Immigration and the Diffusion of Technology: The Huguenot Diaspora in Prussia

Abstract

Standard-Nutzungsbedingungen:Die Dokumente auf EconStor dürfen zu eigenen wissenschaftlichen Zwecken und zum Privatgebrauch gespeichert und kopiert werden.Sie dürfen die Dokumente nicht für öffentliche oder kommerzielle Zwecke vervielfältigen, öffentlich ausstellen, öffentlich zugänglich machen, vertreiben oder anderweitig nutzen.Sofern die Verfasser die Dokumente unter Open-Content-Lizenzen (insbesondere CC-Lizenzen) zur Verfügung gestellt haben sollten, gelten abweichend von diesen Nutzungsbedingungen die in der dort genannten Lizenz gewährten Nutzungsrechte. This paper analyzes the long-term effect of technological diffusion on productivity caused by immigration of skilled workers. In 1685 religious persecution drove highly skilled Huguenots into the backward BrandenburgPrussia where they established themselves and transferred technological knowledge to natives. We find that textile manufactories installed in towns hosting the Huguenots achieved higher productivity than others due to diffusion, even 100 years after immigration. Identification is based on an instrumental variable approach exploiting variation in the settlement of Huguenots which results from population losses due to plagues during the Thirty Years' War, effectively eliminating worries of selectivity in the settlement pattern. Terms of use: Documents in EconStor may

Unbiased Instrumental Variables Estimation Under Known First-Stage Sign

“…First, we consider the data and specifications used in Hornung (2014) to examine the effect of seventeenth century migrations on productivity. For our second application, we study the Staiger & Stock (1997) specifications for the Angrist & Krueger (1991) dataset on the relationship between education and labor market earnings.…”

“…As we can see from Other controls include a constant, a dummy for whether a town had relevant textile production in 1685, measurable inputs to the production process, and others as in Hornung (2014). As in Hornung (2014), all covariance estimates are clustered at the town level.…”

“…As in Hornung (2014), all covariance estimates are clustered at the town level. Note that the unbiased and Fuller estimates, as well as the AR confidence sets, have been updated to correct an error in the March 22, 2015 version of the present paper.…”

“…Section 4 presents simulation results on the performance of our unbiased estimators. Section 5 discusses illustrative applications using data from Hornung (2014) and Angrist & Krueger (1991). Proofs and auxiliary results are given in a separate appendix.…”

The Literary Inquisition: The Persecution of Intellectuals and Human Capital Accumulation in China

“…In contrast, Hornung (2014) examines the positive effect of the Huguenot diaspora on economic outcomes in Prussia. 10 This paper also contributes to a better understanding of the development of China in the period before the Great Divergence.…”