How does exposure to individuals from different ethnic backgrounds during childhood affect interethnic relations in adulthood? Set in late 19th century America, a period of peak immigration of German and Irish families, we analyze the impact of childhood cross-ethnic exposure on later-life decisions to marry an outgroup partner. Using full-count U.S. census data from 1880 and 1900, the initial sample includes over 370.000 American, German, and Irish boys up to the age of 18 across 40 cities. Our analytical strategy includes identification of next-door neighbors based on census entry order and georeferenced data, construction of organic neighborhoods using an innovative machine learning algorithm (Ostermann, 2024), and the linkage of boys in the 1880 U.S. census with their respective entry in the 1900 U.S. census. We create samples of matched boys who are identical in terms of socioeconomic and demographic factors and differ only in whether they grow up next to an ethnic outgroup neighbor or solely next to ethnic ingroup neighbors. The findings reveal significant increases in the predicted probability of interethnic marriage for individuals exposed to neighbors of a different ethnicity during childhood. The effect varies across the three ethnic groups, with American boys showing the most consistent positive effects. Further analysis demonstrates that the impact of interethnic exposure on marriage decisions is moderated by neighborhood ethnic composition. For American boys, the effect decreases as the share of Americans in the neighborhood increases, while patterns for German and Irish boys are more complex. These results contribute to our understanding of how early-life intergroup contact shapes long-term social cohesion, offering insights into the role of spatial proximity in fostering interethnic relationships.
