This paper asks whether voters reward or punish incumbents after natural disasters differently depending on the official’s party. Using Hurricane Sandy and longer-term historical data on U.S. presidential contests, the authors evaluate whether pre-existing partisanship conditions electoral responses to disaster damage and relief.
Examines Hurricane Sandy (2012) Comparative time series 1972–2004 Conditioning by pre-existing partisanship
Data and methods
The study first analyzes the effects of Hurricane Sandy on the 2012 presidential election at the county level, then draws on existing data covering presidential elections between 1972 and 2004 to assess broader patterns. The abstract reports comparisons of disaster-affected versus non-affected counties, distinguishing co-partisan counties (those that previously supported the incumbent’s party) from counties in the opposing party column.
Main findings
According to the abstract, reactions to disaster damage were strongly conditioned by voters’ prior partisanship: counties that previously supported Obama in 2012 "react[ed] far more positively to disaster damage than those that had earlier opposed him." Over the longer 1972–2004 series, the authors report that incumbent-party candidates "performed no worse in disaster-affected co-partisan counties than in non-affected co-partisan counties," but "underperformed in disaster-affected counties safely in the opposing party column."
