Saturday, October 1, 2016

The Argument Against 3rd Party Voting

Central Premises

Here are what I consider to be the central premises of the case against 3rd party voting:

1) There exists 3rd parties where the following conditions hold true:

  • 1a) At least some of their voters would have otherwise voted for a mainstream party…
  • 1b) And voted disproportionately towards the Democratic Party (i.e. more than 50% liberal as opposed to conservative)…
  • 1c) Such that the presence of the 3rd party on the ballot could alter the outcome of the election in favor of the ideologically opposite mainstream party.

2) That the 2016 U.S. presidential election contains examples of (1), namely:

  • 2a) The Green Party represented by candidate Jill Stein
  • 2b) The Libertarian Party represented by candidate Gary Johnson
  • 2c) The Party for Socialism and Liberation represented by candidate Gloria La Riva and similar parties such as the Workers World Party, Socialist Workers Party, etc.

3) That the negative potential outcome multiplied by the probability of (1c) outweighs the sum of the following:

  • 3a) The positive potential outcome multiplied by the probability of the 3rd party winning.
  • 3b) The positive potential outcome multiplied by the probability of the 3rd party threat of (1c) shifting the Democratic Party’s platform.
  • 3c) The positive potential outcome multiplied by the probability that the 3rd party’s campaign will productively highlight overlooked issues on the national stage.

Why lay these out in such overly-formal terms? I think it is worthwhile to see where the arguments are strong and where they are, in fairness, vulnerable. I welcome amendments and good-faith interrogations of these premises.

Strengths and Vulnerabilities

(1a), for instance, is pretty irrefutable. (1b) is sometimes called into question, but generally holds strong. (1c) is arguable but, as I’ll delve into with a later post, it not only can, but indeed has, happened in at least one well-documented presidential election.

It is not common to see arguments against the leftism of the Green Party (2a), but it is actually less pronounced than some might think. (2b), the premise that the Libertarian Party disproportionately siphons votes from the left, is highly contentious, although not without evidence.

There have also been parties like the Reform Party whose voter base spanned the left-right spectrum and the Constitution Party, which traditionally draws from the far right. I am mentioning these to clarify that not EVERY 3rd party draws disproportionately from the left; I am only personally concerned with the ones that do.

(3a) will be the focus of a lot of the mathematical models I will be running. Not stated here, but perhaps strengthening (3a) is the fact that while I might prefer the platform of a 3rd party, that platform may not equate to a highly positive potential outcome. There are cases where I question whether the candidates representing my preferred platforms are well-qualified to handle the job, in particular in the areas of foreign policy, statecraft and consensus-building.

The weakest areas, in my opinion, are the highly qualitative and hard-to-measure (3b) and (3c). In fact, if you are a single-issue voter whose issue falls under (3c), 3rd party voting makes sense for you. Or if you dislike Hillary Clinton and/or herplatform to the same degree as Donald Trump and his platform, than all of (3) collapses.

There are arguments against 1st party voting that are not taken [yet] into account by the premises above, including the possibility of inadvertently selecting a 1st party candidate who yields a negative outcome in practice, of rewarding 1st party corruption, and lending support for contentment with the status quo.


  1. Is (1b) a given, or is it just a given for the circumstances you are interested in? It is my perception that there are more (or more significant) left-leaning 3rd parties than right at present, but I think before 20 years ago or so that was not especially the case. For example, Perot seemed to draw from both sides, but more from the right, and certainly Teddy Roosevelt split the Republican vote.

    1. My latest post, simulating elections like the 1912 and 1992 ones you refer to, looks at both conservative-leaning and liberal-leaning 3rd parties. They definitely both exist, and their prevalences change year-to-year. In fact, some parties like the Libertarians have leaned both ways at different times. But I am mostly only interested in the liberal-leaning third parties, so I admit to focusing on those in most of my posts.