This paper examines geographic concentration in U.S. manufacturing industries using a "dartboard" approach to model firm location choices. The authors propose a new index of geographic concentration that accounts for differences in plant size distributions and geographic area sizes. They find that almost all industries are localized, although the degree of localization varies. The index is designed to allow meaningful comparisons of geographic concentration across industries. The paper also explores the nature of agglomeration forces, the geographic scope of localization, and the extent to which agglomeration involves plants in similar versus identical industries. The authors develop two models of location choice: one based on natural advantages (e.g., access to raw materials) and another based on spillovers (e.g., localized knowledge). Both models can explain geographic concentration, and the authors show that regardless of the mechanism, their index can control for plant size distribution and geographic area size. The paper also presents data on geographic concentration in U.S. manufacturing industries and finds that the level of concentration varies significantly across industries. The authors conclude that geographic concentration is widespread, although the degree of concentration varies. The index they propose allows for meaningful comparisons of geographic concentration across industries.This paper examines geographic concentration in U.S. manufacturing industries using a "dartboard" approach to model firm location choices. The authors propose a new index of geographic concentration that accounts for differences in plant size distributions and geographic area sizes. They find that almost all industries are localized, although the degree of localization varies. The index is designed to allow meaningful comparisons of geographic concentration across industries. The paper also explores the nature of agglomeration forces, the geographic scope of localization, and the extent to which agglomeration involves plants in similar versus identical industries. The authors develop two models of location choice: one based on natural advantages (e.g., access to raw materials) and another based on spillovers (e.g., localized knowledge). Both models can explain geographic concentration, and the authors show that regardless of the mechanism, their index can control for plant size distribution and geographic area size. The paper also presents data on geographic concentration in U.S. manufacturing industries and finds that the level of concentration varies significantly across industries. The authors conclude that geographic concentration is widespread, although the degree of concentration varies. The index they propose allows for meaningful comparisons of geographic concentration across industries.