The pro-trade effect of immigration on American exports during the late nineteenth and early twentieth centuries

The belief that immigrants generate beneficial externalities in their host countries, specifically in the form of an increased opportunity and ability of firms to expand their foreign trade, has recently been challenged by George Borjas in Heaven's Door (1999, p. 97) as having no empirical support. Borjas' assertion ignores several recent papers that provide precisely that evidence of a powerful pro-trade effect of international migration. Here we extend that body of evidence by looking to history. We show that immigration, primarily from Europe between 1870 and 1910, had an important pro-trade effect on American exports. Our data set spans the exports of 44 commodities to 17 countries observed at 5 year intervals. We use a modified gravity model to examine the migrant stock-export relationship and find that United States exports to a country were positively related to the size of the migrant stock of immigrants from that country. The estimated strength of the effect varied across "Old" Europe, "New" Europe, and non-Europe groupings of the trading partner countries. Exports were also found to have been greater to English-speaking countries, and to countries with per capita incomes similar to the United States. This relative per capita income effect became stronger during the latter part of the period, whereas the migrant stock effect diminished after 1885.