Geographic clustering and productivity: An instrumental variable approach for classical composers

Abstract

It is difficult to estimate the impact of geographic clustering on productivity because of endogeneity issues. I use birthplace–cluster distance as an instrumental variable for the incidence of clustering of prominent classical composers born between 1750 and 1899. I find that geographic clustering causally impacts productivity: composers were writing around one additional influential work every 3 years they spent in a cluster. The best composers and those who migrated to Paris appear to be the greatest beneficiaries of clustering. Placebo tests disclose that the effects are attributable to locating in contemporaneous cluster cities, as opposed to historical cluster locations or large cities in general.