"In the long-run, the effect of immigration on native wages is moderated by the pace of new capital investments. Historical immigration flows contributed to the transformation of American manufacturing from small-scale artisanal shops to large factories engaged in mass production (Hirschman and Mogford, 2009). Unskilled immigrants appear to have been complementary with investments in assembly-line machinery (LaFortune, Tessada and Lewis, 2014). Kim (2007) shows that firms in counties with a higher share foreign born in 1920 were larger, more productive, and more likely to be organized as factories; he uses the settlement pattern of immigrants in 1850, as well as distance from the port of New York, as instruments for the later immigrant share of the population. Yet, Hatton and Williamson (2006) argue that land was a more important input in production in the past and hence capital mobility was less effective in dampening the wage effects of migration. Historical immigration was also associated with higher rates of both trade and innovation, which may have contributed to economic growth. Dunlevy and Hutchinson (1999) find that immigrants increased trade flows between the US and Europe in the early twentieth century, perhaps by providing information about and network connections to their home markets (see also O'Rourke and Williamson, 1999, chapter 13). Moser, Voena and Waldinger (2014) analyze one immigrant flow that would be particularly expected to increase innovation in the US economy: the arrival of Jewish scientists forced to flee from Nazi Germany in the 1930s. US patenting in categories associated with the dismissed scientists increased by 30 percent after 1933."