Why Obama’s poll numbers didn’t get a boost from the Libya intervention.

A week into American and allied action in Libya, one political result is already clear: Barack Obama has not benefited in the polls. If anything, Obama’s Gallup approval numbers are actually down a few points since American involvement in Libya began.

We can look to political science to understand this trend—specifically, to the idea of the “rally around the flag” effect. A rally effect, by definition, is when a president’s approval numbers increase during a national security event. Unfortunately for Obama, there’s been no rally effect this week. Which, of course, begs the question: Why?

Perhaps the most famous rally effect, and certainly the most pronounced, was from the September 11 attacks in 2001: George W. Bush’s Gallup approval rating immediately went from 51 percent to 86 percent—on its way to a record 90 percent approval recorded September 21-22, 2001. Some have speculated that such effects are caused by patriotic reactions to wars and other national security events: Perhaps Americans naturally switch from thinking of the president as a party politician to imagining him as the symbol of the nation in times of trouble. If that was so, however, all similar events would cause positive reactions—and that turns out not to be true. For example, neither of two Libyan events during Ronald Reagan’s presidency—the downing of two Libyan fighter planes in August 1981 and the bombing in 1986—produced jumps in Reagan’s approval numbers.

Nor does it turn out that successful military actions produce better approval ratings than failures. During the Carter presidency, two fiascos—the initial hostage-taking in Tehran, and the failed rescue attempt—actually helped Carter’s approval ratings, at least initially. So did the Bay of Pigs for John F. Kennedy, and the U-2 incident for Dwight Eisenhower.

So what does cause a rally effect? In 1989, Stanford political scientist Richard Brody discovered that there was one key variable: the reaction of politicians and leaders in the party out of power. Brody looked at 45 possible “rally” situations, from the announcement of the Truman Doctrine in March 1947 through the Iran-Contra affair in November 1986, checking the Gallup polls before and after the event and examining coverage in major newspapers and on network television. If elites in the party out of power supported the president, the result was a rally effect in the polls; if not, there was no effect. So, for example, Brody found that, after the downed U-2 was first revealed on May 6, 1960, the first negative comment from a Democratic leader didn’t show up in The Washington Post until May 17, and didn’t make it to the front page for several days after that. Ike’s approval stood at 62 percent prior to the news, and rose to 65 percent the next time Gallup checked, in late May. Similarly, after the Bay of Pigs, Dwight Eisenhower, Richard Nixon, Barry Goldwater, and Nelson Rockefeller “publicly pledged full support to Kennedy”—and his approval rating went from a honeymoon 78 percent to an even higher 83 percent. On the other hand, Republicans began attacking Lyndon Johnson’s handling of the Pueblo incident in January 1968, four days into the story, and Johnson’s approval ratings fell seven points.

That opposition leaders matter makes sense in light of what is generally known about public opinion. Most people, on most issues, follow opinion leaders rather than learning the facts of situations and coming to conclusions on their own. That’s particularly likely to be the case when people have little or no prior knowledge of a subject—which is true for most people on most subjects, but is particularly true about foreign affairs. Very few Americans had opinions about humanitarian interventions and Libya before last week, or about the military risks of establishing a no-fly zone, or about likely reactions from other regional actors. The low-cost way of forming opinions about such things, should they be needed (say, because suddenly Libya is all over the news and a pollster is asking), is to find sources one trusts and adopt their views.

Why are opposition leaders the key here? Virtually all people in the president’s own party will look to him for opinion leadership, and therefore support him, in these situations. And, in most cases, adherents of the party out of power will continue to dislike the president and disapprove of his job performance, whatever the initial reactions of any politicians—even those of their own party. But true independents and others who may have mixed feelings on the president will take their cue from whether or not there is unanimity among opinion leaders. If there is unanimity, they are cued to believe that the president’s actions are uncontroversial and acceptable.

To apply this to the current situation, when the United States acted in Libya, many prominent Republicans criticized the intervention—as they were likely to do, considering the extreme, anti-all-things-Obama sentiment that’s taken hold of the GOP. Newt Gingrich, for one, called the operation as “badly run as any foreign operation we’ve seen in our lifetime.” Speaker of the House John Boehner, meanwhile, said he was “troubled” by the fact that Obama undertook action “without clearly defining” the mission. Per Brody’s theory, Obama hasn’t gotten a poll bump.

Looking to the future, as the U.S. political scene becomes increasingly hyper-partisan—with each side of the aisle refusing to find common ground on practically any issue—we can expect that rallying around the flag will become less and less likely. Obama’s polling numbers after the Libya intervention could be just the tip of a future iceberg.

Jonathan Bernstein blogs at A Plain Blog About Politics.