because the proliferation of local immigration policies, including OIAs, is taking place across the United States. Additionally, a national sample avoids introducing unobservable individual traits associated with living in a city with an OIA. Lucid recruited respondents through an online opt-in model.Footnote 6 The sample was balanced across experimental groups.Footnote 7 The analyses focused primarily on differences across partisan and ethnoracial groups. We bundled all non-white respondents because of statistical power limitations. We thus compare attitudes among white and non-white respondents.Footnote 8
Results
Experiment 1: Immigration Policies and Favorability
In the first survey experiment, we introduced respondents to a fictitious town in the “American Midwest” with attributes commonly favored by the public: good schools, affordable housing, and a growing economy (Knight Foundation 2010). The city also has a growing immigrant population. We randomly assigned respondents to one of two policy response treatments. In one treatment, the city’s local leaders “passed a city ordinance that creates an office of immigrant affairs that helps immigrants integrate into the community.” The other treatment stated that local leaders “passed a city ordinance to cooperate with federal immigration enforcement to identify and detain immigrants for deportation.” We then asked respondents whether they had a (1) favorable or (0) unfavorable view of the city using a 5-item Likert scale.Footnote 9
Figure 1 displays variance in the city’s favorability for the pooled sample and by partisanship. The results broadly support H1: respondents view a city that establishes an OIA more favorably than one that cooperates with federal immigration enforcement. Following H1a, Democrats view a city that creates an OIA 30pp more favorably than one that cooperate with federal immigration enforcement (p <0.05). The difference in favorable versus unfavorable views shrinks to 15pp among independents. However, it remains statistically significant (p <0.05). There is a large drop in favorability among Democrats and independents when the city cooperates with federal immigration enforcement. Even so, the favorability among Democrats and independents remains above a majority, suggesting that these groups are not entirely opposed to a city that enacts enforcement-oriented policies.
Figure 1. Impact of the Presence of OIA on Favorability toward the City by the Partisanship and Experimental Group
Note: Displays mean responses for the question: “how favorable or unfavorable do you view this city?” using a 5-item Likert scale ranging from (1) very favorable to (0) very unfavorable. Includes 95% CI. See Supplementary Tables A6–A7 for full results.
As predicted in H1a, Republicans viewed a city with an OIA 10pp less favorably than one cooperating with federal immigration enforcement (p <0.05). Still, even when a city creates an OIA, Republican mean favorability is roughly 65pp.
Figure 2 presents results comparing white and non-white respondents. Similar to the results in Figure 1, both groups view a city establishing an OIA more favorably than one cooperating with federal immigration enforcement. Moreover, contrary to H1b, there are slight differences in the ethnoracial background, but they are not statistically significant. Thus, the ethnoracial background, by itself, does not have a statistically significant effect on perceptions of cities with an OIA.
Figure 2. Impact of the Presence of OIA on Favorability toward the City by the Ethnoracial Background and Experimental Group
Note: Includes 95% CI. See Supplementary Table A10 for full results.
However, attitudes toward a city with an OIA differ when accounting for the interaction between the ethnoracial background and partisanship (see Figure 3 and Supplementary Tables A11–A12). Following H1c, white Democrats had a more favorable view than any ethnoracial/partisan combination. White Democrats showed 5pp more support than non-white Democrats (p <0.05). Compared to all other groups, white Republicans showed the least favorability toward a city with an OIA. White Republicans registered a mean favorability 10pp lower than non-white Republicans (p <0.05). Notably, white Republicans’ mean favorability of a city that establishes an OIA was roughly 62pp.
Figure 3. Impact of the Presence of OIA on Favorability toward the City by the Partisanship, Ethnoracial, and Experimental Group
Note: Left panel includes non-White respondents, right panel includes White respondents. Includes 95% CI. See Supplementary Tables A11–A12 for full results
Experiment 2: Impact of Immigration Status on OIA Attitudes
In a second experiment, we study whether respondents’ favorability of OIAs is conditional on the clients’ immigration status. To do so, we asked respondents to imagine that their city planned to establish an OIA with three primary functions (we did not spell out these functions in the Experiment 1 prompt): hold English classes, connect immigrants to local public services, and help integrate immigrants into the local community. A control group viewed only these three roles. The inclusion treatment group received additional information: the OIA would provide “aid to immigrants regardless of legal status,” suggesting that both documented and undocumented immigrants have access to services. The exclusion treatment group received information that the OIA would provide services “only to immigrants that are in the country lawfully.” We then asked respondents to what extent they would (1) support or (0) oppose the creation of an OIA in their city using a 5-item Likert scale.
Figure 4 illustrates the general treatment effects across experimental groups and partisanship. Compared to the control, support for establishing an OIA remains high regardless of the immigration status of clients (p >0.05). Following H2, support in the pooled sample is greatest when access is limited to authorized immigrants. Democrats and independents are equally supportive of OIAs regardless of legal status provisions. In contrast, Republicans are more supportive of the OIA if benefits are limited to authorized immigrants. Despite the decrease in support when access is not conditional on legal status, Republicans’ favorability toward OIA is 54pp. The results show that OIAs enjoy a high floor of support across the political spectrum, even when benefits are available regardless of clients’ legal status.
Figure 4. Impact of Access on Mean Support for Establishing an OIA by Partisanship and Experimental Group
Note: Displays mean responses for the question: “to what extent would you support or oppose the creation of an Office of Immigrant Affairs in your city?” using a 5-item Likert scale ranging from (1) strongly support to (0) strongly oppose. Includes 95% CI. See Supplementary Tables A13–A14 for full results.
Figure 5 presents the results across ethnoracial groups and partisanship. Overall, there is little difference in support across the experimental groups. The results, however, suggest that ethnoracial background shapes attitudes only when the OIA offers benefits to all immigrants. Like results from Experiment 1 (see Figure 3), white Democrats are more supportive than non-white Democrats (p <0.05). The opposite pattern appears among Republicans: compared to white Republicans, non-white Republicans are considerably more supportive of the OIA when it provides benefits to all immigrants (p <0.05).
Figure 5. Mean Support for OIA by the Ethnoracial Background, Partisanship, and Experimental Group
Note: Left panel includes non-White respondents, right panel includes White respondents. Includes 95% CI. See Supplementary Tables A15–A16 for full results
DISCUSSION AND CONCLUSION
Understanding Americans’ policy attitudes toward immigration requires going local. Fraught political debates over immigration policy have consistently blocked large-scale reform. Municipal and state governments have filled the void (De Graauw Reference De Graauw, Caponio, Scholten and Zapata-Barrero2018; Pham and Van Reference Pham and Van2022; Varsanyi et al. Reference Varsanyi, Lewis, Provine and Decker2012). Here, we examine support for OIAs compared to local cooperation with federal immigration enforcement. In our sample, respondents show far more favorable views of a city with an OIA than one cooperating with immigration enforcement. We also show that the overall support for establishing OIAs is high, regardless of the legal status of immigrant clients. Partisan differences in our sample reflect broad trends in attitudinal polarization about immigration (Sides, Tesler, and Vavreck Reference Sides, Tesler and Vavreck2018). Democrats and independents favor a city that establishes an OIA compared to one that cooperates with federal immigration enforcement. These partisan groups are also inclined to support an OIA regardless of the recipients’ legal status. Republicans, on the other hand, have a more favorable view of cities that cooperate with federal immigration enforcement and are more supportive of an OIA that serves only immigrants in the country lawfully. Where ethnoracial background is concerned, we find no difference between whites’ and non-whites’ favorability of a city that establishes an OIA. But among whites, large partisan differences appear. White Democrats had the most favorable views of cities with an OIA and showed the most support for establishing an OIA, irrespective of the legal status of immigrant beneficiaries. Our findings are consistent with other research showing that among all ethnoracial/partisan combinations, white Democrats have the most liberal views on immigration. Meanwhile, white Republicans exhibit the least accommodating opinions (Jiménez et al. Reference Jiménez, Schildkraut, Huo and Dovidio2021; Kaufmann Reference Kaufmann2018).
Notably, however, there is a high level of favorability of cities with OIAs and support for OIAs in general. To put this in perspective, the group in our sample least supportive of cities with an OIA, white Republicans, still has a mean favorability of 62pp (compared to 51pp among white Democrats for cities that cooperate with immigration enforcement). Moreover, Republican favorability remained above 50pp even when an OIA included immigrants regardless of their legal status. Though we cannot test explanations for why we find relatively high support even among Republican respondents, our findings are consistent with national trends in views about immigration, positive reactions to immigrant integration, and trust in local government to carry out policy aims. Where attitudes about immigration are concerned, Americans have become more accommodating in recent years. Driving attitudinal polarization is Democrats moving in that direction far more rapidly than Republicans, who have remained split in their views (Wright and Levy Reference Wright and Levy2020). Specifying the integrating functions of an OIA may have also raised the support floor. Research on immigration-driven demographic change shows more positive reactions to narratives that emphasize integration (Levy and Myers Reference Levy and Myers2021). That OIAs work to integrate immigrants, as we specify in both experiments, could increase the overall support. Finally, our findings may also owe to partisan differences in support for federalism. Conservatives, in contrast to liberals, have a principled preference for local and state governments to implement a range of policies (Glaser, Berry, and Schildkraut Reference Glaser, Berry and Schildkraut2022). If that principled preference extends to immigration, which is yet to be tested, then immigration policy done locally, rather than federally, may garner more favor among Republicans. These possible explanations should be considered against the backdrop of a divide between elite and rank-and-file Republican views about immigration (Thompson Reference Thompson2020). Restrictionist policies pushed by Republican elites activate the most vocal among the rank and file to raise their voices against immigrants, lending to the perception that extreme restrictionist views prevail (Flores Reference Flores2017). In fact, there is a critical mass of Republicans who, as our findings suggest, would be inclined to support accommodating measures.
Our study is the first, to our knowledge, to examine support for OIAs. Where our study is not exhaustive, it points the way for further research driving views about OIAs and the cities in which they reside. Providing respondents with information about the characteristics of immigrant recipients, asking individuals about a proposed OIA in their own city (as opposed to a near-ideal fictional city), and giving respondents information about how the city would pay for the OIA could add texture to our findings.
Nevertheless, our results dovetail with research showing that integration policies foster equity between immigrants and established populations, leading to improved intergroup relations (Jiménez et al. Reference Jiménez, Schildkraut, Huo and Dovidio2021; Kende et al. Reference Kende, Sarrasin, Manatschal, Phalet and Green2022). Together, these findings suggest that OIAs can foster integration with bipartisan support.
SUPPLEMENTARY MATERIAL
To view supplementary material for this article, please visit http://doi.org/10.1017/S0003055423001077.
DATA AVAILABILITY STATEMENT
Research documentation and data that support the findings of this study are openly available at the American Political Science Review Dataverse: https://doi.org/10.7910/DVN/UXB857.
ACKNOWLEDGMENTS
The authors thank the following individuals for their input at various stages of this research: Ariela Schacter, Deborah Schildkraut, Aliz Toth, Alice Wang, Soyoung Lee, Rachel Perić, Meg Shoemaker Little, and anonymous APSR reviewers.
FUNDING STATEMENT
This research was funded by Stanford University.
CONFLICT OF INTEREST
The authors declare no ethical issues or conflicts of interest in this research.
ETHICAL STANDARDS
The authors declare that the human subject research in this study was reviewed and approved by Stanford University (IRB Protocol 62402). The authors also affirm that this article adheres to the APSA’s Principles and Guidance on Human Subject Research.
Open access
Comments
No Comments have been published for this article.