The Blog

Has Superstorm Sandy Chosen Who the Next US President Is?

Extreme disasters which occur just before an election, do appear to sway voters, the latest research has found.

Extreme disasters which occur just before an election, do appear to sway voters, the latest research has found.

Andrew Healy and Neil Malhotra from Loyola Marymount University and Stanford Graduate School of Business, recently published a study where they examined the effect of tornado damage across the USA on presidential voting, from 1952-2004.

They found in the year preceding a presidential election, that a 10% increase in financial cost of tornado damage, causes the incumbent party to lose approximately 0.15 percentage points of the vote. The incumbent party did, on average, 2.02 percentage points worse in the 50 highest-tornado-damage counties than it did nationally in those years.

The study entitled 'Random Events, Economic Losses, and Retrospective Voting: Implications for Democratic Competence' argues that if random events beyond the control of a politician can dramatically influence an election, then this has ominous implications for democracy.

If voters choose irrationally, democracies deliver bad Governments.

Superstorm Sandy therefore provides a natural experiment on the psychology of voting. Many political psychologists believe events such as natural disasters prove the electorate is fundamentally 'emotional', rather than 'rational', when it chooses between candidates.

For example, Christopher Achen and Larry Bartels published a Princeton University Working Paper entitled "Blind Retrospection: Electoral Responses to Drought, Flu, and Shark Attacks", they concluded; "voters regularly punish governments for acts of God, including droughts, floods, and shark attacks".

Andrew Healy, Neil Malhotra and Cecilia Hyunjung Mo recently published a study on voting behaviour where they investigated the electoral impact of local college football games just before a US election. They argued such games are irrelevant events that government has nothing to do with, and for which no political response could be expected.

Between 1964 and 2008 they found that a local college football win in the 10 days before Election Day, causes the incumbent to receive an additional 1.61 percentage points of the vote in Senate, gubernatorial, and presidential elections.

The study entitled 'Irrelevant events affect voters' evaluations of government performance' found the effect of football results on elections is larger where the teams are more locally important, and the fans more passionate.

This powerful effect is well known in psychology as a human tendency to 'transfer emotions' from one part of life to a completely separate area. For instance, in one famous psychology experiment, after being given a free gift, people were more likely to claim that their automobiles and TV sets performed better and required fewer repairs.

Healy, Malhotra and Hyunjung Mo replicated their football finding by examining a 2009 national men's college basketball elimination tournament. One key result from the study published in 'Proceedings of the National Academy of Sciences', is that when your team does unexpectedly well, therefore producing a bigger boost in mood, such wins significantly increased approval of President Obama.

This effect of mood transfer influencing voter decisions is below conscious awareness and therefore could be particularly prone to manipulation by politicians. How to resist the machinations of spin doctors is also suggested by the experiment.

Some were randomly but directly given information about the outcomes of their team's games. Making subjects aware of the reasons for their mood decreases the tendency to mistakenly attribute those moods to irrelevant events. Those who were explicitly told the scores, showed no impact of the positive mood induced by the team win on their political evaluation of Obama.

If respondents were not explicitly told their teams score, their approval rating of Obama hiked by 4.6 percentage points. The results show that making the game outcomes more salient, eliminated their influence. By moving subconscious considerations into consciousness, this allows people to cease irrationally transferring mood. Their team's fortunes stopped influencing their political judgment.

Psychology is going to be a vital part of political and psychological strategy in the way rival candidates strive to not be blown off course by Superstorm Sandy. But voters can resist manipulative tactics.

In the tornado study, Healy and Malhotra argued that if incumbents are punished for deaths related to natural disasters, voters are reacting emotionally to the disaster. Fatalities, they contend, are generally caused by the strongest tornadoes and thus are largely not preventable by government policy. There is also less scope for effective official responses to address fatalities, than in the case of monetary damage.

Healy and Malhotra compared whether number of deaths caused by tornadoes, or economic damage, was more important to voters. They found economic damage affects incumbent vote share, but deaths do not. Voters punish incumbents not just for damage inflicted on themselves or in their own county, but also for harm in nearby counties--which they can observe and use to gauge government performance.

However, the study, published in the 'Quarterly Journal of Political Science', finds that when the President issues an official disaster declaration, thereby demonstrating responsiveness to the tornado and releasing federal funds, tornado damage actually increases vote share.

Conversely, when no such declaration is issued, voters punish incumbents at the ballot box for economic losses. These results suggest voters do not punish governments for circumstances beyond their control in a knee-jerk emotional response. Rather, the electorate is influenced by whether the incumbent robustly responds to the disaster.

Superstorm Sandy is being compared to Hurricane Katrina which had a massive impact on the US population blaming various levels of Government for failures in response - but then as now - will Democrats simply find Republicans to blame and vice versa?

Neil Malhotra and Alexander Kuo from Stanford University designed an experiment in which respondents ranked seven public officials in order of how much they should be blamed for the property damage and loss of life in New Orleans after Hurricane Katrina. Information provided was manipulated on purpose, with some respondents receiving the officials' party affiliations (Democrat or Republican), others receiving their job titles.

The study entitled 'Attributing Blame: The Public's Response to Hurricane Katrina' found party affiliations do cause individuals to tend to blame officials of the opposite party. But the authors also argue their results suggest citizens are not absolute and blind party loyalists. Instead, respondents also took into account the positions of the officials, and used that information to form a reasoned opinion.

Decision making for voters in a national campaign where there is no recent natural disaster, like a Superstorm, is beset with difficulties. The nation's economic and political performance is not just complicated, but who to blame for what, is also thorny.

In contrast, a major environmental tragedy may provide a useful natural experiment on which to judge a Government. Their specific reaction is more easily observed to such a discrete event, compared to their response to long-standing political or economic predicaments.

Might voters not in fact be at their best in terms of making better more rational decisions, when evaluating a candidate on their response to a natural disaster?

In which, case elections should be routinely held just after major catastrophes. Then, voters might elect better, more effective governments.