This paper investigates the causes of industry agglomeration using coagglomeration patterns. The authors analyze data from the U.S. Census Bureau's Longitudinal Research Database (1972–1997) to compute pairwise coagglomeration measures for U.S. manufacturing industries. They assess the importance of three theories of industry agglomeration: (1) transport cost savings from proximity to suppliers or customers, (2) labor market pooling, and (3) intellectual spillovers. Using regression analysis, they find that input-output dependencies are the most important factor, followed by labor pooling. They also use U.K. data as instruments to address potential endogeneity issues. The results suggest that industries with strong input-output relationships and shared labor needs are more likely to coagglomerate. The study highlights that coagglomeration patterns are not random, and firms tend to locate near their customers and suppliers, as well as near industries that share labor or ideas. The findings support the importance of transport cost savings and labor market pooling in explaining industry agglomeration.This paper investigates the causes of industry agglomeration using coagglomeration patterns. The authors analyze data from the U.S. Census Bureau's Longitudinal Research Database (1972–1997) to compute pairwise coagglomeration measures for U.S. manufacturing industries. They assess the importance of three theories of industry agglomeration: (1) transport cost savings from proximity to suppliers or customers, (2) labor market pooling, and (3) intellectual spillovers. Using regression analysis, they find that input-output dependencies are the most important factor, followed by labor pooling. They also use U.K. data as instruments to address potential endogeneity issues. The results suggest that industries with strong input-output relationships and shared labor needs are more likely to coagglomerate. The study highlights that coagglomeration patterns are not random, and firms tend to locate near their customers and suppliers, as well as near industries that share labor or ideas. The findings support the importance of transport cost savings and labor market pooling in explaining industry agglomeration.