Poli-Sci Perspective is a weekly Wonkblog feature in which Georgetown University's Dan Hopkins and George Washington University's Danny Hayes and John Sides offer an empirical perspective on the issues dominating Washington. In this edition, Sides looks at why campaign tactics don't always work as well as we'd think. For past posts in the series, head here.
The narrative after the 2012 campaign was that President Obama's victory was due to his superior campaign -- better messaging, better technology, better organizing, better everything. But this narrative had a circular logic to it: Obama won because of his superior campaign, and we know that his campaign was superior because he won.
Several initial forays into the data on advertising and field offices suggest a much more qualified conclusion: yes, campaigning did matter, but it was not decisive in 2012. This parallels what Ezra wrote earlier last week: we got too excited over money in this election -- and by extension the electioneering it could buy. In this post, I build on some previous work for Wonkblog and work with Lynn Vavreck for our book on the 2012 campaign, The Gamble. I’ll briefly explore 4 reasons why it is hard for all that money -- in particular the ads and field organization, or what I'll call "campaign activity" -- to be the "game-changer" it is so often made out to be.
Reason #1: Campaign activity can be overwhelmed by other events in the campaign.
Throughout the 2012 campaign, candidates were spending millions every week -- blanketing key states with advertisements -- only to find that other events swamped any advantage they might have had on the airwaves. Consider what happened to Mitt Romney before his loss to Newt Gingrich in the South Carolina primary. Below are South Carolina polling and advertising data that was gathered by Kantar Media|CMAG and obtained by the Washington Post. (I analyzed these data throughout the fall of 2012 for Wonkblog.)
In the weeks leading up to the South Carolina primary, Romney was coming off a solid showing in Iowa and a victory in New Hampshire. He also aired much more advertising than Gingrich in South Carolina. But it didn't matter. As Vavreck and I discuss in our book, Gingrich's performance in the two South Carolina debates earned him favorable news coverage, and this was enough to push up his poll numbers.
The same thing happened in the general election. Obama had an overall advertising advantage in early October. But that didn't keep the first debate, plus the favorable news coverage it generated for Romney and unfavorable coverage it generated for Obama, from shifting the poll numbers in Romney's favor, even in battleground states.
In other words, advantages in campaign activity can be neutralized by other campaign events that sometimes affect the polls (if not the election outcome), like debates.
Reason #2: It can be hard to win the messaging war.
CMAG provided the Post a sense of the major issues discussed in each of the ads. I looked at those issues in the summer of 2012 -- a period when the Obama campaign's ads were thought by many to be particular potent because of his attacks on Romney’s tenure at Bain Capital. But what emerged from the summer data on advertising content was mainly two lines of argument about the economy -- with each side struggling to win the larger debate.
Below is the percentage of each candidate’s ads mentioning various issues in May, June and July. Because ads could mention more than one issue, the percentages do not add up to 100.
The economy’s preeminence as a campaign issue is plain to see. Obama argued that the economy was improving and that he deserved reelection. Romney argued that the economy was not good enough and Obama should be thrown out. Each could find reasons for both solace and concern in public opinion.
The public was pessimistic about the economy and only a minority approved of Obama's handling of the economy. That was good news for Romney. But the public also tended to blame George W. Bush a bit more than Obama and tended to believe that Obama was more in touch with the problems facing the middle class. That was bad news for Romney.
In other words, while it is tempting to think that candidates can discover some secret sauce that makes their message irresistible, they often confront a public beset by ambivalence and conflicting opinions. Crafting a message that leverages the nuances of public opinion is tricky. It's not surprising, then, that the polls during these summer months didn't move that much.
Reason #3: The effects of campaign activity are short-lived.
A few weeks before the campaign, the National Journal's Charlie Cook wrote:
If Obama ekes out an electoral-vote win, look back to last spring and summer, to the Romney campaign’s decision not to define him in a personal and positive way and the Obama campaign’s decision to roll the dice by spending an enormous amount of money to discredit Romney in the swing states, as the factors that led to the outcome.
There are two reasons why this is probably not true. First, the ads Cook refers too -- attacking Romney's tenure at Bain Capital -- did not air for all that long. They were prominent battleground state campaign activity and in news coverage for only a brief time.
Here is the percent of all Obama ads mentioning Bain and the percent of all news media mentions of Romney that included mentions of Bain, drawing on the media coverage data gathered by General Sentiment.
This graph shows that Romney's time at Bain Capital was a major part of advertising and news coverage only in July In part, this was because the Obama campaign emphasized Bain less as the general election campaign heated up. In part, this was because the news coverage of Bain Capital -- stimulated in July largely by a Boston Globe story – waned as the media moved on to other topics.
Moreover, as ours and others' analyses of 2012 advertising effects as well as previous research demonstrate, the effect of ads dissipates quickly -- within a few days or even a day. Cook's characterization suggests that your television is like a nuclear reactor and the summer Obama ads were essentially irradiating you and permanently altering your (political) body chemistry. In reality, ads are more like Tylenol: they work for a short time and then they wear off and you need to see more.
The same thing holds for other kinds of campaign activity as well. In Sasha Issenberg's The Victory Lab, one member of Obama's 2008 campaign team, Aaron Strauss, put it this way:
The effects of any campaign activity are ephemeral. Just because you touched someone in, let’s say, the beginning of October doesn’t mean they’ll be with you two weeks or even one week later.
Which leads to the final point...
Reason #4: In most presidential elections it is hard to get a big advantage in campaign activity.
Even if ads are like Tylenol or campaign activity has an ephemeral impact, a candidate could still benefit if he had a sizable advantage in campaign activity day after day. But that was hard to come by in 2012, particularly with regard to advertising.
Even though Obama had an overall advantage for a sizable part of the campaign, and Romney had the advantage at the very end of the campaign, it was relatively rare for either of them to have a large advantage in very many places. For example, Lynn Vavreck and I found that if on the day before Election Day, Romney had an advertising lead in a media market that amounted to an additional three ad viewings per capita, he would have received an additional three-tenths of a point of vote share. That was not enough to make up his deficit in key battleground states, but neither was it nothing. The bigger problem was just that he had an advantage that size or greater in only 15 percent of battleground state counties.
When both candidates and their allies have a billion-plus dollars to spend, each side's efforts mostly neutralize the other's. That is what seemed to happen in 2012. Both campaigns were sufficiently effective that any effects of campaign activity were not large enough to be decisive. The growing economy set the stage for an incumbent party win, and the two campaigns ultimately reinforced that fact.
So why bother to raise and spend all this money? The problem is that letting your opponent get a larger advantage is exactly what might hurt you on Election Day. No one wants to be Al Gore and lose the Electoral College by 537 votes. In essence, presidential candidates raise money to ensure a tie -- or at best gain the sort of small edge that wasn't decisive in a race like 2012, where the final margin simply wasn't that close. But a small edge could be decisive in elections like 1960, 1968, or 2000. In fact, not all of those three elections were initially predicted to be that close. It was the candidates who helped make them close.
So, where’s the smart money on campaigns making a difference? Campaign activity can decide an election’s outcome. But it likely didn’t in 2012.