Voting Out Incumbents Could Worsen Partisanship

Voting Out Incumbents Could Worsen Partisanship

By David Byler - October 16, 2014

Americans overwhelmingly disapprove of Congress. The two major political parties are more partisan and polarized than ever and neither is much more popular than the other. But would ousting sitting politicians in the upcoming election actually bring about the sort of change Americans seem to want? Specifically, would Congress become less polarized if Americans voted out their representatives in favor of general election challengers?

If Americans got what they wanted and denied re-election to large numbers of current members of Congress, it would likely have no effect on partisanship and polarization in Washington or it could even widen the gap between the parties.

This is because challengers currently tend to be even more partisan than the current members of Congress, according to data from a technology startup called Crowdpac.

We can illustrate this by examining two scenarios. For the first one, we consider a scenario where only vulnerable Republicans and Democrats lose; for the second, we consider a scenario where a large number of Republicans and Democrats are unseated in favor of their opponents.

To help figure out what the effects of these scenarios on polarization would be, we examined the Crowdpac data. According to company co-founders Steve Hilton and Adam Bonica, their mission is “to level the playing field with the big donors and special interests by putting non-partisan, objective information in users’ hands.” To this end, Hilton and Bonica have created a mathematical system that uses data on donations to and from candidates and campaigns to assign an ideological score to politicians. This data, while not complete, covers well over 80 percent of recent major party general election candidates. Crowdpac provided RCP with all available scores for congressional candidates from 2004 to present.

Scores typically range from -10 to 10, where more liberal politicians get a higher negative score and conservative politicians get a more positive one. For example, according to data from 2012, former Rep. Allen West of Florida, a conservative, had a score of 7.14 and Rep. Michael Fitzpatrick, a moderate Republican from Pennsylvania, had a score of 3.64. House Minority Leader Nancy Pelosi had a score of -6.2 and moderate Democratic Rep. John Barrow of Georgia had a score of -2.38. According to Bonica, slight changes in a score -- as little as one point -- can result in observable differences in how candidates vote, or would vote were they to make it to the House.

We then placed these scores alongside RCP polling averages to gauge what would happen if the first scenario -- vulnerable members of Congress losing to their opponents -- came to pass. We calculated the average ideological score of current House Republicans and Democrats and compared that to the average score for both parties if the most vulnerable members were replaced by their challengers in the upcoming election. If 44 competitive House races (the current number featuring an incumbent and that are not safe for either side) all flipped against sitting representatives, the average ideology of each party would move less than half a point. On a -10 to 10 scale, that shift is practically unnoticeable.

In other words, unseating all vulnerable incumbents would do little to decrease the ideological gap between the parties.

Now suppose the situation were different: that large numbers of sitting members were unseated in favor of their challengers, regardless of party or level of vulnerability. This would likely increase polarization.


In every election from 2004 to present, House candidates who challenge sitting representatives have been, on average, more extreme than the House members from their own party.

This initially seems counterintuitive. Conventional wisdom holds that a number of factors -- human geography (unintentional gerrymandering), the creation of safe, gerrymandered districts built to protect ideologically extreme congressmen, and a host of other factors -- have led to a polarized Congress representing only the fringes of both parties. But the data suggest that challengers are even more extreme than the current sitting congressmen. How could that be the case?

One plausible answer is that safe, well-gerrymandered congressmen tend to draw out ideologically extreme opponents. Representatives who are safe from challenges may ward off all but the most ardent “true believers” from the other side, thus making challengers on average more extreme than one would expect. This theory would need to be tested further, but it makes sense of the data up to this point.

Either way, Crowdpac data point towards a dismal conclusion. Though Americans have long believed that their one option to try to change Washington is to throw the bums out, doing so seems only to open the door for an equally unsatisfying officeholder. If a large number of sitting congressmen lost, more ideologically extreme challengers would take their place. The gap between Republicans and Democrats in the House would almost assuredly widen, and Congress would likely become more polarized.

David Byler is an elections analyst for RealClearPolitics. He can be reached at Follow him on Twitter @davidbylerRCP.

A Reply to Rush Limbaugh
Mona Charen · December 9, 2014
Against the Wave: Three Winning House Democrats
David Byler · December 14, 2014
Obama, Congressional Dems Show Cracks in Unity
Jim Kuhnhenn · December 7, 2014
Democratic Louisiana Senator Mary Landrieu Defeated
Melinda Deslatte and Bill Barrow · December 6, 2014

David Byler

Author Archive

Follow Real Clear Politics

Latest On Twitter