Political Agency Under Hostility: How Elite Rhetoric Reshapes Migrant Engagement

Research on political hostility focuses predominantly on its effects on majority populations. This paper shifts the lens to those most directly targeted, investigating how local manifestations of anti-immigrant hostility—physical, institutional, and discursive—shape the strategic responses of migrants across the transnational field. I argue that hostile environments operate as contextual signals that update migrants' expectations about the host society's long-term reliability, triggering a strategic recalibration of resources, attachments, and political engagement. I develop and test a framework distinguishing two core responses: "opting-out," where migrants reorient toward the country of origin through return intentions or strengthened homeland identification, and "voicing-out," where they assert political presence in the host society through increased engagement or compensatory party identification. The choice between these pathways is structured by both the type of hostility signal and migrants' generational position, which proxies for the relative viability of exit versus voice. To assess this, I employ a novel longitudinal design. Individual-level panel data from the German Socio-Economic Panel (SOEP) is linked to subnational indicators of far-right electoral strength (institutional signal), hate crimes (physical signal), and anti-immigrant rallies (discursive signal). Using two-way fixed effects models that exploit within-individual variation over time, I identify how changes in local hostility shape migrant outcomes across three domains: opting-out behaviors (return intentions, remittances), identity and belonging (attachment to Germany and country of origin), and voicing-out behaviors (political interest, party identification, ideological placement). The study makes three key contributions. First, it provides the first systematic evidence that different types of hostility—physical, institutional, and discursive—trigger distinct response patterns, revealing that migrants distinguish between threats to personal security, political membership, and public discourse. Second, by tracking the same individuals longitudinally, it moves beyond cross-sectional research to establish that migrants adjust their strategies as their local environments change. Third, it reveals systematic generational heterogeneity: first-generation migrants translate institutional threats into both concrete exit orientations and compensatory host-country engagement, while second-generation migrants—lacking exit resources—respond with symbolic homeland identification alongside stable political participation. Together, these findings advance a model of migrants as strategic actors who interpret environmental signals and reallocate resources across borders in structured, theoretically predictable ways.