Rally Effects and Michael Jackson
It is axiomatic in quantitative political science literature that there is a strong relation between foreign crises and presidential approval rating. To summarize a vast field of literature, presidents enter office with a relatively high approval rating. Over the course of a president’s administration, however, his popularity invariably trends slowly downward, and is often closely correlated with economic indicators.
Occasionally, however, a president will receive a sudden boost inpolling. Political scientists call these “rallies.” Almost every rally effect measured over the course of seventy years of extant records has followed a highly publicized international incident. The logic, therefore, goes that the public is prone to “rally around the president” in times of international crisis or war, giving the executive an otherwise improbable short time surge in popularity.
Political scientists have made careers out of defining what is and is not a “rally” and how to properly measure a “crisis” and the public’s perception of it and whether or not foreign crises stand to help or hurt a president. Because political scientists like making models that possess predictive power, some of the most sophisticated of these scholars have attempted to ascertain under what conditions a rally may be most likely to occur, employing time series models that divide years and administrations into halves and quarters. In these brackets, they track the number of crises that occur and their impact on a president’s approval rating while controlling for a host of economic, political, and international factors. The tautology inherent in this approach is evident and the methodological issues with it should be apparent to anyone with a basic grasp of quantiative literature.
Nonetheless, for the sake of this brief essay, I will put these aside and make a BOLD CLAIM
Coverage of foreign affairs usually boosts a president’s approval rating
Over the past three weeks, however, sabberrattling in North Korea, a media frenzy over Iran, and a quasi-revolution in Honduras have done nothing for President Obama’s approval ratings, in fact, they have gone slightly down, according to Gallup, the poll of record for this literature.
Why?
An easy explanation would be that an economy in a recession renders moot any nominal boost from crises abroad for a president’s approval rating. The Obama Administration has also attempted to mitigate the importance of each of these affairs and remains heavily invested in an ambitious domestic agenda. Voters, the argument would follow, simply are not paying attention to the international arena, or if they are, it is not reflected in the polls because the economy renders this an abnormal period in polling numbers. Obama recieved no boost from the Somalian pirate episode in April, which even the strictest political scientist would have to consider a prototypical matieral rally jumper.
Another explanation could be that President Obama’s approval ratings are artificially high due to his recent electoral victory and triumphalist victory tour. The luster will come off the administration at some point over the next twelve months and those supporters who are not diehards will only “approve” of the president’s performance when foreign news dominates the headlines. The president is still in his “honeymoon” phase, which makes results here atypical.
Perhaps Obama is so innately polarizing that his polling numbers are in no way correlated to anything his administration accomplishes domesitc or international. This seemed to be the case during President Bush’s second term, which would suggests a rally effect is highly contingent upon the domestic political environment, which further suggests the concept of a “rally” might be a relic of the Cold War era, with exceptions made for September 11 and the Iraq War.
Maybe with two wars going on, the new president is riding a perma rally effect? Meh . . . impossible to disprove.
Or maybe Michael Jackson and Farrah Fawcet died last week and no one is paying attention to anything else right now. In deciding what constitutes a crisis, many quantitative use of force scholars will only use incidents appearing on the front page of the New York Times as their units of analysis. This appears antiquated. Maybe scholars should look at the number of “tweets” a news item recieves to judge public perception of it?
Conclusion:
Any number of contingent explanations could inhibit a predicted rally effect, if we assume foreign crises invariably boost presidential popularity, as many political scientists do. The obvious flaw in this line of logic is that it can explain away any dissonance by saying “something else was more important,” which means making a predictive model is quite difficult. This, of course, is what historians like John Lewis Gaddis have been saying all along, that there are no “independent variables” in reality, all are dependent in some way on another one. This is not to suggest there is no value in this methodology. Quantiative studies can help a scholar only able to view a few trees to get a better sense of a forest, but to anyone familiar with a particulars of an era, explaining why or why not a president might not have recieved a boost in approval with reference to other times and other eras risks overlooking a host of otherwise imperctible proxmial causes.
