A Libertarian Spoiler in North Carolina?

One surprising state where the GOP is still struggling is North Carolina, where recent polling has Republican Thom Tillis trailing Democratic incumbent Kay Hagan by 3 or 4 points.

Interestingly, Hagan’s numbers are not substantially better than candidates in states where Democrats are trailing.RealClearPolitics reports her standing as 44.5 percent, compared to 43 percent for Bruce Braley in Iowa and Mark Udall in Colorado.

Hagan’s lead in North Carolina depends in large part on the presence of libertarian Sean Haugh, whom Politics1 describes as an, “Ex-Libertarian party national political director, pizza deliveryman, & frequent candidate.” Currently, Haugh pulls in 5 or 6 percent of the vote.

If Haugh continues to do this well, that will make it substantially tougher for Tillis to close the gap. The question is: will he?

Answer: it is hard to say. History gives contradictory indications.

For starters, 6 percent is a hard figure for a third party candidate to retain as we get closer to Election Day. It is greater than the percentage of true third partiers in the electorate, those who vote that way knowing their candidate is a sure loser. In North Carolina, third party voters have averaged about 1 percent in presidential elections over the last decade. A figure as high as 6 percent must therefore include people who do not usually waste their votes. But when a candidate only polls 6 percent, everybody paying even a modicum of attention knows that their vote will be wasted, so there is a gravitational pull back down towards the core third party vote.

This is a big reason why some notable third party candidates have disappointed in recent years. Ralph Nader’s 2000 insurgency, although at 2.4 percent nationwide was big enough to spoil a Democratic victory, was still only about half of what final polls had predicted. By the same token, third party candidates underperformed more recently in the 2013 Virginia gubernatorial election and the 2009 New Jersey gubernatorial election.

In each of these cases, a good chunk — maybe 40 to 60 percent — of poll respondents claiming to support a third party candidate ultimately thought better of it. Presumably, they did not want to waste their vote, and instead backed the candidate who most closely reflected their views. This is the “rational” move to make in a winner-take all race, which usually become two-way affairs. Indeed, we are seeing a weird version of this tendency in Kansas, only there it is the Democratic candidate who was squeezed out. In parliamentary systems where parties that cross certain thresholds are rewarded with legislative seats, third parties are sustainable. In the United States, where you get nothing for losing, most would-be third partiers tend to fold into the main two-party battle.

All of this would suggest that Haugh right now is over-performing. Compounding this is that his campaign does not seem to consist of very much at all. Here is his rather unimpressive web page. And his filings to the Federal Elections Commission report no raised money to date. The 6 percent or so of North Carolinians who claim to back Haugh are thus not doing so because of his active efforts. So, it is reasonable to expect them to bail at some point in the next couple weeks, with Tillis being the likely net beneficiary.

But this is far from a guarantee. In 1992 Ross Perot won upwards of 20 percent — enough support to make a credible claim prior to the Election that he could win. Four years later, he won only 8 percent — such a small amount that his supporters had to have known he was going to lose, and thus they were “wasting” their votes. Ditto John Anderson, the former Republican turned Independent who snagged 7 percent in 1980. Going back further in time, the presidential votes for George Wallace in 1968 as well as Populist James Weaver in 1892 suggest that some candidates who are bound to lose can nevertheless do better than the fractional, hard core third party vote. (Steve Rosenthal of the University of Minnesota wrote a great book on this about twenty years ago.)

More recent examples are of particular relevance to the battle in North Carolina. In 2012 Democrats won four Senate seats in states that Mitt Romney carried — Indiana, Missouri, Montana, and North Dakota. In all four of those races, Republicans ran notably inept campaigns. Two campaigns, by Rick Berg in North Dakota and Denny Rehberg in Montana, were conventionally bad: lousy campaign strategy, failure to nationalize the race, an inability to gauge the strengths of their opponents. Technically speaking, you could argue that Rehberg’s candidacy was a bit worse than conventionally bad; after all, his company had previously sued the city Billings, the largest city in the state. Still, he cannot hold a candle to the other two, infamously bad campaigns; Todd Akin of Missouri and Richard Mourdock of Indiana foolishly made themselves targets of the “War on Women” counterattack by Democrats.

In three of these races there was an Libertarian on the ballot — Indiana, Missouri, and Montana. Across these states, the Libertarian averaged about 6 points on Election Day (a narrow range of 5.7 percent in Indiana to 6.6 percent in Montana). So, the most obvious theory is that some proportion of the electorate that usually does not waste their vote on a third party candidate simply could not bear to vote Democratic or Republican for Senate. Presumably, the lion’s share of these voters were Mitt Romney backers who, if the GOP had run a better campaign, would have supported the Republican. What is especially troublesome for Republicans is that, at least in Montana, the Libertarian might very well have played spoiler. Akin and Mourdock lost so badly that the Libertarian candidate did not make a difference, but the Libertarian vote in Montana — at 6.6 percent — was greater than the margin that separated Rehberg from Jon Tester. Even when we factor in that Montana has a stronger libertarian streak than other states, the Libertarian may still have played the spoiler.

The good news/bad news for Republicans in North Carolina is this. Good news: Tillis is not a terrible candidate by any conventional standard. He is not a superstar like Joni Ernst, but he is not nearly as awful as Mourdock or Akin, and he has run a better campaign than Rehberg or Berg (not suing your state’s largest city is a good first step!). Bad news: it may not matter. The Democratic party has really singled out Tillis this cycle; perhaps anticipating that North Carolina would be its main firewall in holding the majority, Democrats have spent a princely sum to make Tillis out to be the devil incarnate. Compounding that, he had a somewhat difficult primary season, and bruised feelings do not always heal in time. Compounding that, he is the face of the state legislature, which is not popular in the Tar Heel State.

Probably the single best metric for gauging Senate elections this cycle is Obama’s job approval number. In North Carolina, it is a sure bet that it is terrible — in the low 40s, just like the country at large. That suggests a ceiling for Hagan that should make the race quite winnable for the GOP. But if Republicans cannot dislodge maybe half of the Haugh supporters, the Democrats might yet hold the state.

Unfortunately, the evidence is mixed as to whether the GOP can succeed.