The Long-run Impact of Historical Shocks on the Decision to Migrate: Evidence from Irish Migration

This study investigates how negative historical shocks can explain migration in the long-run. We construct a unique dataset based on the 1901 and 1911 Irish Census data and a selection of the Ellis Island Administrative Records which allow us to test whether the Great Irish Famine (1845-1850), one of the most lethal starvation in history, has shaped the decision of migrating to the USA in the following 60 years. We control for several set of individual and geographical characteristics and we find that the Irish Famine was an important significant driver of individuals’ migration choices. Instrumental variable analysis based on the exogenous spread of the potato blight provides consistent results. The results are supported by various instrumental variables regressions based on the exogenous movement of the potato blight. We are also interested in understanding how the potential long-run impacts of the famine can impact the decision of migrating. The mechanism we have in mind and would like to test is that the migration networks formed soon after the famine still have the driving force two generations afterwards. More in detail, during the 50-70 year period followed by the Famine, we believe that the Irish counties hurt most had more short-term migration, which created even more migration due to the network effect and there has been an accumulated network effect. We suppose that county level inequality has resulted in bigger network effects for counties hurt most. Furthermore, there might have been a famine-induced increase in county level inequality. This has resulted in more migration from the counties that have been hurt most and "trapped" in a bad economic situation, resulting in more emigration.