The Lasting Electoral Effects of Natural Disaster Responses

As President Obama visits Memphis and promises relief to the victims of the Mississippi floods, we may ponder what the electoral consequences of the federal response might be. Dan Hopkins has blogged earlier that voters distributed the blame for the Katrina response across different political actors. Yet, the Republican party still suffered considerably in the presidential race. But what if the disaster response had been effective?

A new paper (ungated) by Michael Bechtel and Jens Hainmueller forthcoming in the American Journal of Political Science shows that the electoral bonus of good responses to natural disasters can be substantial and can last beyond the first election following a disaster. Bechtel and Hainmueller examine the response to the 2002 flooding of the Elbe river in Germany. The government responded with a rapid mobilization of rescue forces and was quick to disburse large amounts of relief aid. Bechter and Hainmueller estimate that the main incumbent party, the Social-Democratic SPD, received a seven percentage point increase in their vote share in the affected areas during the 2002 elections. About a quarter of this bonus was still present in the 2005 elections although it had dissipated by the 2009 elections. Further tests show that most of this effect was due to persuasion of existing voters rather than mobilization of new voters.

The difference between a good and poor response would likely be larger than the seven percentage points estimated here. It is, of course, difficult to generalize across countries and disasters but it nonetheless suggests that aside from the humanitarian stakes the electoral stakes are also high. The abstract is below:

Dominant theories of electoral behavior emphasize that voters myopically evaluate policy performance and that this shortsightedness may obstruct the welfare improving effect of democratic accountability. However, we know little about how long governments receive electoral credit for beneficial policies. We exploit the massive policy response to a major natural disaster, the 2002 Elbe flooding in Germany, to provide an upper bound for the short- and long-term electoral returns to targeted policy benefits. We estimate that the flood response increased vote shares for the incumbent party by 7 percentage points in affected areas in the 2002 election. 25 percent of this short-term reward carried over to the 2005 election before the gains vanished in the 2009 election. We conclude that, given favorable circumstances, policymakers can generate voter gratitude that persists longer than scholarship has acknowledged so far, and elaborate on the implications for theories of electoral behavior, democratic accountability, and public policy.

[Cross-posted at the Monkey Cage]

Erik Voeten

Erik Voeten is the Peter F. Krogh associate professor of geopolitics and global justice at Georgetown University.