So You Want a Third Party

PZ has just saved me work, summing up most of my thoughts on third-party presidential candidates.

I’m not saying that we’re doomed, though, just that the presidential race is the wrong place to effect change.

The right place is everywhere else. Maybe the primary campaigns would be better: we need to get candidates in place that don’t require us to hold our noses in order to vote for them. The Republican field is always a race to find the one candidate just crazy enough to satisfy a badly deranged base, while not so obviously crazy as to alienate everyone else, so forget them. The Democrats always seem to be looking for the moderate who won’t really change the system (that would be scary) and who will inspire just enough to squeak into office…but not inspire so much that people will wake up to our problems. I suspect that both parties will fundamentally resist change.

So maybe that’s not even the best place to work on fixing the election system. Especially in this election, the power of incumbency is so great that no one was even going to look seriously at an alternative to Obama.

You know where the elections really matter, where you really have a choice? At the local level. The Green Party is stupid to throw so much effort into a presidential campaign right now — they ought to be focused on building a base. I would vote for a Green for city council or district representative in a heartbeat. And once they’ve built a deep party structure, they become serious candidates for higher office, because they will have the backing of people doing good work on the ground.

There’s more. Read it.

Since PZ did so much, let me add a few specifics on leveling the playing field at a local and state level.

  • Volunteer, donate, and run: This should be a no-brainer. You want a party for you? Help build it.
  • Funding: Does your state do any public funding of elections? If it doesn’t, work on fixing that. Support it at the federal level too. Entrenched parties attract entrenched interests with the money to fund what they consider to be a sure thing. They don’t take much in the way of risks. Public funding doesn’t care who’s going to win as long as the party or candidate meets some minimum level of support. If your state does have public funding, know what is required for a candidate to qualify.
  • Debates: Many sponsors of local debates are open (or at least more open) to including more than two parties’ candidates. If they aren’t, put some pressure on them. These organizations exist to inform you as a voter. Stand up for your interests. Then attend, watch, or listen to the debates.
  • Press: Our corporate press frequently doesn’t feel much need to cover third-party candidates. Bug them about that. They, like the organzations that sponsor debates, are supposed to be representing your interests. If they don’t, take matters into your own hands. Find your local political blogs and pass around what they have to say about the third parties. Write up what you hear from the debates. Figure out what issues interest you and consolidate candidates’ statements from their websites in blog posts or on other social networking sites. If they don’t have information, write to the candidates with your questions and share their responses with your friends. Go around the gatekeepers.
  • Major-party status: In order to be included in debates or public funding, parties typically have to meet some threshhold for votes in statewide elections, registered members, or some other arbitrary qualification meant to weed out vanity candidates. Find out what those qualifications are locally and keep them in mind as you go about your political life. Wide-open races, like Minnesota’s current Senate race, may throw you opportunities for making a worthwhile protest vote.
  • Have standards: The reason you want a third party is that you want viable candidates who can exert an influence on the public discourse, if not govern outright. Don’t vote for bad candidates just because they’re neither Democrats nor Republicans. If they win, you’re screwed. If they don’t, no one takes you seriously. The Grassroots party held major-party status for two years in Minnesota when, during one of those wide-open elections, they ran a candidate who had a coherent social libertarian, socialist platform instead of just “legalize eet”. It would have been a sea change if he’d been elected, but it wouldn’t have been a disaster. Encourage that.
  • Update: Stacy reminded me in the comments that I forgot to mention our system for determining a winner in an election. Do you have local instant runoff elections? No? See about fixing that. Many locales have standard runoff elections, but these are hugely expensive on any large scale. Instant runoff elections allow you to vote for your first choice without dooming your second, but they’re assumed to be too confusing for many voters. We need instant runoff to have a history of success before we can advocate for it at the national level. 

Third parties aren’t impossible. They’ve happened more than once in U.S. history. Neither party has been particularly well regarded by the public for decades. This can be done if we want it. We just have to build for the long haul, and that takes smart work.

So You Want a Third Party

21 thoughts on “So You Want a Third Party

  1. 1

    And so we end up in a Catch-22: Do not vote for or otherwise support third parties, which lets both wings of the Two Party move even farther to the right, which makes it more important not to vote for or otherwise support third parties.

    Until we hold the Democrats’ collective feet to the fire, nothing will change. And we cannot hold their feet to the fire if we are too spineless to vote for someone who actually supports progressive values. We must not sacrifice long-term change to assuage our short-term fear.

  2. 3

    And Gregory comes in and demonstrates the problem: “We cannot hold their feet to the fire if we are too spineless to vote for someone…”

    NO!!! Didn’t you read the post?!?!? Voting is only a small part of the process. Much of the rest is there in Stephanie’s first bullet: “Volunteer, donate, and run”.

    I think the Tea Party makes a fairly good Exhibit A. Those people got involved (actually, had been involved, just not as vocally) two years ago. And look at how many people they got elected. Look at how many politicians they got to move to the right to meet their demands. They didn’t get to where they are by simply casting their ballot every election.

  3. 4

    We still need to change the voting system. You don’t just want the occasional successful third party candidate. At best, with the system we’ve got, a third party might replace one of the Big Two as one of the major parties, but the problem–the inevitable marginalization of other voices–will remain. People who don’t like one of the Two will still find themselves in a bind, unable to vote for their preferred candidate because that would amount to a vote for the candidate they like least. With a First Past the Post system, we’re doomed to have two majors battling it out.

  4. 9

    Instant runoff voting is a terrible system. It’s still susceptible to Duverger’s law just like plurality voting, which ought to be enough to stop advocating it. But besides that it’s got all sorts of ridiculous, illogical features.

    – People can change the winner to their favorite candidate by ranking her last (in large enough numbers)
    – People can change the winner to their favorite candidate by not showing up (in large enough numbers)
    – My favorite: the winner can be invariant under reversal of all the ballots’ rank orderings (!!!)

    … Besides more logistical disadvantages like the lack of precinct subtotals and the incompatibility with voting machines.

    I sincerely hope you don’t get it installed anywhere. I see Stacy already linked to the better alternative.

  5. 10

    Fundies realized a long time ago that the Wingnut Party(they would have probably named their third party something different) wasn’t going to work. They took over the Republican Party. Why can’t Progressives take over the Democratic Party? At one time the Republican Party was much more Progressive than the Dems are now. It could happen!

  6. 11

    On a day-to-day basis, local government decisions has a greater effect than state or federal governments, and requires comparatively little resources due to the number of people per electoral district being relatively small, so this is where attention should be directed. If it became more profitable for builders to build upward rather than outward, through elimination of height and minimum restrictions, integration of residential and commercial use of land, maximum automobile parking of 0, decreasing impact fees to 0 for infill and increased for low density, and basing property taxes on the value of land alone rather than land and building, there would be improved economic, environmental, health and social qualities, resulting in improved voting decisions at all levels of government. With improved governance and lower cost of housing and transportation, more people will move to the area, resulting in improved influence of these areas in state and federal governments. With a unified voice coming from urban areas in the majority of states (the suburbs and exurbs would no longer exist), there will be sufficient demand to enfranchise all American residents, including those under the age of 18, non-citizens, felons and ex-felons, and the District of Columbia and the five territories including Puerto Rico. Eventually, there will be sufficient support to eliminate the bureaucracies known as the state and federal senates, resulting in appropriate representation of the more populous regions/states, and the implementation of appropriate changes will become significantly easier, including national popular vote for president rather than electoral college and proportional representation/instant runoff voting.

  7. 12

    There are problems with all voting systems. The math and science behind the systems and their results is fascinating, and much more complicated and problematic than people realize.

    Indeed. Once you get to Arrow’s Impossibility Theorem you have to start wondering if a dictatorship wouldn’t be much, much simpler.

  8. 13

    Joshua Adams, are those pathologies *typical* of IRV? What would you prefer?

    If one wants theoretical elegance, one could consider Condorcet methods, which turn preferences into a virtual round-robin competition. If one candidate beats all the others in these virtual 1-on-1 contests, then that one has unambiguously won. But if no candidate does so, then one must use some algorithm for disambiguating such cases. Various algorithms have been devised for doing so, like Schulze beatpath and ranked pairs, but they tend to be rather hairy.

    IRV and Condorcet methods are improvements on FPTP, but they still don’t do much for third-party friendliness. For that, one needs multiseat electoral districts, and the most reasonable choices there are Single Transferable Vote and party-list proportional representation.

    STV is a sort of multiseat version of IRV, with winners as well as losers removed in each round.

    Party-list PR is very common, and a variant that preserves local representation is mixed member: a mixture of district seats and at-large or list seats.

    As to Kenneth Arrow’s Impossibility Theorem, it’s about always satisfying certain “nice” criteria at the same time. So one has to pick what one is willing to sacrifice.

  9. 14

    lpetrich, I advocate range voting. It’s a beautiful, simple system that does not suffer any pathologies (Those impossibility theorems only apply to rank order voting systems, which range isn’t). It also lacks the features that lead to two-party domination, because candidates are scored independently from one another, meaning there’s no reason to throw your favorite third party under the bus in favor of the safe candidate. A strategic voter on the left could score Stein at the maximum, and then also give Obama the same score while giving Romney 0. That way you can help Stein get a strong showing while not risking helping Romney over Obama. The only sense in which Obama is hurt by your vote is if Stein ends up defeating him, in which case, mission accomplished! She probably wouldn’t, but at least her score would look a heck of a lot more impressive than simply getting a fraction of a percent of the vote share does. This should increase the perceived viability of the parties in the next cycle, which should ultimately help get their foot in the door. The beauty of the system is that you can essentially throw some score their way at no strategic cost, so it stands to reason that plenty of people would, at least once they realized this. And once they look definitely competitive, people may start decreasing the scores they give to the “safe” major candidates and actually letting the third parties win.

    It’s speculative analysis, of course, but it seems like the best shot of any voting system to help third parties*. And it has nice properties like monotonicity, and electing the Condorcet winner, and stuff. I wouldn’t settle for anything else, really, if you’re going to the trouble of agitating for change. Except maybe approval voting, which is similarly elegant. However, some studies were done via exit poll to see how third parties would have fared in a range or approval vote, and apparently range helped them more than approval. So it makes sense to hold out for the best.

    As for whether the pathologies are typical in IRV, I’m not sure that question makes sense. No matter what the election results are, you could do a thought experiment where you change the result by some completely counterintuitive means. I’ll cite an example from the CRV website.

    Concerning Ireland’s 1990 presidential election:

    “If a whole new county had magically appeared, all of them supporting Lenihan’s most-hated rival Currie and ranking Lenihan in last place… then Lenihan would have won! If a slightly-smaller Currie-county had appeared and 12% of Lenihan’s voters had instead voted for his most-hated rival Currie while ranking Lenihan last… then Lenihan would have won! So, since those handicaps actually were not present – why didn’t Lenihan win?”

    I think these kinds of questions demonstrate the logical incoherence of IRV. And certainly the “Best=Worst paradox” does so. Put simply, the way it chooses winners doesn’t make any sense. Asking whether a specific election was essentially pathological is to miss the point that the entire system is screwed. (However, see here for a discussion of how often pathologies occur: And, again, it just plain doesn’t help third parties, so why on earth is it being advocated in a post about how to help third parties?

    * Simply changing the voting system, of course, isn’t enough. Not all of the blame lies with strategic plurality voters for why third parties don’t win. You would need to do other things like get rid of gerrymandering, lower the ballot access requirements that the major parties throw up specifically to keep their monopoly, create real bottom-up infrastructure by winning local and state seats, etc. I’ll also plug CRV’s page about replacing gerrymandering with a simple redistricting algorithm: Look at those nice, clean maps.

  10. AJS

    Any voting system has problems, but the Rest Of The World likes the Single Transferrable Vote. You give everybody a preference from 1 to however many (you can leave some candidates with no preference). All the “1”s are counted up.

    The weakest candidate is eliminated. Their votes are redistributed according to the “2”s on the paper (if there is no “2” on the paper, the ballot goes on the “spoiled” pile.)

    Eliminations and redistributions continue until there is only one candidate (plus the “spoilt” pile) remaining. If there are more valid votes than spoiled ones, that candidate is elected; otherwise, the election is void and nominations are re-opened.

    Note that the counting may be stopped early (and usually is in practice): Any candidate is mathematically certain to win if they have polled more than 50% of the vote (e.g., if there were 400 votes in total, the 201st ballot paper landing on a candidate’s pile would decide the winner).

    Note also that it is possible to fill multiple seats in one STV election, by transferring votes from the winner down to the next preference.

    STV also has the advantage that it is highly amenable to hand counting (which is highly parallelisable, scalable and generally faster than machine counting). With the candidates and their representatives doing the counting, nobody trusts anyone else at the same table; so they only result on which they can all agree is the truth.

  11. 16

    I’m for Lottodemocracy.

    Just picking people at random would ensure a perfectly representative government.

    It would save untold millions in election costs.

    Big money would lose because it’s easier to catch bribery than cough campaign contributions cough

    The executive could spend its time actually doing the work instead of campaigning all the time.

    Picking people at random couldn’t possibly be worse than the dreck we get now.

  12. 17

    @Stephanie – I read. Your thesis was to leave national politics to the Two Party and focus exclusively on local races. I disagree with that thesis.

    Local and district party organizations are already pretty responsive to the values of the people: they have to be, if only because they see their rank-and-file constituents every day. State parties are less responsive, but the know that it would not take much effort to overthrow them: witness how quickly Talibangelicals, then the Tea Party, took over the GOP.

    At the national level, however, the politicians and policy makers are very isolated. There is no threat to their mode of operation, as they can reliably depend on the “A vote for X is a vote for our archenemy!” crowd to keep the sheeple in line. They can reliably depend on our vote no matter how objectionable their views might be.

    That is not going to change on the national level until we are willing to act on the national level.

  13. 18

    I am in full support of members within this community choosing to run for office, but I have to ask: why run as a third-party candidate when the Democratic Party is very welcoming to rational candidates of all types?

    If you live in a strong Democratic area, I think someone from the FTB community would be able to have a fair chance of success in a primary, and then would have a very good chance in the general election.

    If you live in a Republican area, then you probably wouldn’t even face a primary within the Democratic Party, and while you probably wouldn’t win in the general election, I don’t think your chances would improve if you ran as a third-party candidate.

    I personally think it is a more viable approach to help move the Democratic Party to better policies from within, rather than trying to create a viable third-party. And I don’t think the Democratic Party at the local level is resistant to intelligent people becoming more involved and helping move the party in the correct direction.

    But regardless of the party affiliation, I do think that the solution for a better future does require us becoming more involved by volunteering, donating, and running for office.

  14. 19

    The inherent structural problem is this: in any representative democracy, to obtain election success commensurate with with electoral support, it is necessary to win the support of greater than one third of the active electorate. Do the sums, please …

  15. 20

    I love the hate IRV gets. It’s enormously amusing., quoted/linked by Joshua Adams, has an IRV hate page called “A completely idiotic Instant Runoff Voting (IRV) election” ( And they’re right! It is idiotic: of 27 possible IRV voting patterns, all the voters picked the weirdest combination of three patterns available. Not to mention the desperate construction to make the voting look pathological. Then the page finishes off with “YOU WANT COMPLEXITY? That was only a 3-candidate election with only 3 kinds of voters.” Yes it was; where were the other 24 voting patterns? Why are we studying an election with log10(voters) < 4? It's an interesting bit of tortured math, but it's not a likely outcome in an actual election.

    Then in, the author talks about IRV spoiling. Specifically, "Favorite Betrayal" and "Participation Failure". Funnily enough, the small text under favorite betrayal states that in a real system (Australian HoR elections), this never happens. I'd imagine that if you stop trying to consider a 3-way election as a propsition between three voting patterns and instead consider all 27 patterns, participation failure is minimised as well. is more interesting. But all that states is that both holding a final 2-candidate run-off election and using FPTP seem to produce wrong results on a regular basis. One example in over 20 years of "funny" results for IRV, and that's only if you propose weird alternatives and pretend that the people who vote for one candidate as first preference all vote the same way. In practice, this doesn’t happen.

    Range/approval voting, IRV/STV, MMP, some variety of condorcet? All fine voting systems in real life. All of them can produce funny results, though, if you torture the math enough. Very amusing to see people hate on IRV so hard though, just because bad results are easy to make in toy examples where you ignore 89% of possible votes.

    In my opinion, approval voting is a very simple system that produces fair results – it’s essentially range voting with a maximum score of 1. I think better of humanity than to think many would be confused by a more complicated system, though, and I prefer New Zealand’s Mixed Member Proportional system most of all. It has the unfortunate side effect of codifying parties into the system, but since they’re here to stay anyway, we may as well make use of the bastards, no? kevinalexander’s system is also interesting, but I have to wonder if it wouldn’t become basically hated in the same way that jury duty seems to be. Especially with a large and powerful anti-corruption organisation meaning people couldn’t ‘win’ the lottery and treat it as if they’d won a lottery for money…

  16. 21

    […] talking about it nor not talking about it seems to do it justice. I had some advice for people who want a strong and enduring third party. Along with the rest of the country, the fallout after the Newtown shootings prompted me to talk […]

Comments are closed.