The Spoiler Effect

In Virginia's 2013 gubernatorial election, Libertarian candidate Robert Sarvis captured over 146,000 votes--almost three times the 56,000-vote margin that separated Republican Attorney General Ken Cuccinelli from the victor, former DNC chair Terry McAuliffe. His presence on the ballot may well have cost Republicans the governor's mansion. Last week, Cuccinelli consultant Chris LaCivitas had this to say about Libertarian candidates:

"Libertarians and quote-un-quote libertarian-minded Republicans exist for one purpose and one purpose only, and that is to keep Republicans from winning general elections." -- Chris LaCivita

However, LaCivita was speaking not about Cuccinelli's loss in Virginia last year, but about Libertarian candidate Sean Haugh, who's been nominated to run against Sen. Kay Hagan and Speaker Thom Tillis in the Senate election in North Carolina. Regardless of the election, though, LaCivitas is referring to a problem that crops up frequently under first-past-the-post, winner-takes-all electoral systems. It's known as the spoiler effect, and in short, it refers to the phenomenon in two-party systems where a vote for a third-party candidate is effectively a vote for the major party that is less aligned with the third party's platform. While it's harder to gauge the effects of candidates like, say, Ross Perot (who likely took votes from Bush and Clinton equally in battleground states in 1992), Libertarian candidates seem to draw largely on the Republican voting base. It makes sense in historical context, in which Republicans like Rep. Ron Paul and Democratic fundraising boogeyman Charles Koch have run on Libertarian tickets for president (Rep. Paul for president in 1988, and Koch for vice president in 1980).

It's an effect that briefly came up for discussion in some analyses of House Majority Leader Eric Cantor's upset loss in the VA-7 GOP primary: what would happen if Rep. Cantor, like Sen. Murkowski in Alaska's 2010 Senate race, mounted a write-in campaign to beat both Profs. Brat and Trammell in the general election? Considering that Rep. Cantor won with over 222,000 votes in 2012 while only 36,000 voted for Prof. Brat in the June 10 primary four days ago, that was a possibility, but the real worry for Republicans in that scenario would have been that Rep. Cantor would suck away just enough Republican votes to give Democratic Prof. Jack Trammell a plurality of the vote in a state that doesn't mandate runoff elections when no candidate has received a majority--Rep. Cantor would spoil the candidacy of the official Republican nominee. (Of course, all of this is moot at this point because Rep. Cantor quickly announced that he would not be running in November.)

The effect is a product of our winner-takes-all electoral system, where voters for losing candidates receive exactly as much representation if their candidate loses by 2 or 3 points as they would if their candidate lost by 20 or 30 points (but that's a story for another day). When it comes down to a close race (North Carolina) that could very well decide which party controls Congress (it's got a fairly strong chance at being the sixth pickup Republicans need to win the Senate), potential spoilers definitely need to be watched in the polls, which is why pollsters in North Carolina (looking at you, PPP) ought to start including Mr. Haugh as a candidate in their survey matchups. (More next post, however, on how accurate those polls will actually be with Haugh included.)