The Washington legislature’s 18th district is a safely Republican district. Earlier this year, the incumbent for one of the two State House seats in that district said she would not run for re-election. Soon after, six Republicans announced that they would run for the seat (House Seat 18-1). Each Washington state legislative district elects two House members. The two contests are separate from each other on primary and general election ballots. Candidates either file for seat (1) or seat (2). The voters elect one candidate in the 18-1 contest, and one candidate in the 18-2 contest.
Republican Party officials have begun to worry that if there are six Republicans in the primary, the party may lose the seat. Washington is one of two states that uses the top-two system, in which only the two candidates who get the most votes in the primary may be on the general election ballot.
In November 2008 the vote on this seat had been 60.0% for the Republican and 40.0% for the Democrat. Filing doesn’t close until June. If two Democrats file, and there are six Republicans and two Democrats on the August 2010 primary ballot, it is highly likely that the two Democrats will place first and second, leaving no Republican on the November ballot. So a district that leans Republican could have no opportunity to elect a Republican in November, except by write-in vote. See this story. Republican Party officials will wait until they know if two Democrats will file. If two Democrats do file, the party will then undertake an informal process to limit the Republican field. Thanks to Krist Novoselic for the link.
Perhaps one of the Republicans might – gasp! – go talk to Democratic and independent voters and seek their support.
This is not a flaw in the “top two”, this is exactly how it makes only appealing to one’s partisan base a losing strategy. Want to win? Find a way to represent the most voters, not just those in your party’s base.
That gives the independent voters a potentially huge role there. Of course it’s not enough to just change the election rules, we have to organize voters to participate in the process.
Hey, I know; the six Republicans could have a pre-primary, and the top two from THAT contest could run in the open primary; that’s the ticket…
Primaries, open primaries, top-two primaries, even all third-party harming ballot-access laws; it’s all a cludge that fails to acknowledge the real problem: The fact that plurality sucks when there are more than two real options.
Approval (or score) voting! For a democracy that can count to three!
I’m sure they would all go and talk to Democratic and independent voters and seek their support, if they all remain in the race. But what is more likely is that most of them will drop out and voters won’t have any opportunity to support them. Ironically, top-two increases the power of the “party bosses.”
This seems cool to me. I want more Democrats in my state!
Elections seem increasingly irrelevant. Democrats, Republicans, variations on the tune are minor. I think the two corporate parties should decide in a back room with their corporate masters who will be the representatives for particular districts. I no longer vote in general elections in WA State, except for referendums or initiatives. Only the feeble minded and those who learned of voting via American Idol, vote when the choices are only Democrats or Republicans, or two of one or the other of the corporate parties.
#3:
You’re probably right Richard.
Open primaries come to this; where before you had an either-or choice between one D and one R, now you can safely choose between two of them!
…Unless _both_ of them run two (or more) candidates.
…Or if there’s any third-party options running.
…Or if you care which member of your party wins, not just that some member of your party wins.
In any of these cases, you go right back to square one. So, in practice, top two has few upsides, and occasional devastating downsides (wizard vs. lizard).
Actually, I think it shows the flaws of the the two-party system. Candidates are afraid to run as a different or independent party because the public has been brainwashed into believing there are only two viable philisophical political parties. We need to change the mentality of the public to believing there are alternatives to the Demorepublicratican Party.
Furthermore, IRV would prove beneficial in this situation and in most elections with a plurality of candidates.
Oooh, ooh, Rich, I was hoping we could be buddies here…
One, the two party system isn’t a brainwashing: the way our voting system works, the only game-theoretical rational response to it, is a two-party system. It’s sad, but no amount of crying or wishing it weren’t so can change that. We have to change the system.
But IRV doesn’t fix the problem. IRV still collapses toward two-party domination, just like plurality. (See: Australian house.)
Top-two open primary is, in the 3-candidate case, functionally equivalent to IRV. But even in that case, neither TTOP nor IRV can guarantee that the best candidate wins, only that the worst candidate doesn’t. Voters still get pushed towards picking “the lesser of two evils”; that’s what “wizard vs. lizard” illustrates.
If you really want an alternative to the Ds and Rs, educating the people won’t help; IRV won’t help; you need a voting system that can actually pick between THREE candidates, not just two, and then the people’s own self-interest will allow them to break the cycle.
And the system that can do that is approval voting (or score voting).
Proportional representation would also work.
#10. Yes, proportional representation would work. But you should strike the word “also” because no voting rule in single member districts is going to elect legislators from more than two parties — not IRV, not approval, not score voting.
Even plurality voting can elect legislators from more than two parties under some circumstances — districts small enough, support for a party geographically concentrated enough. Changing the voting rule doesn’t alter these circumstances (which we should pay more attention to).
IRV would give everyone a better idea of the true preferences of the voters, and allow small parties to play a new role as junior coalition partners. That doesn’t get their candidates elected. Approval/score voting, I’m not so sure. The strategic incentive to bullet vote might overwhelm them, leaving us exactly where we are now.
#11:
You’re confusing me.
You first say single-member districts can’t elect members of multiple parties; which I take to mean that, of course, since there’s only ONE winner, they must be from a SINGLE party; is that what you’re saying?
Because then you say that plurality CAN elect two parties; but then you’d have to be talking about some proportional plurality system, or something? You lost me.
Anyway; IRV. It’d work “fine” as long as you assume the open primary is full of multiple candidates from the same two parties and all voters consistently vote along partisan lines, or at least that every of every third party falls-back to the same major party. But those assumptions are both wrong and do nothing to actually get third-parties elected. This is where I see a lot of IRV supporters coming from, intellectually: they confuse independence of clones (which IRV does have) with independence to favorite-betrayal (which IRV does not have unless you assume third-parties will never beat either of the two major parties, i.e., third parties can only win when you pre-suppose that third parties can’t win).
(I cover this more, and describe it better, at my blog (click my name).)
Now, as for score: let me tell you what you’re actually describing when you talk about the incentive to bullet vote. You’re describing the fear of accidentally electing your SECOND favorite choice instead of your FIRST choice. You’re afraid that, if you mark both, then you could have had it BETTER.
In every other voting system, the fear voters must contend with is of accidentally electing their LEAST favorite choice instead of their first.
Approval (and score) are a huge improvement because of this simple fact. It means third parties, and moderates, actually have a fighting chance, instead of practically none at all.
The State and the Grange need to settle the current lawsuit by giving parties back their associational rights. This way the hacks, bosses and whatever derogatory name you need to give party members can be free to nominate candidates who might not speak to the majority of voters. Let them nominate losers I say.
Top-Two is a majority voting system where you can vote your conscience in the first round, and if that candidate doesn’t get to the runoff, the voter has a second choice.
If a candidate loses the private nomination, they should be able to run as independents or third party.
But will the two major parties accept such a settlement? I don’t know – they’re lost in the illusion that the Open Partisan Primary will come back. If that ever happens, say hello to a credible effort that proposes Washington go all the way non-partisan.
#1 So turning primaries into crapshoots is a feature of Top Two, not a bug? The concern that Richard expressed here is quite valid. Had Illinois used Top Two this year, there would have been two Democrats on the ballot for Governor even though more than 45% of the votes in the primary were for Republicans:
Democratic primary
Quinn 460,391 50.4%
Hynes 452,304 49.6
Republican primary
Brady 155,327 20.3%
Dillard 154,907 20.2%
McKenna 147,599 19.3%
Ryan 130,400 17.0%
Andrzejewski 110,735 14.5%
Proft 59,178 7.7%
Schillerstrom 7,388 1.0%
In the Lieutenant Governor’s race, there would have been two Republicans on the ballot, even though a majority of the votes were for Democratic candidates:
Democratic primary
Cohen 212,902 26.0%
Turner 182,432 22.3%
Hendon 113,273 13.8%
Boland 105,551 12.9%
Castillo 105,056 12.8%
Link 99,972 12.2%
Republican primary
Plummer 237,646 34.0%
Murphy 232,924 33.3%
Tracy 79,870 11.4%
Cole 61,165 8.8%
Cook 55,205 7.9%
White Sr. 32,184 4.6%
The current plurality system is pretty bad, but disenfranchising huge swaths of the electorate in the general election because of how candidates ended up splitting votes in the primary is even worse.
P.R. and A.V. — all the other stuff is too complex and/or dangerously obsolete.
#12: I thought that what I wrote was reasonably clear, if not exactly elegant. I still think so. Single-winner voting rules have a very strong tendency to limit the viable choices in any one district to two. The tendency is so strong that we often drop the word “tendency” and say things like (quoting myself) “no voting rule in single member districts is going to elect legislators from more than two parties”.
The two viable choices in one district may be different from the two viable choices in other districts. That was the point of my comment about district size and geographic concentration. The U.K. has plurality voting in single-member districts and still has three national parties, plus some regional ones, that get seats (although grossly disproportionate numbers of seats). The U.K. House of Commons also has one seat for every 95,000 persons, the U.S. House of Representatives one seat for every 710,000 persons.
No voting rule can overcome this tendency, holding district size and political demography constant. Dale and I agree that IRV will not do so. He thinks that approval/score voting can do so. I do not.
Let’s make this easy and let Microsoft run for office.
What an absurd system.
Hey, Dale, I have a question. How about this proposal?
– You can vote for or vote against any candidate
– Candidate with the highest rate wins
– Rate score formula: (Votes for+ Blank Votes/2 / Votes against)
“Top Two” will always, in the end, result in a one-party power sharing structure.
“TopTwo” is designed to create a one party, fascist-socialist-communist state.
Bob Richard said:
No voting rule can overcome this tendency [toward duopoly], holding district size and political demography constant. Dale and I agree that IRV will not do so. He thinks that approval/score voting can do so. I do not.
But Dale actually has evidence to support his position, whereas you do not.
There is absolutely no evidence that some property of single-winner districts inherently leads to two-party domination. And you have not even attempted to offer any evidence to that end.
But there are properties of voting methods which account for this tendency. We have even cited the fact that most countries with ordinary runoff elections (not IRV) have escaped duopoly in single-winner districts — so we have refuted your position with concrete evidence.
And we all know perfectly well why plurality voting gives us duopoly. Even staunch advocates of IRV have made the point. It’s simple game theory. A vote for anyone other than the first and second place candidates is “wasted”. Thus people tactically vote for their favorite among those two considered to be the front-runners, regardless of who their overall favorite is. Thus one of those two front-runners almost always wins.
But IRV has the same property, although it is not obvious in cases where the minor candidate(s) are weak. This is just an empirical and mathematical fact. We have ballot data from the last IRV mayoral election in Burlington VT that gives us a particularly telling example. We have decades of IRV use in Australia to support this position. There is just no genuine controversy on this issue. That explains why you never offer any evidence to support your position, but instead just assert things, like “Dale and I disagree”.
Score and Approval Voting have properties (namely, passing the FBC) that make them seem almost certain to avoid duopoly. I cannot think of any theoretical reason why we would still expect to have duopoly if we were using Score or Approval Voting. And apparently you can’t either, because you’ve never offered one. You’ve just asserted that they won’t change things. I think the reason you haven’t offered any evidence to support your position is because there just isn’t any.
Krist said:
Top-Two is a majority voting system where you can vote your conscience in the first round, and if that candidate doesn’t get to the runoff, the voter has a second choice.
That is simply inaccurate Krist. TTR (“top-two runoff”) is only free from strategy in the second round, because there are only two choices. Here’s a concrete example proving this. Say you have blocs with these preferences:
40% Progressive > Democrat > GOP
21% Democrat > Progressive > GOP
39% GOP > Democrat > Progressive
The bottom bloc of voters honestly prefers the GOP candidate, but if they pay any attention to polling on head-to-head match-ups, they will see that it is futile to waste their vote on the GOP candidate, because he will almost certainly lose in the runoff, regardless of who he faces. Thus they are better off to support the Democrat.
Nevertheless, TTR is more conducive to multiple parties than is IRV, because it gives voters an incentive to strategize about the second and third place candidates in the first round, rather than the first and second placing candidates. To help or hurt the first-place candidate is pointless, because even if he gets knocked down to second place, he’ll still make it to the runoff. So voters need to be tactical about which candidate they want to challenge him in the runoff, thus they need to think tactically about the second and third placing candidates. But then, doing that can bring down the first-placing candidate, so these tactics cause diverging instead of converging. This probably explains the root reason that most of the world’s 27 TTR countries have broken free of duopoly.
#13 Voters in Washington are quite free to associate with one another in support of political candidates, providing financial and other support, recruiting candidates, handing out campaign literature, etc.
#23: The parties in Washington state want to be able to have their members associate in order to officially nominate candidates. That’s why the parties are litigating against the “top two.”
That’s also why the parties in California are opposing Prop. 14– Maldonado’s Folly.
#24 Why should a State officially recognize the nominations of a political party, for ballot access purposes? It is simply unwarranted interference by the State in the political activities of its citizens.
The political parties oppose Proposition 14 out of their own self interest, or perhaps the self interest of the official party hierarchy.
#25: All 50 states and the District of Columbia officially recognize the nominations of political parties for president.
49 states– all but Washington– officially recognize the nominations of political parties for Congress.
48 states– all but Washington and Louisiana– officially recognize the nominations of political parties for all or most state offices.
If you’re going to reverse this trend, you’ve got a long, hard road ahead of you.
“The political parties oppose Proposition 14 out of their own self interest…”
That’s especially true of the small parties, since their very survival is at stake.
#26 Once upon a time there was slavery in a lot of States.
NOT any more — via the very hard way in 1861-1865.
How about the nonpartisan top 2 stuff in the NE legislature ???
#26 Are you sure about Arizona presidential candidates? See their certificate of ascertainment for 2008.
Pingback: Ballot Access News » Blog Archive » Republican Legislative Candidates Begin Dropping out of Washington State Legislative Race, as a Consequence of “Top-Two” System
San Francisco had top two voting and a Green got elected, supervisor, and received 47% in the mayors race after receiving 18% in the first round, and being outspent 10-1.
Generally, in the top two, you have a front runner and a group of challengers.
The ability of people to vote for their favorite candidate in the first round and then make a lesser evil vote in the second round means that you get people deciding between two choices they might not otherwise make. Third parties have historically done best under this system.
San Francisco has since instituted IRV and is back to a one party town.
Just a data point.