I could be talked into a jungle-primary in which the top two vote-getters proceed to the general election. It all depends on the details. That would likely cause a bit more moderation - in liberal districts, the relatively small number of Republicans might help choose between two Democrats; and in conservative districts, the relatively small number of Democrats might help choose between two Republicans. In both cases, the more moderate choice would likely win out.

But a nonpartisan ballot? No way.

The first thing you learn when walking door to door for a candidate is that 80-90% of the people you meet say, "Is he a Democrat?" and when you respond affirmatively, it's "Thanks! I'll vote for him!"

We politicos fight over the last 10-20% of the electorate - the swing. But that means that 80-90% don't care about the campaign; they just vote the party. To remove that information from the decision making process is a disservice. In fact, it'll just empower the powerful - as people will then make their decisions based on endorsements... "Gee, if Kitzhaber endorsed him, he must be the Democrat."

Voters need more info, not less. In fact, I'd support allowing people to earn the endorsements of multiple parties (at least, making it easier than it is now) and making it easier to create a party in Oregon. Then, you'd have 20-30 parties - and candidates would run under lots of banners; giving voters even more info.

More parties; not less.

Then to play devil's advocate, do you think Portland City Council members and the mayor are elected by an uninformed electorate?

Nonpartisan elections can function more easily in city council settings because the city is a small enough political entity. Candidates can go around to all parts of the city relatively easily, cheaply, and quickly. Thus, they can campaign everywhere within their district. Further, the small scale of the district allows for greater citizen involvement, information, and comment. You don't need a partisan affiliation when you know very well how the candidate stands on all the issues. This is made all the easier by Portland's relative political homogeneity and activism. People take the time to know their candidates, and, really, they're generally just choosing from flavors of liberal. Finally, once on the five-member council, the small size makes party coalitions less important - everyone knows everyone and their stances.

Statewide, such a program is harder to implement. When a legislator works in Salem and represents Burns Junction, or Lane County, it's harder to get out there and meet with voters and other legislators, get them to know you and your positions. There may be fewer people running for office; the array of positions is far greater both in larger districts and within the state legislature. Once in office, there's 30 senators and 60 representatives - knowing the D or R (or G or L) can help a lot when lawmakers try to work together.

I'm sympathetic to both Keisling's and Ringo's bills. However, I think parties do serve a purpose so I don't know if we really need to go all the way towards non-partisanship. Keisling's idea would create more involvement and debate in races that are overwhelmingly Democratic or Republican, where right now the race is really decided in the primary. It would be better to have the general election mean something.

What I would really like to see is proportional representation in the House. Eliminate districts and have people vote by party and a list of names as in a parlimentary system. Parties get seats according to the percentage of the votes they get. If the Libertarian or Pacific Green parties can get at least 1/60 of the vote, they would get a seat. We already have the Senate based on geography so leave that as it is.

Proportional representation would let parties be parties, rather than the ideologically vague coalitions they are today. It would mean parties would campiagn on ideas during the election, and form coalitons after the election, rather than worry about geography and having to form some kind of coaliton before the election.

The House would be more about parties and ideas, while the Senate would be more about geographical representation and individuals. I could see making the Senate non-partisan in this system.

So there you have it, make the Senate non-partisan and turn the House into a european style parliment with proportional representation!

Adam

LT, you asked, "what county?" did I find 80-90% Democrats.... My example should have been more clear. Typically, when one goes canvassing for Democrats, the canvasser spends most of their time on 3/4 or 4/4 Democrats....

I suppose the opposite is true for Republican campaigns. My larger point is this: Most voters in most elections, even in heavily 'swing' areas vote a party line. Political observers spend all their time analyzing the swing in the middle, but let's not make it more difficult for the 80-90% that vote more-or-less straight Democrat or Republican.

I would definitely favor a 'top two' primary, although I am fine with keeping the party labels on the ballot.

Imagine some of those exurb and rural races where we might wind up getting a moderate republican against and right-winger. Then the D's can help elect the moderate in the general, rather than losing to the extremist who wins the R primary today!

I think the polarization of politics is the central problem facing the legislature. I don't think it arises from partisan politics, but addressing partisanism may be the way to begin to fix things.

As America becomes more fragmented, we increasingly can avoid any engagement with the "other." This is a fairly well-developed theme, so I won't harp on it. The upshot is that our natural inclination toward tribalism finds plenty of cultural support right now.

I'd be in favor of anything that undermined this tendency, particularly if it was a statutory solution. Easy enough to scrap it if it is a failure. I'm also not persuaded by Cody's analysis that legislative and city races are politically different animals. Only a tiny fraction of voters ever see the candidates, even in city campaigns.

In fact (to add to my last statement above), an argument could be made that nonpartisan campaigns would increase voter participation and engagement because candidates could cross the line into territory staked out by what would formerly have been the opposing candidate's home base.

It's not the parties, it's the rural-vs.-urban. Ted Ferrioli by any other name would still be a royal pain in the keester. So, for that matter, would Charlie Ringo. 8c)

Yeah, but in rural districts, it's not a Dem vs. Republican, it's a moderate vs. hard-right Republican. And the same dynamic in SE Portland, reversed. The hope with these solutions (either variant) is to tip the balance away from fanatacism. I think it's worth a shot.

It's interesting to think how Randy Leonard might have fared in a partisan election. He might well have faced a far-left candidate with impeccible party credentials who earned endorsements from the bulk of the party establishment. Not only is Leonard able to run on his own idiosyncratic slate of issues, but he also is abetted by the nonpartisan elections in his effort to reach out to diverse constituencies. Counter-intuitively, the result is that he's garnered enormous far-left respect.

Prof, actually this open discussion was posted by Jeff Alworth without any prompting from me.... Great minds think alike, I guess...

"If we have a non-partisan legislature, on what basis would the legislature organize itself? How would committee assignments be handled? Would there still be party caucuses?"

Holding non-partisan elections would not eliminate factions or parties. Legislators would still form caucuses on a variety of issues or identities as they do now, in addition to party. If parties didn't exist, someone would immediately invent them. Holding non-partisan elections isn't the same thing as eliminating parties or caucuses. We still have freedom of association in the Constitution. But nothing in the Constitution says that government and taxpayers have to facilitate and subsidize the organization of parties and the election of their nominees.

Our legislators are smart people. I'm sure if we elected them in non-partisan elections they would still be able to figure out among themselves who to choose as leaders. People would still know generally what faction or party any given individual was associated with.

Non-partisan elections would enable a more dynamic and free flowing party and faction system where legislators could form their own constituencies rather than be restricted by party label. But they would still be restricted once they developed a particular identity and associations with certain groups.

Proportional representation would achieve essentially the same thing, because it would be easier for more parties to gain seats. Voters wouldn't have to choose just between the two big tent coalitions, Democrats and Republicans. Voters would choose from among a variety of parties that more truely represented their beliefs and interests, knowing they weren't throwing their vote away or being a spoiler to anyone. The parties that won seats would then form coalitions after the election.

I also like the idea Michael touched upon above, which is to create a non or bi-partisan panel to objectively create legislative districts, both at the federal and state levels. We've seen how partisan the redistricting process has become in several states around the country, most notoriously in Texas. As long as we have districts, they should be created and drawn as technocratically as possible rather than with an eye towards giving one party an unnatural advantage over the other. Politicians shouldn't be able to choose their own voters by rigging districts. The job should be handed over to an independent panel with equal representation. This would inevitably result in more competitive districts and fewer super-safe districts.

Thanks, Rick. One question - if someone gets over 50% in the May primary, is it over?

I'd oppose Keisling's and Ringo's measures, both, but especially Keisling's.

Keisling's idea being like Louisiana's system is a bad thing, IMHO. All it takes to ruin the system is for one party to clear the decks and unify behind two candidates, while the other party runs more than two candidates, diluting the votes accordingly, and ensuring that only one party is represented in the general.

It amounts to unilateral disarmament by one party, and I think we all know which party that's going to be, because the Republicans damn sure aren't going to dry up and blow away.

Furthermore, if your party is not representing your views and running candidates in tune with them, you have every right to switch, or even start a new party. Was it really that far back we had Al Mobley representing the OCA in the general? I can't even remember which Repub's candidacy he torpedoed (Frohnmayer?), but I'm damn thankful that our side didn't split off from Barbara Roberts, allowing her to be one of my favorite Governors ever.

(Why my brother got to hear her speak at his grad. from Tech, while I got a Construction Company owner or whatever for mine at Western, is another matter, which I will happily bring up when next I talk to her.)

As a Democrat, I shouldn't be allowed to "help" the Republicans decide who they want to stand up in the general without my input. And neither should the Republicans decide my candidates.

You can't legislate more support for third parties where before there was less, either. They've either got it, or they don't. People don't stop standing behind their values, or listening to the presentations, just because the labels have been removed.

But you can sure wipe it out:

Keisling's proposal narrows it down to two in the general, so if anything, that ensures that a Green, New Party, Constitution, Reform, or whatever candidate will never appear on the general ballot. Even the worst major party effort in a statewide race in Oregon gets 30%, usually, a nut even the Reformistas couldn't crack with Perot.

So why don't we propose second-choice, instant-runoff voting instead?

Hi Rick,

One critique of Keisling's primary proposal is that it seems to heavily reward party discipline, which tends to happen more on the R side than not (as John D pointed out).

It's not hard to imagine a scenario in which there are two Republicans and three Democrats running in equally competitive primary races. Assuming a competitive primary in both, the Dem candidates could split their vote in thirds, while the Republicans split their primary vote 51-49%. Numerically, the two Rs would be the top vote getters.

Even if over 50% of primary voters cast ballots for Dem candidates who shared basic progressive values, the General election would end up being a contest between two Republicans.

Again, this is just one scenario in which a majority of voters casting ballots for progressive candidates would have to face a general election between two conservatives.

No system is perfect, but I wonder if there aren't some unintended consequences that might make this system less fair and less democratic.

I think Alworth is monitoring our email!

Mwoh ha ha!

Charlie,

You capture precisely the direction of my thinking.

Phil wants to implement this system because he believes it will reduce partisan polarization in the legislature. That is the primary rationale, as far as I can tell. The second is to enhance participation and provide voters more influence over the final outcome.

With respect to the first, Phil's proposal has a number of embedded assumptions, and until I see these tested, I remain skeptical. Rick implies the same when he says that, under this proposal, candidates would be encouraged to construct winning coalitions for the general earlier (the implication being that candidates who could appeal to a broader set of citizens would be more likely to finish in the top 2 in the primary and thus be in the general).

This depends critically on three things: a) the distribution of preferences in the population, b) the rate of turnout in the primary, and c) the number of candidates.

First, the Keisling proposal implicitly assumes that voter preferences in Oregon follow a "bell shaped" (flat would also work) distribution, with lots of voters clumped near the median.

If, however, the distribution of voters is bimodal -- and presidential voting patterns in Oregon indicate that they are -- then candidates will not "run to the middle" in the primary, but will instead run to the centerpoint of their support coalition.

Second, the Keisling proposal does not consider that turnout in primaries tends to be much lower than in the general. They may reply that this is precisely because Independents can't vote in the primary, and this may be true. However, even in states with open primaries, turnout is much lower.

The strategic situation for a candidate once again may be this: run to the midpoint(s) of the preferences of the primary voters. And all extant evidence is that primary voters tend to be more extreme ideologically than general election voters.

Finally, the number of candidates. In a multicandidate race, you are not bound by the logic of 50% + 1, which means that, in a two candidate race, you have every incentive to run to the middle, since the voters on the wings have no other alternative to choose from.

In a multicandidate race, a winning strategy could be to construct a disciplined support coalition totalling 20 or 30% of the voting population. And, given everything else I've written about the distribution of preferences in Oregon (ideologically and geographically) and the first stage results from Louisiana, this seems a very likely outcome.

So here is, to me, a possible statewide scenario. You end up with Kevin Mannix on the right, drawing his support from a highly disciplined christian and economically conservative segment; and Erik Sten on the left, drawing his support from the liberal activists in Multnomah, Lane, and Benton counties. (I'm not trying to bash Mannix or Sten here, but I don't think either would deny that they are relatively conservative or liberal).

<h2>I suppose I need to be convinced that this is not only a possible, but in fact a likely scenario. I know what kind of evidence I could use to test the claims I make above (especially the geographical and ideological distribution of voters / preferences in Oregon), but I don't have the time or the data to look at it right now.</h2>