Monday, June 20, 2011

Get This Party Started: Future History

History of American Political Parties: I - II - III - IV - V - Future

About a year and a half ago, I wrote a five-part series looking at the historical transitions of the American party-system. At the end, I promised a look at the future, of what the next transition might look like, and an assessment of whether or not it could happen soon. But when I started looking in to it, I discovered that it was hard to find information about voter's views that wasn't broken down first by party affiliation, and mothballed that idea. But then, just a few weeks ago, the new Pew Research Political Topology Report was released, which has precisely the information I was hoping to see.


First, a quick (and what I hope will be seen as unbiased) summary of the 5 previously-examined transitions.

  • 1st to 2nd: One major party disappears in treasonous-embarrassment, it's members joining up with what is now the only game in town; one-party rule lasts 20 years before falling apart into four-way presidential race, partially along the old party lines; the winner gets just 30.9% of the popular vote.
  • 2nd to 3rd: Both major parties are unable to adequately address the nations largest economic and social issue: slavery; major presidential nomination fights occur in both; one party's leadership collapses completely, the other breaks into factions; a new party is created with an explicit platform on the issue, and they win the presidency on their second try, with just 39.8% of the vote.
  • 3rd to 4th: Economic arguments (over the gold standard and inflation) find split support in both parties; a splinter 3rd-party spoils the election for its parent, taking four (of 44) states, and in the next election, there is mass cross-party voting based on this issue; the crossover becomes permanent.
  • 4th to 5th: A massive economic downturn permanently pushes many voters (the working poor and academics) away from the party they blame for the crisis; despite massive 3rd-party voting, a majority (57.4%) vote for the opposing major party's candidate.
  • 5th to 6th: Massive social change and accompanying laws (culminating in The Civil Rights Act) permanently push many voters away from the party that enacted them, while attracting others; in the course of just two presidential elections, all but 5 (of 48) states reverse their partisan presidential leanings.

What's Your Issue?

After that first transition, there's a strong re-occurring theme in there, namely that of a major issue which cuts across the constituencies of both existing major parties. Only once did it completely collapse a major party, while the other three times it caused a major swap of supporters. And this is where the Pew Report comes in, because it examines the American electorate, and tries to break them down into a small number of issue-groups, from which we should be able to identify a nascent party transition.

The good news, if you're a third-party advocate, is that Staunch Conservatives and Solid Liberals only make up 11% and 16% of the electorate, respectively; this will jive with your belief that there aren't really that many people who are actually that devoted to the two major parties as they currently exist.

The bad news is that, to my eye at least, it doesn't seem like there's any serious cross-party split out of which you could expect to see the sort of transitions we've come to expect over the last 100 years. At least, not in the short term. For instance, Pew identifies Libertarians (10%) as one of its groups, but when you dig into the issues, they tend to hold identical views as Staunch Conservatives except for the issue of gay rights; similarly, we see New Coalition Democrats (9%), who are Solid Liberals except for the issue of gay rights. While that could be the makings of the kind of grand constituency-swaps we've seen since the creation of the Republican party, it hasn't yet risen to the level where this one issue can drive the outcome of every election.

The same story can be told about welfare, and other social safety-net programs. The so-called Disaffecteds (11%) are Staunch Conservatives, except they support (and often are supported by) these programs, while Post-Moderns (14%) are Solid Liberals, except that they would scale back these programs (which they are generally well-off enough to not need to use them, and question if anyone else needs them either.) This could also grow to become a defining issue that leads to a long term exchange of party support, but it's just not quite there yet. In both these cases, the issue at-hand would have to overcome voters' concerns on every other issue, and I have a hard time imagining, for example, that an overwhelming number of Libertarians would vote for the full Democratic platform because of its stance on gay marriage, or a Democratic candidate soften the party's stance on every other issue in an effort to attract them.

The other two groups which Pew identified were Main Street Republicans (14%; Staunch Conservatives except for their stance on the environment) and Hard-Pressed Democrats (15%; Solid Liberals except for their stance on immigration.) Expect these four issues (gay rights, social safety-net programs, the environment, and immigration) to be the top issues for the foreseeable future; but while this is what politicians will be talking about, it won't be bringing us any closer to a change in the party system.

Trying to use transitions from before the civil war as a model works no better. There is simply no issue that compares to slavery in its level of importance to voters and its ability to split both major parties. I've written before about how copyright brought me here, but while it is one of many issues which can draw support from voters for both major parties, it is also one of many issues which is not important enough to enough people to build a new, successful, major party on.

The odds don't seem any better for a "centrist" party to rise up in the current environment; if anything, groups in the middle disagree with each other more than the groups on the wings. For instance, Disaffecteds and Post-Moderns are opposed over the same issues that the Republicans and Democrats are, but while they take opposing issues on welfare from their "host" party, they still disagree with each other completely. The same with Libertarians and New Coalition Democrats over gay rights. An attempt could be made to build a pro-gay anti-welfare centrist party out of Post-Moderns and Libertarians (I read their blog), or an anti-gay pro-welfare centrist party out of Disaffecteds and New Coalition Democrats (does such a group exist?), but there is a huge systemic bias against centrist parties, and the Pew report doesn't show anywhere near enough strength for such a group to have any chance to unseat one of the two major parties.


If you accept the fact that we really do have a two-party system, and you should, then you have to accept that issues, and how the issues are tied to parties, and how the issues and the parties are tied to voters, creates an incredible knot of inertia and rigidity in the political system. And currently, that system does not seem to be near any sort of tipping point. So keep hammering away at it, you 73% of voters who disagree with the major parties on at least one critical issue; your moment isn't here yet, but maybe in another eight years, it could be.

That, or you could throw your support behind better election methods, like approval voting and score voting, which allow better support for third parties and for multiple-issue-driven campaigns; then at least we could at least argue about the issues, instead just the "two" parties.


  1. I love how I can prod posts out of people sometimes with just a simple comment ;)

    I suppose one of the failures of the successful regional third party movements we've had in the US (such as the NPL of ND and the Labor Party of MN, amongst others), amongst other successful civic movements, is they were too short sighted to fight for an electoral system that'd allow for a more robust multi party system for the future, or at least to give outsiders more of a chance at representation and influence. Maybe we can get it right this time?

  2. "an anti-gay pro-welfare centrist party out of Disaffecteds and New Coalition Democrats (does such a group exist?)"

    Possibly, but I've not heard of it. There probably exists a number of socially regressive pro social support individuals (A lot of American catholics would likely fit in this camp), but I'm not sure they're trying to build a party, they're probably very scattered about. They exist in other nations though, like Australia where such a group is recently making some media waves.

    "Centrism" in the US seems to be mostly represented by what you describe the anti welfare pro social progressive types. I simply call them "Libertarian lite", which is the group "Riseofthecenter" appeals to.

    I don't think election of "centrist" parties is even ideal anyway though, but a system that allows forth multiple viewpoints across the spectrum and forcing "common ground" compromises to govern. That's just me though...

  3. Interesting how both of your examples of successful third parties, are third parties that joined with a major party. The state Democratic party in ND *is* the Democratic-Nonpartisan League, and the state Democratic party in MN *is* the Democratic-Farmer-Labor party.

    (Which I think you know, I'm just pointing that out for anyone following along in the comments.)

    Their success was based initially on being a credible threat the major parties, by virtue of the issue they rallied around and its importance at the time. But their long-term success came by being re-captured by a major party.

    The whole NPL/FLP movement is interesting, because it picked up with the 3rd-4th transition and lasted all the way through to the 4th-5th transition; Teddy Roosevelt's failure with the Progressive party comes out of that period as well, and MN and the Dakotas are where he had some of his strongest support. So I think there really was a sustained and earnest desire for "something else" in that time and place. And while it had regional success, it had to, eventually, be co-opted by the Democratic party for any national impact. (Or, you could say, it eventually succeeded by pulling the Democratic party to its way of thinking.)

    Nothing is as cut-and-dry as I've made it out to be, but sacrifices must be made for the sake of succinctness; but the apparent long term (if regionally-restricted) three-party system in the American midwest is something worth examining in more detail.

  4. Yes, I knew both of those parties merged with their state Democratic affiliates, I just thought they were worth bringing up because they were successful to the point where the Democrats felt the need to co opt them, and my feeling they were short sighted in the long run. Interesting enough, the NPL populists eventually spread up to Saskatchewan and formed the CCF, which eventually became the NDP, which is now Canada's second largest party (whether that will last remains to be seen).

    I have a personal hypothesis (I have to do a lot more reading and research before I vet it out) that one of the reasons successful third party movements in the US never resulted in a change to the voting system is that a two party "system" can work when society is far more simpler than our own, in both technological complexity and electorate complexity (when only white males could vote for example). Typically a successful third party in the US would get their main focus adopted by one of the major parties and just disperse or merge then (NPL's famous achievement is the Bank of North Dakota), I feel they did this because they figured I suppose that the conditions for these events could simply naturally happen indefinitely when they need to happen in US politics. Again, somewhat short sighted...

  5. India is an interesting example: they have a huge number of successful 3rd parties. But they do it by having just two major *national* parties and zero to two *regional* parties in every state. India is very divided regionally (by language, culture, religion).

    There's also the issue of communication; as the ease of communicating across the continent improved, it became harder and harder for 3rd parties to succeed by the regional route.

  6. No doubt, strong regional differences also create a more viable parties in an otherwise duopoloistic setting. Britain in the second half of the 1800s was dominated by the Liberals and the Tories (Conservatives), yet a third party began to rise when they instituted secret ballots for Irish votes who in droves voted for Irish nationalists. The lack of such strong differences probably is another reason why we don't have any viable third parties outside Vermont (though I would think Hawaii would be an likely example for a regional based third party to rise)

  7. What I particularly find interesting is how there aren't really any well organized third parties left in the United States (with some exceptions, like the VPP of Vermont and the MNIP of Minnesota). Even in authoritarian dictatorships there are (underground) well organized opposition parties. I'm not sure why, even with the odds beat against them, most of our third parties are pathetic in terms of organization.

  8. Perhaps we need to push one of these issues up into a major division.

  9. Speaking of Midwest, I guess we should bring up the Independent Party of Minnesota, which has done fairly well for a third party in the US. Perhaps there's still some latent expressed desire in the MidWest for "something else"?

  10. @daniel noe:

    Yup; that'd be one way to go about it.

    Unfortunately, at some point, that usually means people get shot.

    One of my big hopes for a better voting system, is that it would make these issue-related party changes not require physical violence.

  11. @TF,
    I think the strength of the US presidency and our regular nat'l presidential elections have secured two-party dominance in the US more so than Canada or the UK.

    What if we made electoral reform the key issue? The problem then being getting folks to converge on electoral reform at least at the nat'l or local levels.

    Improved single-seated rules (IRV, AV, SV or IRV3 with an AV3 first stage...) tend to be favored more so by those on the left and strongly opposed by those on the right. And then you have the acrimonious back and forth between advocates for different electoral rules. FairVote has a marketing and first-mover advantage so why not get AV/SV advocates to switch to pushing FairVote to use IRV3 with an AV3 first stage? Then we could work together instead of against each other.

    Methinks though that 3-seated state representative elections that'd pretty much guarantee the election of a Republican and Democrat from each state-district and that'd make the 3rd seat competitive would have more potential for getting support across party lines. Especially, if it were explained well how it would subvert the cut-throat rivalry between the two major parties.

    As for which type of 3-seated election, we could use citizen councils at the state-level as were used in British Columbia to push for BC-STV. Except, they would decide which 3-seated election to push for in the state.


  12. ps, I think that if we frame electoral reform as in the tradition of the civil rights movement that it is inevitable that there will be blood-shed.

    We need to make sure we follow MLKjr, not Malcom X or the Arab-springs in this regard...

  13. @DLW

    I don't want to get into the "acrimonious back and forth" (you and I have done that already) but you said:

    "Improved single-seated rules (IRV, AV, SV or IRV3 with an AV3 first stage...) tend to be favored more so by those on the left and strongly opposed by those on the right."

    I have not found that to be the case. For instance, a few years ago IRV was proposed in Alaska; it was supported by Republicans and strongly opposed by Democrats. The real pattern seems to be that the *larger* of the two major parties supports IRV, and the *smaller* is against, whichever side of the partisan line they are.

    I also want to point out that MLK was shot and killed himself. Abjuring violent methods for yourself doesn't mean your opponents always will. Everyone has to believe that violence is not the answer; which means everyone has to believe that the political process is just.

  14. I didn't want to incite more acrimonious back and forth. It seems to me like there's good reasons to favor IRV3/AV3 as a way to unite the two biggest bands of electoral reformers in the USA.

    If the goal is to secure a permanent majority then it makes sense that the stronger party would favor IRV if third party dissenters tend to be more likely to rank them higher than the other major party candidate.

    Here's a good question to ask though, "Who'd benefit [the most] from AV/STV?" and get them to support the advocacy, after all money makes the world go round...

    I agree wrt MLKjr. My point is that since the fund$/capacity for violence favor$ those in power, including their control of the MSM $pin of any such violence, protestors who want to move the center or make society deal with an issue need not to return violence with violence, which unfortunately means going against our deeply ingrained fight-or-flight impulse.


  15. @DLW

    "I think the strength of the US presidency and our regular nat'l presidential elections have secured two-party dominance in the US more so than Canada or the UK. "

    Yeah, there's some kernel of truth there, amongst other factors like debate restrictions, restrictive ballot access laws, etc. etc.


    Compromising on AV in some form or fashion might be worth considering, given FairVote's muscle and organization. It'd let us practice AV out in some fashion in a real political setting.

  16. @TF
    It's partially a matter of what's causing what. When two parties dominate, they often get to set the rules to their mutual favor as well...

    That's not necessarily true, but our third parties have been following too much the way of Ralph Nader to play the two major parties for seemingly modest but significant reforms that'll help move the center more in the long run.


  17. @DLW

    "It's partially a matter of what's causing what. When two parties dominate, they often get to set the rules to their mutual favor as well... "

    Well yeah, they set up the rules at the exclusion of everyone else. I think it's many more factors though than a singular source.

    I think one of the reasons many third parties aren't organized even despite the odds against them (with many more nations having them even when the game is stacked against them) is most Americans when despaired don't organize, they just drop out. If that attitude changed, we'd probably see some better organized third parties, and they'd probably change the discussion (even if just a bit) more into their favor.

  18. Also, I make a distinction between exclusive and "dominated". UK and Canada are two party "dominated" states, largely a product of their plurality voting system, the US is a two party "exclusive" (with a few exceptions like Vermont) state, a result of more than just a plurality voting system, but an obvious source. Lots more dirty tricks and just plain ol' undemocratic tampering that results in an exclusive system rather than a dominated state.

    I'd like to see well organized third parties holistically challenge such rigging, I don't think it'd be hard to get people to angry at such things if they knew how rigged it was.

  19. Dale, are you familiar with this paper?

    It seems to suggest that Approval Voting won't end two party domination either and is still susceptible to the sorts of problems found in other election rules.


  20. Nobody seems to be aware that in New York State, third parties do succeed. Most of the time, they only join with the major ones, but they can elect people like James Buckley. The fusion systom is why it works there.

  21. @Opinionator
    Fusion voting is a formalization of strategic voting in some elections while contesting other elections that lets third parties display their relative strength.

    I like it. I think it's easier to get if there is already substantial third party support and what-not...

    @DHS I think that there are lots of 3-seated STV elections used in Ireland. This could provide raw data for testing the impact of 3-seated STV Hare, even though it's arguable that voting strategies could shift with the change in election rules. But it'd likely be more convincing relative to simulated data. We could even limit the number of candidates per party by getting rid of their least popular third or second candidate.

    Now if I could only become a graduate student again with a focus one electoral reform...

  22. Hi Dale,
    I was reading "Majority Judgment" at the suggestion of Stephen Brams and I gleaned from it that when preferences are uni-modal that ranked-base election rules work more comparably with non-ranked-based election rules.

    Has it been tested out how the BRs for election rules change when preferences over the candidates are constrained to be unimodal?

    Also, what do you think of taking the median of a score-vote, as opposed to the use of Approval Voting or Score voting?


  23. This comment has been removed by a blog administrator.

  24. DLW, thank you for pointing out that paper. I think it's quite a vindication of some of my arguments!

    For one, they point out that under approval, more candidates can run without spoiling the election. Sounds good!

    Two, they conclude that under approval, the winning position will be more moderate. Sounds good!

    And third, they point out that 3 candidate elections actually work under approval, although 4 candidate elections may exhibit problems. (Remember though that 3 is more than 2, even if 3 is less than infinity!)

    I think you're conflating the "only two possible winning positions" with "two party domination", because as more-centrist 3rd party citizen/candidates enter the race, the SUPPLANT one of the existing two parties. In other words, 3rd parties WORK when the existing parties drift away from the consensus. And that's great!

    I'll want to double check my math, but I think they made an error in their "approval fails Condorcet" example; I don't see why the type-1 voters would continue to approve the type-2 candidate in equilibrium for the faction sizes they describe. If they don't do that, they each gain 4 points of utility, and the Condorcet winner is elected; problem solved!

    Check me on this: if the factions sizes are 3, 1, 5, 4, that satisfies their inequalities, and if type-1 votes only for the type-1 candidates, position "small" (the CW) wins.

  25. Also: Yes, uni-modal distributions have been tried. The results are consistent with other BR voting-utility distributions. (I think it was a uni-modal example, actually, that had honest-IRV still coming in behind tactical-Approval.)

  26. @Dale I was not asking about the rankings, but rather the relative "score values" when preferences are unimodal. As I understand it, it does nullify the threat of non-monotonicity.

    I haven't read the paper myself. I wanted your feedback first.