Moving to Higher Ground: Migration Response to Natural Disasters in the Early Twentieth Century

Abstract

Areas differ in their propensity to experience natural disasters. Exposure to disaster risks can be reduced either through migration (i.e., self-protection) or through public infrastructure investment (e.g., building seawalls). Using migration data from the 1920s and 1930s, this paper studies how the population responded to disaster shocks in an era of minimal public investment. We find that, on net, young men move away from areas hit by tornados but are attracted to areas experiencing floods. Early efforts to protect against future flooding, especially during the New Deal era of the late 1930s, may have counteracted an individual migration response.

Do Floods Have Permanent Effects? Evidence from the Netherlands

“…One possible explanation is that the occurrence of a disaster provides an impetus for governmental action aimed at reducing hazard. Boustan et al (2012) argue that government disaster relief programmes may partly have cancelled out individual responses, leading people to return or move to hazard-prone areas. Furthermore, flood mitigation measures in the form of large infrastructure projects may have a transformative role on local economies.…”

“…Moreover, although large scale natural disasters are rather unique events, especially in developed countries, if a disaster does occur it directly creates a demand for policy intervention to mitigate the impact of the disaster, including any potential future disasters, possibly further increasing the clustering of population in risky areas (Boustan, Kahn, and Rhode, 2012;Kahn, 2005). This raises the question to what extent governmental disaster protection is actually partially counteracted by individual responses as revealed by household location decisions.…”

The impact of natural disasters on energy consumption: An analysis of renewable and nonrenewable energy demand in the residential and industrial sectors

“…First, "creative destruction" leads to capital upgrading, as infrastructure that has been destroyed during a natural disaster is rebuilt (see Skidmore and Toya [1], Hallegatte and Dumas [2], and Noy and Vu [3]). The second hypothesis that the economic slowdown associated with a natural disaster is followed, at best, by convergence to predisaster levels (see Smith et al [4], Vigdor [5], Belasen and Polachek [6,7], Hornebeck [8], Strobl [9], and Boustan, Khan and Rhode [10]). We extend the above literature that has explored the impact of natural disasters on GDP, to study the impact of natural disasters on energy consumption in the short-and long-run.…”

Evolutionary modelling of the macro-economic impacts of catastrophic flood events

“…Secondly, self-protection measures may include migration due to public perception where flood risk levels are considered unacceptably high (and as a result reduce self-protection costs, for instance in embankment systems, to zero). The attraction of living along the coast has encouraged, for example, many people in the past to move to areas at risk from flooding, increasing the economic costs of disasters (Boustan et al, 2012).…”

“…The main challenge ahead lies in integrating different types of non-economic knowledge and information, such as geophysical land use data, with economic data and equations related to changing risk perceptions and behavioural adjustments. Exposure to natural disaster risk such as floods, and ultimately economic damage costs, are a function of both an individual's private choices and government decisions over land use zoning and infrastructure investments (Hallegatte and Dumass, 2009;Boustan et al, 2012). Empirical studies emphasize the role of socio-economic factors (personal characteristics, risk perception, behaviour in relation to flood damage) in reducing flood damage, in addition to structural measures (Shaw et al, 2005;Botzen et al, 2009).…”