That Lingering Whiff of Scandal Lasts About 4 Years
If politicians can survive the initial storm over their bad deeds, it takes roughly four years before voters will forgive and forget.
Just for the sake of argument, let’s say you’re a seven-term congressman and you’ve got a little problem with the House Ethics Committee. How long before the folks back home forgive and forget?
About two terms, according to a new study in the Social Science Quarterly. This assumes that you stick around—about a quarter of House members resign or retire when scandal comes a’knockin'—and that you’re among the half who survive the next election.
Given that initial election to the House is as good a promise of continued employment as modern America can provide these days, such results are pretty telling, even if it seems obvious that a scandal ought to get you booted out. And it puts an expiration date on Fitzgerald’s old (and misused) dictum about second acts in American life.
“Incumbents involved in scandals were about three times more likely to resign or retire and three times more likely to be defeated in the general election.”
For their research, University of Connecticut political scientists Rodrigo Praino and Vincent G. Moscardelli, along with Daniel Stockemer from the University of Ottawa, looked at all the U.S. House of Representatives races between 1972 and 2006. In that period, 88 members had issues that were referred to the Ethics panel, which was the researchers’ metric for a scandal. That won’t net every observed transgression, especially those with a sexual edge—Anthony Weiner’s softcore escapades in 2011, as an example outside the study period, saw him resign before anything hit the committee—but it does ensure, the authors write, that the allegations are publicized and “reasonably founded.”
An ethics probe has never been an automatic bar to retaining a seat, but it does hurt. “Incumbents involved in scandals” according to the study, ”were about three times more likely to resign or retire, three times more likely to be defeated in the general election, and 11 times more likely to be defeated in a primary than other incumbents.”
Given Americans’ famously short attention spans, the trio of political scientists asked not just how much the scandal hurt in the very next election cycle, where the damage has been studied before, but in subsequent election cycles. (Remember that all 435 seats in the House are voted on every two years.) On average, they found the incumbent’s election margin—the difference between the incumbent’s vote total and the challenger’s—dropped 12 percentage points in the immediate race and, assuming the wounded candidate retained the seat, five percentage points in the next. “It is not until the second postscandal reelection bid (year 4) that the average scandal-plagued incumbent returns to his or her prescandal margins,” they wrote.
(It’s not quite confirmation, but New York Times’ stats wizard Nate Silver points out that four years after Mark Sanford was shamed out of South Carolina’s governorship, he won a House seat there by about 13 percentage points fewer than party registration would have suggested.)
That a candidate could lose a dozen percentage points and still win would be amazing in more competitive races, like the presidency, where it’s been just shy of 30 years since the winning margin was greater than 12 points. But in the average House race, the margin of victory was about 33 points. Talk about a victory for those who argue for safe seats!
The researchers did factor in a number of controls in generating their results, including the partisanship of a district and of the candidate, how well a candidate did in elections before the scandal broke, campaign spending by both sides, and how strong a challenger was put up by the opposition. They also controlled for seniority—a lousy human being who brings goodies to the district, after all, is a lousy human being we can re-elect—and for the fact that while re-election is the rule, victory margins do taper off as House members grow longer in the tooth. Scandals on average pop up in the seventh term, the researchers learned.
A scandal also increases voter turnout at the district level. The bump is a modest (and statistically insignificant) 1.6 percent, but all the increase goes to the opponent.
If we assume that anything that doesn’t kill a career makes it stronger, almost everything about a scandal seems modest with the passage of time. Right after they set out that the aftertaste of scandal depresses expected margins of victory for four to six years, the political scientists noted, “Incumbents who survive their initial reelection bid do recover a substantial amount (just under two-thirds) of the immediate loss after just one cycle, quickly moving the average incumbent back out of any ‘danger’ zone into which he or she might have tumbled in the immediate aftermath of the scandal.”
This paper, of course, was focused just on the U.S. House of Representatives, which only covers some American political scandals, i.e. Rangel and Renzi and Cunningham and DeLay and Gingrich, but not others, i.e. Vitter and Spitzer and Sanford. But that idea of four years or so for political reconciliation seems like a handy rule of thumb: Mark Sanford went from doghouse to the House (presumably via the Appalachian Trail) in that span, while headlines this month reported that George W. Bush’s popularity has rebounded from its 2008 low.