Weatherproofing our elections: There ought to be a law — and there is

Everyone complains about the weather, but few do anything about it.

This may not be true for local election officials’ response to the challenges Hurricane Sandy posed in conducting the 2012 election. Researchers have consistently found that inclement weather on and before Election Day has had a significant and negative effect on voter turnout; this was no less true in jurisdictions most adversely affected by Hurricane Sandy last fall. In a report prepared for the Presidential Commission on Election Administration, I found voter turnout declined 2.8 percent between 2008 and 2012 in counties most impacted by the storm. This effect persists when controlling for a host of factors specific to the 2012 election, individual counties and their respective states.

So what, if anything, did local election officials do to mitigate the negative effect the hurricane had on voter turnout? Local elections officials in states and counties affected by Hurricane Sandy found a way to mitigate the hurricane’s depressing effect on voter turnout. The negative effect the storm had on voter participation was significantly muted by two conditions in these states and counties: early voting and the consolidation of the Election Day voting places.

Among counties most adversely affected by Hurricane Sandy, voter turnout declined significantly less in areas that afforded their voters opportunities to vote before Election Day (i.e., early in-person voting). Being able to vote several days before Election Day and at a host of different locations gave voters many more opportunities to vote, especially when voters’ assigned voting locations may have been inaccessible and unusable. The availability of early voting in counties adversely affected by the hurricane, however, was limited to just one state, Maryland, who adopted early voting for the first time in 2012.

Local election officials also had significant success at reducing the negative effect Hurricane Sandy had on voter participation. By either choice or necessity election officials in counties hit hardest consolidated the number of Election Day polling places. In these counties, the mean number of polling places per 1,000 voters was reduced by 10 percent, while the number of poll workers per voting place increased by more than a third. The result was a smaller number of larger voting locations with more staff to help voters. This dramatic change in the configuration of polling places had a positive effect on voter turnout. Moreover, offering a smaller number of larger voting places had a positive effect on voter turnout in all other U.S. counties unaffected by Hurricane Sandy, albeit a much smaller effect than was observed in affected counties. Why?

A small number of larger polling venues increases voter access to polling places and produces a more efficient processing of voters. Elsewhere my co-author Greg Vonnahme (Stein and Vonnahme 2008;2012) and I have shown larger and more visible polling sites can reduce informational costs voters incur when attempting to find a polling location. More parking spaces, voting machines and poll workers reduce the time it takes voters to cast a ballot. With more staff at each polling location, poll workers are able to specialize in certain tasks, such as checking in voters or assisting them with their ballots, which should lead to more efficient operations and improved service to voters. The efficient use of poll workers should aid errant voters needing information about where they can vote on or before Election Day and what information they need to establish their identification.

In the aftermath of natural disasters and emergencies, we might expect larger and more centrally located polling places (e.g., hotels, supermarkets, stadiums and large public buildings) to be more accessible and powered. Anecdotal evidence suggests county election officials in hurricane-affected areas relocated small polling places at neighborhood elementary schools to larger, centrally located facilities where voters had greater access. Whether this was intentional or necessitated by hurricane damage is uncertain. What is clear is that the reconfiguration of voting locations can — and did — have a mitigating effect on Hurricane Sandy’s negative impact on voter turnout.

Next time you complain about the weather, ask your local election officials what they can do about it.

For more information, read:

Robert M. Stein and Greg Vonnahme,Engaging the unengaged voter: Voter centers and voter turnout,” Journal of Politics. 2:487-497 (April 2008).

Robert M. Stein and Greg Vonnahme “The effect of election day vote centers on voter participation,” Election Law Journal, 11(4):291-301 (September 2012).

 Predicted 2012 Voter Turnout (%)

Non-disaster counties                                          Disaster counties











[Shaded area represents 95 percent confidence interval.]


Robert Stein is the Baker Institute fellow in urban politics and the Lena Gohlman Fox Professor of Political Science at Rice University.