With President Barack Obama currently in the process of filling vacancies in his cabinet—already one current United States Senator has gotten the nod—and with the retirement of Senators Rockefeller, Harkin, and Chambliss, there has been a good amount of chatter about the ramifications of opening up a U.S. Senate seat. While every state is a unique case, we started wondering what the actual numerical advantage is when it comes to incumbency.
The answer is 7 or 10, or 9 or 16, depending on the situation. While this is a little less prophetic than the answer to the Ultimate Question of Life (42 for all the non-Hitchhiker’s Guide to the Galaxy fans out there), there is an interesting pattern that has developed over the past six years or so.
From a 10,000 foot view, there is little doubt about the power of incumbency. It gives a candidate excellent name recognition, franking privileges, the ability to perform constituent services, and access to donors, just to name a few of the benefits. The re-election rates alone are argument enough. Indeed, from 1962-2010, Senators have been re-elected nearly 89 percent of the time.1 (Just as a comparison, the numbers are slightly higher in the House of Representatives where generally around 90 percent of incumbents earn re-election.2)
Until recently, both parties tended to embrace these odds and very rarely did Senators have a meaningful primary challenge. However over the last few elections, the two parties, especially Republicans, have taken to running primary opponents against incumbent candidates. This dynamic has set up a very interesting view of the power of incumbency, since we can now look at the difference in results after a Senator loses in a primary versus when a Senator leaves the seat and a Special Election occurs.
MEASURING THE LOST ADVANTAGE
When examining the outcome of an election, there are two ways to look at how well a candidate has performed: the margin of victory and the share of the vote. By either measure, the loss of an incumbent reduces that party’s success in the next election. Since 2006, there have been nine Senators in situations that allow us to measure the importance of incumbency. Five of the Senators were incumbents who lost in their primary (Lieberman, Specter, Murkowski, Bennett, and Lugar), and four were incumbents who moved on to positions in the Obama administration (Obama, Biden, Clinton, and Salazar). In each of these nine scenarios, a “replacement candidate” ran on incumbent party’s ticket. We then measured how the replacement candidates performed as compared to their predecessors. Given the fact that this sample size is small and we do not control for mitigating circumstances (i.e. three-way races or the Christine O’Donnells of the world), we do not want to imply causation; however, this comparison does give us an interesting view into the statistical power of incumbency.
If we look at all nine Senators as one group and compare the percent of the vote earned by the incumbents in their previous elections to the vote share in the next election garnered by the replacement candidate, the disadvantage in losing incumbency status is clear. However, the two situations are so different we do not feel looking at them as a group is the best view. Breaking the nine races into two different groups—incumbency lost to appointment versus incumbency lost due to a primary—we see a slightly different picture between the two. The candidates who won deeply partisan primaries garnered a vote share that was on average 10 percentage points less than their predecessor. Conversely, the vote share of replacements for new members of the Obama Administration declined by only 7 percentage points. Clearly, an additional three points can make the difference when the goal is to get to 50 percent, but this is not in the category of monumental shifts.
However, in politics we do live in a world centered on how vulnerable or beatable a member of Congress is. Therefore, it is also important to look at the effect of incumbency not only on level of support but also on the margin of victory. The numbers paint a similar pattern; in races where the incumbent moved on to the administration, the performance is slightly better than those races where an incumbent lost in the primary. At the same time, the drop in vote share from an incumbent to a replacement candidate is still quite strong. In Senate races where the incumbent moved down Pennsylvania Avenue, replacement candidate’s average margin of victory dropped by over 9 percentage points. While this is a significant drop, it is better than the replacement candidates who were involved in primaries who saw their margin of victory decline by nearly 16 percentage points.
The 2012 election highlighted short-term and long-term challenges created by losing incumbent Senators with the failure of Richard Mourdock to retain Dick Lugar’s seat after winning the Republican primary. Mourdock would not have been the 50th seat for Republicans, but Joe Donnelly’s win for the Democrats makes the process of gaining control of the Senate a little more difficult for the GOP in both 2014 and 2016. Being able to hold solidly Republican states like Indiana is essential for Republicans to regain the chamber.
More immediately, President Obama has tapped another former Senate colleague to fill a cabinet position. By selecting John Kerry as his next Secretary of State, Obama has technically put a Massachusetts Senate seat in far more danger of going back to the GOP. Kerry would have easily won re-election, but his replacement will face a considerably tougher task in keeping the seat in Democratic hands. This is a tricky situation for a sitting President who undoubtedly would like to place the most qualified individuals in influential positions, but as shown above, there are very real political consequences to these choices.
As the parties look to the future, from a statistical perspective, the reduced performance of replacement candidates compared to incumbents puts the party at more risk of ultimately losing control of that seat. Joe Manchin and Susan Collins may infuriate Democrats and Republicans respectively for their occasional “heresies,” but members of both parties might want to think twice about what the true outcome of a primary challenge would be.
Additional Tags: power of incumbency, power of incumbency ap gov, power of incumbency in congress, power of incumbency wiki, power of incumbency meaning, power of incumbency define, power of incumbency be reduced, the power of incumbency essentially means that, power of congressional incumbency, how could the power of incumbency be reduced, what does power of incumbency mean