Research

Each year, the United States resettles thousands of refugees from around the world. Despite the long history of the resettlement program, we know little about how United States citizens react to the diverse population of refugees. Using refugee placement data, I study how the resettlement of refugees impacts U.S. natives' voting behavior at the county level in Presidential, Senate, and House elections between 1992 and 2016. To address the non-random placement of refugees, I employ a shift-share instrumental variable approach based on the initial county-level distribution of refugees in 1991 interacted with national refugee flows. I find that refugee resettlement results in an increase of support towards the Republican Party, which historically has platformed more anti-immigration policies. To understand why resettlement increases Republican support, I exploit variation in the geopolitical relationship between the U.S. and refugees' countries of origin. I find that the effect is driven entirely by refugees from adversarial countries, nations at odds with U.S. foreign policy, rather than from friendly countries. These findings suggest that political responses are shaped not only by the number of refugee, but also by the perceived political ideological differences.

Throughout history, organized labor unions have been advocating for workers' rights in the labor market. Despite this, union membership across OECD countries has declined since 1985. Previous work on union membership determinants have considered most socioeconomic factors, except for religious affiliation. Religious teachings have been shown to not only be a source of guidance in the household, but also in economic outcomes such as work. This paper tests whether Catholic affiliation, relative to Protestant, predicts union membership. Using six waves (2001–2019) of the German Socio-Economic Panel, I estimate an individual fixed-effect model and find being Catholic increases the probability of union membership by 4.8%. To further address endogeneity concerns, I instrument a county's contemporary Catholic share with the religion of its historical territorial ruler under the Peace of Augsburg (1555). I assume that a ruler's decision in 1555 does not correlate with contemporaneous decisions to join a labor union. The two-stage least squares estimates confirm the main result as do the robustness checks. The findings establish religious affiliation as a novel determinant of union membership.

Following the Geneva Conference in 1954, Vietnam was divided along the 17th parallel, splitting the country into North and South Vietnam, and allowed for 300 days of open borders during which citizens from either side could freely move between regions. Fearing religious persecution and political repression under the communist leadership of North Vietnam, over 800,000 Northern Vietnamese citizens chose to migrate south. This massive migration posed an enormous challenge for South Vietnam, a newly established state with limited resources, even with the support of the French and U.S. military. Despite both Northern refugees and native South Vietnamese being anti-communist, the South Vietnamese saw their Northern co-patriots as financial burdens and completely dependent on domestic and foreign aid. Adding to this, the refugees were perceived as poor, unclean, and culturally different. Many scholars have theorized that this divide, coupled with resource scarcity, created an atmosphere of mistrust and resentment, undermined South Vietnam’s stability and its efforts to resist the growing insurgency of the Viet Cong. I test this hypothesis by first digitizing novel refugee settlement data from the Michigan State University Vietnam Group Archive. This archive contains information on 180 refugee hamlets established in 1955, of which 101 were geo-located. These hamlets were then linked with data from the quarterly Hamlet Evaluation System (HES) from the U.S. National Archives. The HES provides data on the political and economic structures of South Vietnam hamlets, as well as the frequency and intensity of Viet Cong activities. Preliminary linear regression shows that there is a positive and significant result between refugee numbers and a South Vietnam hamlet's Viet Cong activities.