The weather was clear all across Massachusetts and New England, perfect for voting as far as the crest of the Alleghenies. But from Michigan through Illinois and the Northern Plains states it was cloudy: rain in Detroit and Chicago, light snow falling in some states on the approaches of the Rockies. The South was enjoying magnificently balmy weather which ran north as far as the Ohio River; so, too, was the entire Pacific Coast. The weather and the year's efforts were to call out the greatest free vote in the history of this or any other country.
—Theodore H. White, 1960, from The Making of the President
Political pundits devote countless hours to analyzing the factors that go into deciding people's vote on election day. Race, geography, social values, and financial concerns are some of the most common factors that come up for discussion. But what about the weather? We have only to look back to the recent Republican Convention in August 2012 to understand that just the threat of storms such as Hurricane Isaac influence the schedule of events that lead up to a Presidential election. And for the election itself, the weather in early November offers a full ballot of propositions to consider across the nation, ranging from tropical storms to early blizzards.
Early American Voters
In rural America of the early 19th century, with the harvest complete, a landowner might well have departed his farm to vote at a courthouse the Monday before a presidential election—held on the first Tuesday after the first Monday in November, as it still is today. In fact the weather and changing seasons helped to initially determine a specific date for presidential elections, which was set down in an 1845 law. In early agrarian America, the harvest needed to be complete, no one could ride to town on the Sabbath, and travel would become treacherous and voter turnout light with the arrival of the icy gales of late November. So the date was determined to be the Tuesday after the first Monday in early November.
As painted by George Bingham, elections in Missouri and many other states in the late 1840s saw male land owners calling out their choice for office outdoors on the court house steps. The secret ballot was a decade or two away, and while the “viva voce,” or word of mouth, vote offered transparency to the process, the townspeople heard your very public vote. In the Bingham painting, one wonders: If the cumulus clouds in the distance built into thunderheads, did the polls close early?
Caption: County election painting by Bingham.
Weather has likely influenced election turnout, for various and complex reasons throughout the history of presidential elections. Americans vote state by state for representatives to an electoral college to determine the next president. More often, the overall popular vote does not produce a close tally. But in the 2000 Presidential election we remember that was not the case.
Foul Weather as a Friend
Political scientists, pundits, and politicians have long speculated on the weather's influence on Republican and Democratic turnout, but no one has proven definitively that weather decided a close election and probably no one ever will. Elections, like weather patterns, are too complex.
One of the oldest political theories says that Republicans “should pray for rain,” since the wet weather will ostensibly discourage older and poorer citizens (predominantly Democrats) from venturing to the polling place on Election Day. In 2007, three political scientists published a study of the nation's weather in presidential elections from 1948 to 2000, where they concluded that rain in some Florida counties in the 2000 election may have deterred voter participation and influenced the final result to the benefit of the Republicans. Other political scientists have published different conclusions on the effect of rainy weather (if any) on turnout by political party preference. Small sample sizes in these studies make it difficult to draw conclusions that would apply to future election days in November. In addition, unlike the postman and his appointed rounds, voters could potentially be kept from the polls by all extremes of snow, rain, heat, or gloom of night. Landfalling hurricanes in early November, while relatively rare, do occur as often as every other year, impacting Florida (Hurricane Mitch in 1998) and the Gulf Coast states.
Caption: At one time a Category 5 Hurricane, Mitch crossed Southern Florida on November 5th, 1998 as a tropical storm.
Caption: GOES East Satellite Image of Hurricane Noel on November 2, 2007. The storm became extra tropical but brought hurricane force winds to the New England Coast followed by power outages.
Late October to early November snowstorms, such as the famous Minnesota Halloween Blizzard of 1991, while also uncommon, could certainly disrupt the opening of polls in a Presidential election year. Temperatures in the wake of the Halloween blizzard in both Minneapolis and Duluth fell to below zero.
Caption: Northern Minnesota snowfall totals from the 1991 Halloween Blizzard.
Caption: November 4, 1991 temperatures remained below freezing in the wake of the Northern Plains Halloween Blizzard.
A Wary Eye on Weather Patterns on Past Election Days: 1916, 1960, and 2000
Prior to 1824, not everyone in the United States voted on the same day for President. Weatherwise founder and weather historian David Ludlum, in his book The Weather Factor (1984), compiled a thorough analysis of weather and presidential elections back to William Henry Harrison in 1840—before weather maps were drawn and systematic weather observations were available. Ludlum's description of the weather on Election Day in 1916 in California makes a case for the influence of extreme weather on elections. A winter storm brought 15 to 20 inches of snow to the mountainous counties of the northeast portion of the state, which was predominantly democratic at the time. Woodrow Wilson emerged with a narrow majority of about 2,000 votes in California—deciding the state and the general election. A slight shift in the timing or intensity of the storm could have further diminished the turnout in the mountainous districts, changing the election result to Charles Evan Hughes.
Caption: Daily weather map from the U.S. Department of Agriculture for Election Day, November 7, 1916. Precipitation areas are shaded.
Ludlum also analyzed the weather nationally in the 1960 election, when Kennedy defeated Nixon by only 84 electoral votes. Illinois was a particularly contentious state, under the influence of cold frontal rain on November 8. Ludlum concludes that the rain may have kept voters in predominantly Republican rural areas from getting to the polls, but not cut down on the Democratic turnout in Chicago and Cook County.
The weather map for November 7, 2000'the Bush versus Gore Election Day—depicts a deep northern plains low pressure system that produced one-inch snow amounts for portions of the Dakotas, Nebraska, and Montana. Rain extended from the Gulf Coast through the Ohio Valley. Nationally, voter turnout was high, with 55 percent of the eligible voting population casting ballots. In the states where snow fell, voter participation was near or above 60 percent, and the margins of victory were greater than five percent. Rainfall in Florida, the state with the closest election result, was relatively light and confined to the northeast Atlantic coast. This was not a presidential election where a Hurricane Mitch or a Halloween Blizzard made it difficult for citizens to reach their local precincts to cast a vote.
Caption: A cold front moving into the Mid-Mississippi Valley on Tuesday November 8, 1960, made for a rainy election day from East Texas to the Great Lakes.
Indoor Weather and Election Technology
Weather and climate affect voting systems. High humidity can affect the thickness of paper ballots that are run though computerized optical scanning systems, and ballots can be misread if they do not feed through the scanner correctly. In low humidity conditions, static discharges between a computerized voting system and a voter's hand can initiate software failures in electronic voting systems. Enacted after the disputed 2000 Presidential election, the Help America Vote Act of 2002, also known as HAVA, resulted in updated federal equipment standards for voting systems, including environmental and reliability standards. HAVA addressed usability, computer security, and testing of voting systems, and mandated a laboratory testing program. All of these measures will address the potential effects of localized weather conditions on election equipment.
The HAVA also created the Election Assistance Commission (EAC) to oversee the distribution of funds to states for new voting equipment to replace punch cards and lever machines, as well as the development of voluntary standards for voting equipment in future elections. In order to address weather concerns, the EAC also made available guidance to the states on emergency procedures for elections in advance of and during severe weather events including hurricanes, tornadoes, snowstorms, and floods.
21st Century Voters
Let's fast forward from the 1840s to the 2008 election. Determined voters in Richmond, Virginia, waited in the rain at 6 a.m. on Tuesday, November 4, to cast their votes in the presidential contest between Senators John McCain and Barack Obama, substantiating the point that a motivated voter cannot be deterred from casting his or her vote. Since 2004, a majority of states have liberalized early voting laws, and as many as 30 percent of voters are expected to cast their ballots before Election Day, November 6, 2012, either by mail or in person at early voting locations. Disinterested voters, who are a minority, will continue to use weather as an excuse not to make it to the polls. Nevertheless, extreme weather, particularly a tropical storm or hurricane such as Noel on November 2, 2007, could severely curtail voting on an Election Day. Voting on the Internet still presents computer security challenges and is not likely to offer a solution in the short term to braving snow, rain, heat, or gloom of night on Election Day. Weathering elections, or more accurately, determining the influence of weather on election outcomes, is a hot topic for political discussion, and remains a challenge to be resolved.
ALLAN EUSTIS is a past editor of Weatherwise Magazine, an AMS Certified Broadcast Meteorologist and former Industrial Meteorologist for the National Weather Service. He has worked in local, state and national elections as a poll worker and precinct captain for over a decade.