The dynamics of industry agglomeration: Evidence from 44 years of coagglomeration patterns
分析了美国制造业44年的共同集聚模式，发现投入产出关联和劳动力市场共享的重要性下降，而知识溢出的重要性上升，并与贸易和技术冲击相关。
Evidence abounds that agglomeration patterns have changed over time, but little is known about changes in the underlying determinants of agglomeration. We analyze 44 years of coagglomeration patterns of U.S. manufacturing industries and show that over time, input-output linkages and labor market pooling have become less important determinants of industry agglomeration, while knowledge spillovers have become more important. We show that trade and technology shocks are strongly associated with the decline in labor market pooling and the increase in knowledge spillovers. The downward trend in input-output linkages is associated with an increase in trade competition but not with a decrease in the transportation costs of goods.