Some have suggested that Donald Trump's candidacy signals a future break in the GOP. The basic idea is that Trump's campaign has tapped into, and galvanized, a vanguard of disaffected voters largely within the GOP base that are now so disaffected with the GOP as a party that they will no longer reliably vote for Republican nominees.
There is some evidence of this crack at the state level (for example, see Roy Blunt's battle for reelection in Missouri this year), and the highest levels of the GOP are feeling the pressure. Furthermore, this pressure is probably going to increase between now and the election.
As Seth Masket notes, Speaker Paul Ryan and other GOP luminaries are being increasingly pressed to choose between supporting their party's presidential nominee and the US political system itself. After all, as Joshua Tucker recently noted, Trump's calls for jailing Hillary Clinton clearly smack of electoral behavior in "competitive authoritarian regimes such as Ukraine under former president Viktor Yanukovych, Belarus under Aliaksandr [Lukashenko] or the years of military rule in Myanmar."
But how would such a fissure emerge in practice? The US does not have much experience with credible third-party alternatives at the national level. We do have some experience with third-candidate spoilers at the presidential level (for example, Teddy Roosevelt in 1912, Strom Thurmond in 1948, and George Wallace in 1968 — not to mention Abraham Lincoln in 1860 and Andrew Jackson in 1824). But we don't really have any experience with a true "third party" at the national level. Why?
Fundamental to this is the Constitution itself, which leaves controls of elections to the states. This is understandable: The Constitution is basically a compact between the states (for example, state legislatures essentially control amendments to it). This implies that a third party, to be "national" in the Senate, must get on the ballot in various ways in various states.
As the Republicans have learned since broad discussions of replacing Trump began, "getting on the ballot" is difficult. These difficulties apply more broadly than the presidential race — the states control how they elect members of Congress, too. And a true third party in our system must not only compete for the presidency (just ask the Reform Party) — it has to compete for control of public policy, and that requires competing for Congress.
Getting on the congressional ballot is difficult in any one state. Because of differences among the states, and both geographic and demographic dispersion, it's a Herculean task to do so in more than a handful of them.
For example, just take a peek at the 50 sets of requirements for House races in 2014.
To drive home the scope of the challenge, remember these requirements are for each candidate in each congressional district. Simply put, a third party has to think about getting on the ballot in, and then contesting, hundreds of elections with varying requirements and challenges.
So based on a quick back-of-the-envelope calculation, it would cost at least at least half a million dollars just to get on the ballot in a majority of the House districts. That is before you even think about trying to be competitive in those races — presumably competitive against both incumbents and challengers with strong and efficient party machinery behind them. In the end, it would easily require hundreds of millions of dollars to field a competitive slate of candidates for Congress.
As a result of this structural feature, at the very least, I don't see a serious electoral split threatening the GOP in 2018 or anytime in the near future. That said, Trump's candidacy might still "split" the GOP into the future.
From an electoral perspective, this might be observed in two ways. First, continuing the Tea Party dynamic, GOP incumbents might face challenges in the 2018 primaries based on/energized by the Trump campaign. Second, the 2020 GOP presidential primary — regardless of who wins the 2016 presidential election — might mimic the 2016 GOP primary in terms of unpredictability and volatility.
From a governing perspective, well ... just ask former Speaker John Boehner: There's ferment in the GOP base. Of course, Trump is arguably better thought of as an effect, rather than cause, of this. But his candidacy has stoked this ferment — even if the GOP maintains control of Congress in November, it's going to be a rocky two years.
In the end, the GOP might "split" in many ways, but that doesn't mean a third party will emerge as a result. Rather, the reality is that a serious party must be able to field serious candidates in a serious swath of districts and states across the Union. For better or worse, the Constitution provides a structure that makes this an incredibly difficult task. In other words, if the US electoral system is rigged, it was rigged by the Founding Fathers.