Emergency electoral reform: OLPR for the US House

Because the constitutional emergency is likely too deep to just turn the page, small-d democrats face an emergency of another kind. The need to adopt proportional representation has never been greater. The country simply can’t afford the risk that the Republican Party does nothing fundamental to reform itself, and wins back the House in 2022. A change to some form of moderate proportional representation (PR) is essential.

Given the current balance of power in the House, the Republicans would need to flip only about seven seats in 2022. (There are currently three vacancies.) With rare exceptions, presidents’ parties lose votes and seats in midterm elections. With the balance so tight, there is almost nothing to stop Republicans from winning back control of the House, other than perhaps if they descend into internal party chaos. They just might do that. They might even split. But I don’t like seeing the fate of the republic depend on Republicans finding yet another way to squander an easy electoral win that’s there for their taking.

I am not arguing for a change to PR only for the sake of the Democratic Party. In fact, my argument is that this is a way for Republicans to save their own party. The country needs functioning pro-democratic parties on both the center-left and the center-right. At the moment, it has such a party only on the center-left, and even that is a temporary ceasefire amidst a deepening internal division.

Cleavages in American politics today and the need for PR

I would identify three key cleavages in American politics at the moment. (Note: issue positions and cleavages are very much not my academic speciality at all. I admit I am simplifying, but the divisions I identify should be reasonably accurate as a broad summary.) There is the Republican–Democratic cleavage. This one is almost evenly divided, which explains a lot of the current partisan polarization. Hold together just enough–avoid the proverbial circular firing squad–and you can win. Then there is the democratic–authoritarian cleavage. On this one, the pro-democratic segment extends all the way from the leftmost large-d Democrats to somewhere near the middle of the Republican Party. The pro-Trump, white-supremacist, election-denying wing of the Republican Party has shown itself to be completely willing to set aside democracy, and even to promote/tolerate political violence, in order to advance its political agenda. This wing is a cancer that must be removed from the right-wing bloc that currently consists solely of the Republican Party. Then there is, for lack of a better term, the capitalist–socialist cleavage. This one obviously divides the Democratic Party. On one side are Democrats who generally take a more gradualist view of the need for economic policy change, plus nearly all of the right, in being free-market oriented. On the other, left or “progressive” side are Democrats who emphasize various proposals to remake the economic model (including less commitment to free trade), whether or not “socialist” is the correct term or even the term they favor. Think Bernie Sanders and his supporters, as well as some of the “progressive” wing of the Democratic Party. Basically, the point is that there are (at least) two “rights” and two “lefts” but currently only one party on the right and one on the left. And the emergency is that one of the “rights” has abandoned democracy and shown a willingness to accept political violence.

The need for PR is to let the free-market small-d democrats in the currently existing parties act independently of their more extreme wings. This is precisely what PR systems permit–each side’s extreme can be its own party rather than a wing of one majority-seeking party, without raising concerns over “spoilers” that arise under plurality elections.1

As I already conceded, I am oversimplifying a complex political scene for the sake of argument. I also am not going to go into the details of how actual coalitions would work under this stylized latent four-party system that PR would allow to break forth. Both the need for electoral coalitions in single-winner offices (Senator, President, governors), and forging legislative coalitions among these parties in the House, would complicate the flexibility of alliances that one obtains when PR is used to elect a single dominant institution (as in many parliamentary systems). The point is simply that PR offers the best means of generating center-spanning coalitions to control House majority outcomes, in contrast to the current system’s generation of majorities that include a fringe–a nakedly authoritarian fringe in the case of the party most likely to win a majority in 2022 under current rules.

So, we need PR to save democracy. But what kind of PR? I would take any kind over the system we have now! But I think there is one that recommends itself because it is the easiest to implement, for voters to understand, and for election authorities to administer.

A model of open-list PR for the US

I favor open-list PR not because it is the “best” system or my personal favorite. Strong cases can be made for single transferable vote (STV, which is a form of ranked-choice voting in multi-seat districts) or for mixed-member proportional (MMP). However, open-list proportional representation (OLPR) best meets the criteria of simplicity in implementation, voting, and administration. My argument for OLPR is inspired partly by my own sense of what is workable, but more largely by a post by Jack Santucci.

It literally could be made our electoral system tomorrow, as follows (I am setting aside the fact that there is a reapportionment and redistricting taking place in 2021-22 in my “tomorrow” scenario). Take 3–5 existing contiguous single-seat districts and merge them into the multi seat districts needed for PR. Thus the proposal is for districts with district magnitude (M) of three to five. (Later I will address states that have fewer than three Representatives.)

Each voter would have one vote for a candidate, just as now, but the ballot would list all the candidates of each party that are running in the larger multiseat district (up to M candidates per list). The initial allocation of seats would be based on summing votes of party candidates nominated to each list, using one of the standard PR allocation rules (I’d favor D’Hondt, but various others could be fine). Then, once each list’s seat total is determined through the application of the PR formula to its collective vote total, its top s vote-earners get the list’s seats (where s is simply the number of seats the list has won). This is standard OLPR, or more formally, it is quasi-list PR, because there is no opportunity to cast a vote for the list as a whole.2

An important question is how to handle nominations to the lists. Personally, I’d prefer to get rid of primaries, as when there is a wider range of choice of both party lists and candidates on those lists, primaries arguably are not needed. However, no proposal that abolishes primaries is likely to fly, politically. I would not let that bog the emergency reform down. I propose embracing ideas that are already out there and being pushed by the independent-politics reformers, such as “top two” and “top four”.

How would this work? One could continue to hold a “primary” in each of the existing single-seat districts; I will now call these nominating districts to distinguish them from the larger general-election districts. The goal here is to avoid making it as unwieldy as it could be if primaries were held in the larger districts to be used in the general election. The first round (call it a primary even though it would stretch the definition thereof) would advance the top c candidates from each nominating district, where c could be four but could be some other number agreed upon.3 Presumably, as is the case in California’s “top two” currently, the candidates themselves would indicate what party they affiliated with on the primary ballot, but use of the label would not be restricted by any central actor in the state (or other level) party.4

Then, between this primary and some date in advance of the general election, let the top cM candidates for the larger general-election district negotiate who goes on whose list and how those lists are branded. The party labels could be ones that are registered in advance of the election (i.e., before the primary) as is currently the case in many states, or it could be left completely open for actors to negotiate between rounds. This is an important detail, but not one I think should be essential to advancing the wider proposal. It could even be a matter of individual states to sort out.

The idea here is that the top-c first round in nominating districts, followed by negotiations over lists for the general, encourages those who have advanced to a slot on the general election ballot to cooperate in order to maximize their seat-winning potential in the OLPR process. At the same time, however, it allows these candidates and their allies to reject anyone who has qualified for a slot on the ballot from being on their list if he or she is too extreme for the brand they want to cultivate.

If general-election lists are restricted to M candidates, then in any case where two or more of the same party have qualified from a given nominating district, one will have to be left off the list, unless there is another nominating district where no candidate of that party qualified. The objective here is not to force any set of candidates to run together. Local actors, including the candidates, decide. They have to balance the supporters that a given candidate can bring with the risk that some candidate drives away other voters in a context in which any given list is likely to win 1 or 2 seats in a three-seat general-election district (or 1–3 in a 5-seat district), rather than 100% of the representation of the single-seat districts, as under the current system. I am not wedded to the various components of this idea, and am completely open to other ideas. The wider point is that there are reformers who dislike parties and there are reformers who want stronger parties. I am looking for a way to thread a narrow needle and build a reform coalition–under emergency conditions.

When coordination fails and candidates who have qualified for a given party exceed M in some district, but they can’t agree on which M get to use the name, what do we do? While I would not normally advocate multiple lists within a party, I’d be willing to allow it to make the idea of lists and PR work. Also, any candidates who, having qualified in the primary, do not find partners to go in together on a list should be free to run as independents.

I should conclude this section by noting that my OLPR proposal is totally severable from my nominating-districts and “top c” proposals. If the latter get in the way of OLPR, I am happy to drop either or both. My ambition is to help make the transition to OLPR politically smoother, by retaining smaller geographic entities as politically meaningful aspects of the implementation of PR (through the nominating districts), and retaining the “bottom up” qualification of general-election candidates that is a hallmark of the current system. The overriding objective is to let different wings of current parties compete separately, outside of a majoritarian context in which splits become spoilers, and general-election candidates are sometimes extremists themselves or are in debt to extremists in their party. Avoiding these pitfalls of the current system is the very essence of PR.

Other issues

I am assuming this proposal stays within the current 435-member House. There are arguments to be made in favor of increasing the size of the House, but I have my doubts that a larger House is by itself inherently valuable. It certainly is not worth the risk of its becoming a poison pill that prevents PR. If advocates of electoral reform make a larger House seem like a condition of electoral reform, the cause of reform is probably doomed.

With a 435-seat House, and even with any House of reasonably achievable larger size, there will remain states with only one or two members. These states will obviously not be able to have districts that elect 3–5 members apiece. So what? Many PR systems have a few districts with one or two members, even when their national average magnitude is larger. This is not a reason to reject a proposal for reform. States that have one Representative could be encouraged to adopt ranked-choice voting, but should not be required to do so.

I should address why I do not advocate STV as the overall system for the House, given the current fashion in some circles for ranked-choice voting solutions. This is not the place to go into reasons why STV may not be desirable in its own right. It has some strong positive features, but also some negative ones. The biggest negatives are the need for voter education, substantially changed ballot formats, and already overstretched election administrators having to adapt their routines to make the more complex counts work. OLPR allows all of this to be as close to the status quo as possible, while still getting PR.5

What about MMP? I have been known to argue it is a good system. However, absent substantial increase in House size, it has some real drawbacks. The single-seat districts have to become considerably larger geographically for MMP to work with the existing state delegation sizes. (The list tier for MMP in the US surely would be state-by-state, or regions within larger states, not nationwide or otherwise multi-state.) The OLPR proposal that I am advancing here also means larger general-election districts, but has the advantage of having more than one member elected from each of these larger districts, while also retaining the more compact districts for nominations. An additional drawback of MMP in the American national context is in how you implement the list tier. It is either closed lists, which might be politically unpalatable, or it is open lists alongside the two-tier structure, adding a considerable further complication.6

So, no, I have not abandoned my general preference for MMP, nor am I claiming STV is a “bad” system. I simply am arguing that OLPR is a good solution to an immediate emergency for democracy.


We must find a way to prevent a new House majority from being elected in 2022 that is under the effective control of an anti-democratic wing. The voters who prefer a center-right party are not going to vote for the existing Democratic Party as long as they fear (rightly or wrongly) that that party is coming under the control of its own extreme “socialist” wing. Voters need choices that are more moderate, as well as parties that can represent voters with grievances that lead them to reject mainstream politics. What we need to avoid is a mainstream party winning a majority of seats while under the control of its grievance-based authoritarian extreme.

I am under no illusions that this will be easy. I certainly accept that any PR proposal is less likely to pass than likely. It requires more institutionally oriented Republicans to see a clear and present danger from continuing to work within a party that has a strong and undeniable anti-democratic tendency, as well as to believe that tendency is too large to be contained within. It also is not going to be immediately embraced by the Democratic establishment that just won all three elected components of the federal government, and so requires them to realize just how fragile and transient their control is.

Difficult though it is to get this proposal accepted, we are in a situation where an emergency exists for democracy. So let’s get to work!


[Over the years I have done many posts on the idea of adopting proportional representation (of some form) in the US. Please click here and scroll to see them all.]



  1. Advocates of ranked-choice voting in single-seat districts (also known as the “alternative vote: or “instant runoff”/IRV) will say that their preferred system also avoids the spoiler problem. This is not fully correct. The issue is that this view takes a district-level perspective. The point of PR is to avoid “spoilers” in larger ideological blocs. Getting the same result from IRV requires something approaching uniform distribution of those blocs across districts, or at least for each group within a bloc to have its own local strongholds, so that the parties/factions within a bloc can meaningfully trade preferences. Otherwise, it mostly leads to the same issues as plurality voting, whereby to win, the larger party/faction within the bloc must appeal to voters of the other. The case for IRV in the current emergency would rest on an assumption that, within the right, the authoritarians are the smaller component. If they are not, they will either win from preferences of those on the moderate right, or will potentially win pluralities of the vote when many voters don’t give second preferences. (We can’t be certain that voters for the mainstream center-right will preference a party on the mainstream left. Maybe they will, maybe not.) This brings me to the final issue: IRV advocates tend to overlook that the best case for the system assumes compulsory preferences, which are unlikely to be adopted (and may even be held unconstitutional) in the US. If many voters give only first choices, then IRV is more or less the same as plurality.
  2. Such an option could be added, but I am trying to keep it as familiar as possible while still getting PR.
  3. It might be wise to set c to the same value as the general-election M; it certainly should not be much smaller than M.
  4. I don’t think anything that generates such control over labels is politically palatable in current American politics, even though most political scientists would say it is desirable.
  5. If the reform included a clause allowing individual states to opt for STV instead of OLPR, I would not object.
  6. There is also the need to prevent parties from gaming MMP with “dummy” lists. This has been discussed previously on this blog. It can’t be dismissed as a serious problem, and so I’d rather just sidestep it in designing a proportional system for the US in the present moment.

64 thoughts on “Emergency electoral reform: OLPR for the US House

  1. P.s

    My comment would be following: the risk of factionalism, rise of campaign expenditure…


    On Sun, 17 Jan 2021, 21:30 Gianluca Passarelli, wrote:

    > I d comment first but I d need the password though:-) > > On Sun, 17 Jan 2021, 20:11 Fruits and Votes, > wrote: > >>


    • Strange that Word Press even allows comments on a post one can’t read. Anyway, it is now “live” and I changed the post date to today’s in order for it to appear as if it were new.

      This comment is here just to explain why there are comments that pre-date the post itself, or appear to do so.


  2. I don’t know if I really need to respond to a comment that had not been prompted by anything more than the title, but…

    1. We already have factionalism, and one faction is fascist (I don’t use the term lightly).
    2. We already have ridiculously expensive campaigns.

    OLPR, especially with low M, is not going to make either of these problems worse than the American status quo, and I think a case can be made that it will make them less troublesome.


  3. Pingback: Matt Shugart Makes the Case for Proportional Representation in Congress to Save the Country from Authoritarian Wing of Republican Party | Election Law Blog

  4. I agree with both you and Dr. Santucci. This is likely the quickest and easiest way to go. Barring this, we’ll have to put our faith in Skowronek being right and Biden being a Reconstructive President who consolidates a new Democratic majority in the midterm, a la 1932.


  5. This is an excellent proposal. I believe the arguments about the specific system apply even in a non-emergency scenario, including the very serious implementation challenges involved with STV; I have of course previously expressed reservations about MMP additional to what you mention here. In short, open-list PR is the way to go for the US.

    Moreover, I think the scheme to retain primaries at the SSD level is an admirable way to adapt current practices which may help the transition process along considerably (though I would worry somewhat about issues such as a district not electing any of “its” candidates, as arise using district-ordered lists (see my previous post on this topic here: https://tinyurl.com/y4cngfux). The only thing I would dissent on is using or mandating any system that would weaken control of the party label. Just adapt current laws to say that any top X candidates nominated on the SSD would be on the party’s list, with the SSD nomination process otherwise proceeding just as it does now.

    Liked by 1 person

  6. NOW –



    S-I-M-P-L-E PR —







    Example 100 Votes, Elect 5

    Ratio = 100/5 = 20

    Surplus Moved
    C1 25-20 = 5 Surplus
    C2 19+1 = 20

    C3 14+4 = 18

    C1 20 = 20 Elected
    C2 20 = 20 Elected
    C3 18+2 = 20 Elected
    C4 17+3 = 20 Elected
    C5 15+5 = 20 Elected
    Sum 90+10 = 100
    Losers 10 are moved to elected persons.


  7. I think this is a solid proposal. It’s three important selling points:
    1. hugely better than FPTP
    2. clearly well-thought-through and practical from an implementability perspective
    3. arguably viable as a matter of pragmatic politics (largely because it’s reasonably non-disruptive to incumbent Democrats and to the Democratic party as a whole).

    I’m going to suggest two alternatives, and compare them in terms of pros and cons. I think either of my alternatives share all three of those selling points, though I understand there’s room for debate on that.

    Alternative 1: Mixed Transferable Voting https://electowiki.org/wiki/Mixed_Transferable_Voting . This is a form of MMP with simplified choose-one ballots through pre-declared delegated trasfers.

    Such choose-one ballots resolve a few of the issues Dr. Shugart has pointed out with MMP. For one, it is an open-list method — in fact, even more faithfully open-list than his OLPR proposal, because it uses explicit intra-party transfers. For another, it avoids the Bavarian-style ballot complexity of other open-list MMP methods. And finally, it avoids the “dummy” list problem entirely.

    However, some of the MMP issues do remain. It would require either increasing the size of the House (to at least 650, or a bit more if DC and/or PR become states), or increasing the size of each primary district (thus pitting multiple incumbents against each other in some cases). Furthermore, the predeclared delegation mechanism it uses to achieve its simple ballots is novel from an international perspective, and probably philosophically hard to swallow for some reform activists (even though, at worst, its pathologies would clearly be less than those of FPTP).

    So, if Dr. Shugart is correct that increasing House size would be a poison pill, and if the alternative of throwing multiple incumbents into the same primary would also be such a pill, this is less-viable than his/Dr. Santucci’s OLPR proposal.

    On the other hand, it would have advantages over OLPR. As an at-large method, it would clearly be more exactly proportional in states with over 5 seats total than 5-seat districts would. And as a transfer-based method, it would be more proportional for intra-party factions, even in states with 3-5 seats.

    Alternative 2: PLACE voting https://electowiki.org/wiki/PLACE_FAQ . This is a novel, at-large, biproportional method with choose-one ballots; like the above, it uses pre-declared delegated transfers to get that ballot simplicity.

    This is similar to alternative 1. It avoids the need for a either a larger House or larger districts, but in exchange it has even more novelty and internal complexity (though it remains simple from a voter interface perspective).

    If either of the alternatives I’ve given turns out to be politically viable, I’d prefer them to OLPR. However, if they’re not, I would absolutely support OLPR. In particular, I think that the OLPR proposal here addresses key problems with STV as in the current Fair Representation Act; I think that STV, unlike OLPR or the alternatives I’ve given, would be too scary for both incumbents and parties to get the super-majority of Democrats that it would need to pass.


  8. Can I suggest a sunset clause on OLPR? Quick introduction and ease of use is certainly a major point in its favour. However, it’s worth recalling the example of South Africa.

    Closed List PR is a fine way to elect a parliament at the onset of a democratic transition. However, it has been increasingly questioned since the end of the transition, several inquiries have recommended a shift to more accountable forms of representation, and the constitutional court recently ordered parliament to adopt a different electoral system. Electoral systems seem to become set in stone and while OLPT could get the US through its present crisis, it is not really desirable as a long term system

    In New Nation Movement v President of the Republic the constitutional court held:

    [53] We cannot dismissively say if you stand for political office through a political party, it makes no difference; you still do stand for political office. It may make all the difference to some. And it does to Ms Revell. She explains that, as a representative and leader of the Korana nation, a section of the Khoi and San people, she is averse to forming or joining a political party. Hers is not a for-the-sake-of-it objection. I understand her point perfectly. I read it to mean that, as a leader of a nation, she does not want to be constrained by that kind of partisanship that comes with being a member of a political party. That partisanship makes you ultimately answerable to the party. Being free of those shackles will make Ms Revell directly answerable to her nation, not to a political party. That is the choice she is making. In my book, it is a valid choice. Surely, her example is not isolated. There must be many and varied other examples. Subject to the Lavigne threshold, we cannot make light of them.

    [54] This must not be taken to mean the state is entitled to ride roughshod over associational choices that are not sound. Even if not well founded, choices by an individual may well define her or him. Unless the state can justify interference, even such choices are deserving of protection under section 18. Indeed, I read Woolman to make a similar point.[77] Many of the life practices we follow have nothing to do with well thought-out choices. Some are “arational” choices dictated by the world we live in.[78] One may be a member of a certain religion not because she or he has made a conscious decision in that regard, but purely because that is the religion she or he grew up in. Likewise, one may follow certain cultural practices and thereby commune with certain groups of people in furtherance of their culture. And that may be for no reason other than that that’s what she or he was socialised into. So, some of our associational relations may not have been chosen. “They just are.”[79] But they may be deserving of protection.


    [58] In sum, choosing to associate is an exercise of the right to freedom of association. Choosing to dissociate from that which you earlier associated with is also an exercise of that right. Choosing not to associate at all too is an exercise of the right. A restraint on any of these choices is a negation of the right.

    [59] It is axiomatic then that if the state compels an individual to associate when she or he does not want to, that limits the right to freedom of association. That must mean the reading of section 19(3)(b) contended for by the respondents results in a denial of the right to freedom of association.

    The constitution of South Africa is quite friendly to political parties. The constitution of the United States is not. It is quite hard not to see the US supreme court adopting analogous reasoning about freedom off association when considering longterm use of OLPR. A sunset clause for say 10 years could avoid that.


    • I may not have expressed myself well. I do not discount the differences between CLPR as used in South Africa and OLPR as proposed by MSS. However OLPR has many, but not all, the same freedom of association difficulties as CLPR.


      • OLPR can get around freedom of association difficulties very easily if it allows a “list” of one candidate,. Or even directly allows individuals to be elected as individuals, whether part of a list or not. If the highest D’Hondt average falls to a candidate without a list, she gets a seat. If she gets enough votes for two seats, she would, not having a running mate, miss out on that seat and it would go to the next deserving party or independent candidate.

        A Sunset Clause though is an invitation to chaos. Who can honestly say they won’t see at least a faction of conservatives ready to pounce as soon as the provisions sunset, force the return of single member districts to gerrymander themselves a majority in the House. This is the party after all that will frequently admit that the Electoral College is the only way that they can win the White House, as if the idea of putting up a slightly more appealing candidate is beyond them.

        Liked by 1 person

    • The problem with a sunset clause is what happens then: Back to the status quo ante?

      I think having a mandatory review is not a bad idea. Maybe we are saying the same thing. But I understand “sunset clause” as the law expires. If you have no election law in force, how do you run the next election? Seems rather unwise to me.


      • A sunset clause would not automatically restore FPTP unless it was drafted by an incompetent. I’ve addressed this in a relpy to mark Roth below. I’d also like to adopt jd’s point that the law should have a second chapter providing for what would happen were OLPR (as I think a distinct possibility) struck down by the courts on analogous reasoning to New Nation Movement v President of the Republic.

        I am nervous of easy constitutional solutions. Australia would almost certainly be a republic today had not the Australian Republican Movement decided the easy road to a republic was altering the title of the governor-general and otherwise making minimal changes, despite a large majority of republic advocates wanting an elected president on the Irish model.


    • I think a review clause would be better than a sunset clause. “After 10 years, a Citizens’ Assembly will review this voting method, and will have the power to propose a new one. If over 2/3 of the CA supports a new method but the legislature does not implement it, it will go to a referendum, and be implemented if it passes.”

      (I know, Congress could later pass another law to prevent the referendum or even to ignore its outcome. But if they just dodge or deadlock on the issue, the referendum would happen.)


  9. I am concerned that the anti-democratic win of the Republican Party is far larger than just the Trumpist sector, The Republican Party seems to be essentially an anti-democratic vehicle of representing geography over population.They defend the Electoral College as a “necessity,” they accuse the Democrats of “gerrymandering” by seeking to admit Washington as a state, and they label basic voting rights protections as “partisan.” If this was put before Congress, McConnell would have a great deal of support in labeling and filibustering this as a Democratic gerrymander or some such nonse. This is the party that calls gerrymandering the Electoral College to allow Republican wins in Democratic states “proportional representation” after all.

    On a technical note, if states are to be divided into districts of variable size, what is to stop the GOP from gerrymandering or even tullymandering the districts?


    • A sunset clause would not necessarily mean reversion to FPTP, any more than the constitutional court of South Africa ordered parliament to adopt a new electoral system or revert to FPTP. Under any form of PR it is unlikely that the Republicans, who appear to be in some danger of facing a Trumpist party on their right flank and a conservative party on their left flank, mould be in a position to dictate FPTP after 10 years of OLPR.

      I admit to a certain guilty pleasure thinking about how the 2022 election would look under FPTP if the right divided into 3 parties. I will mention it at my next confession.


      • A GOP facing challenges on the left and right would, in my view be MORE likely to want to go back to single member districts. A return to single member districts would mean the return of primaries, most likely, and not a British style five party race for each seat, Once the primary winnows things down to a single right wing candidate, the GOP might decide to hedge its bets and win a majority with a coalition that includes a few far right wingers.

        Of course my views are very much clouded by my conclusion that the United States doesn’t use FPTP at the moment. It uses a unique two round system where all viable left wing candidates are corralled into the Democratic Party and the Democratic Primary and all right wing candidates into the GOP and its primary. If the Trumpists do break off the from the GOP in the next year or so, my theory might shatter. But I suspect that they will simply launch primary challenges and then, generally, still vote for the winner of the GOP primary even if it is not their guy.

        Liked by 1 person

      • My point was not that the Republicans might not want to revert to FPTP. It was that the emergence of 2 or 3 parties of the right would mean that Republicans would be unable to revert to FPTP.


        • Alan, rereading this very rich comment thread almost a year and a half later (with the emergency still very much ongoing, as today’s hearings of the Jan. 6 committee remind us), I am still unclear on something. What is the sunset clause you are proposing? What is its reversion point?

          I think it is a bad idea, but given that I’ve learned over the years that you have many good ideas, surely I am simply misunderstanding you.


  10. What do you perceive to be the implications of OLPR, in the form you have proposed, for gender and minority representation? If they would/could be adverse, what would be the risk of court challenges?

    Note also that if individual candidates can run as a list, that would open the way for a largest remainder formula to break down, giving rise to a need for recursive formula to be used instead. This was a significant issue in the Constituent Assembly in Timor-Leste in 2001, where one possible candidate had the potential to get a majority of the vote in his own right – though ultimately he didn’t run. See section 37 of the relevant regulation, here: https://peacekeeping.un.org/mission/past/etimor/untaetR/reg20012.pdf. This problem doesn’t arise if a divisor formula is used.


    • This is a good question for the law professors. My sense is that the nomination districts might provide the sort of majority-minority/minority-influence representation we now get in some SSDs. Or the parties could mandate quotas.


      • I can’t imagine how OLPR with nominating districts could be worse for representation of women and racial and other minorities than the status quo. If deemed necessary, the system could certainly have minority-majority nominating districts to ensure (or make more likely) that members of the minority group advance to the general election.

        Going a step further, one could mandate quotas somehow. As Jack says, that’s a question for legal experts, not me.


    • The nomination district idea would work reasonably well with this, since AIUI the Voting Rights Act requires at present the drawing of majority-minority districts which, while ensuring minority representation, also pack minority (generally Democrat) voters into districts. Using these majority-minority seats to nominate candidates would mean minority voters would be able to use these districts to ensure a candidate of their choice made the ballot without any such concerns about diminishing the real effectiveness of their vote.


  11. With single-member districts and 2-party dominance, an effective gerrymander makes almost every seat into a safe seat for one or the other party. The hard fight is to secure the nomination as the party’s candidate, not the general election. A party-based primary system further drives polarization, squeezing out “free-market, small-d democrats”.

    Starting from a condition of 2-party dominance, OLPR with “nomination districts” would still put the squeeze on moderate candidates at the primary stage. There’s no chance to appeal to moderates of the other party if you don’t get past the primary.

    Primary candidates would have different incentives if primaries were M-seats wide. Candidates could declare themselves using any label — registered party or other. The top “c M “ vote-getters of a jungle primary would be the candidates in the general election, subject to a limit M of the candidates who campaigned under a given label.

    Your strategy in a primary can be much different if you aim to be one of the top 10 candidates than to be one of 2 candidates (for c=2 and M=5, say).


    • This is an important question, thanks for bringing it up. Personally, I favor starting with single-district primaries with a “soft c” of 2-4 (via a vote threshold), but in the longer term, leaving room for some experimentation at the state-by-state level.

      Partisan primaries are a key part of the broken polarizing incentives of today’s politics. If, under OLPR, there were single-seat-district primaries with c=2, those polarizing incentives would mostly continue. Clearly, we should aim for something better than that.

      But I fear that going to multi-seat-district primaries with, say, mC = 10 might be too far in the other direction. This would really weaken parties by a lot, even compared to their already-relatively-weak position in the US. Candidates might be incentivized to campaign — and then legislate — on narrow issues and charisma, not broad platforms. You might end up with a hyper-fractured party landscape, a la Israel when its thresholds were even lower.

      Note that this is not just arguably bad intrinsically; it’s also pretty undeniably bad from a perspective of the political viability of the reform. Most current incumbents do not want to see their currently-well-established party fracture into shards.

      Yes, the OLPR round would still have an effective quota of around 16%, so the specter of Israel may be an exaggeration here, but I think my point is clear.

      I think that a good place to start would be single-seat-district primaries with a threshold of, say, 20%. That is to say: the top two candidates, plus any candidates with over 20% of the primary vote, would progress to the OLPR round. To keep the threshold meaningful even in case of lower-turnout primaries, you could say that if turnout was under half as high as the previous general election of the same type, the threshold would instead be 10% of the previous general election. This would mean that in practice, c might vary between 2-4, depending on the district. I think this threshold is low enough to leave “room in the middle” when partisan candidates become too polarized, but high enough to still encourage big-tent parties and campaigns.


  12. Pingback: Cosa succede dopo un tentativo di colpo di stato? – Ultime notizie – Leggi il mondo oggi. – Notez.eu

  13. Pingback: What happens after a coup attempt? - That One Media

  14. Pingback: Coup attempts usually usher in long stretches of democratic decline, data shows – FREEMEDIA

  15. Pingback: Was passiert nach einem Putschversuch? – Letzte Neuigkeiten – Lesen Sie die Welt heute. – Giveawaybots.com

  16. Pingback: Analysis | Coup attempts usually usher in long stretches of democratic decline, data shows - Pehal News

  17. We need an end to Winner take all electoral votes from each state. Winning by congressional district would bring the electoral college much closer to election by popular vote.
    Most Americans don’t understand the function of the electoral college, rail at the lack of a plurality election, none recognizing that the title president of the United States means president of a state by state country not president of the American people.


    • President of the United States means president of the state called “the United States of America”. It does not say anything about how this officer is chosen. If anything, the title suggests appointment by Congress, which would indeed be the best.

      If one does buy the idea that the title should say something about the means of election, election by Congressional district would make it the President of the Swing Districts of America, just as it now is the President of the Swing States of America.


      • Isn’t there a constitutional right for the minority party to gerrymander its way to the White House and false congressional majorities? The “republic,not a democracy” crowd and “the presidents of the states not the people” crowd tell me this all the time.

        Why not just give each county one vote while we’re at it.

        On a more serious note: The only claim of the ant-democratic crowd, regarding presidential elections, that doesn’t just sound like naked racism is their claim that the president should have a national coalition behind him. Now, to me at least, there is nothing about a minority coalition with more dirt under it that makes it more legitimate than an actual majority of the voters voting for the first place finisher. But an OLPR House can check the presidency by making it much easier for a nationwide coalition, that earned an actual majority of the votes of real people, to have a House majority.

        I would get along with a single amendment that enshrines some form of actual PR in the House and provides that the House elects the president. We don’t need a nationwide direct election if a third of the country hates the idea but the voting majority still gets it proportionate say.

        Liked by 1 person

  18. Pingback: A modest and timely proposal – voteguy.com

  19. Sometimes it is better to take 70 per cent of what you want on the grounds that 70 per cent is better than zero per cent. At other times, it is better not to take 70 per cent because taking the 70 per cent will delay for a long time or stop forever the achievement of the 100 per cent.

    Th single transferable vote is a much better system than list PR systems because the voter can rank all candidates in any order at all, and no one will be elected until a quota is reached or so many preferences have exhausted that there are no full quotas left to get.

    Whatever system is adopted, it needs the end of gerrymandering. Even multi-member seats can have their boundaries drawn to increase one party’s chances of winning a majority of seats. I understand the US Constitution provides that Congress can make laws regarding elections. It can surely set up an American Electoral Commission to draw boundaries and run elections.


    • Yes, if there were actual political will to adopt my proposal–or any PR system–there surely would also be a majority for requiring proper boundary commissions.

      The question of whether STV or any type of list PR is “better” is very much a contested point, and also beside the point of this post for reasons I tried to articulate within it. The important questions are, which one can be adopted most readily and, which one would be less vulnerable to being overturned in the future? I am pretty sure the answer to both questions it not STV.


  20. Pingback: Speaking of Reform

  21. “… The most basic obstacle facing right populism has been around for decades: the people who matter on the right would rather get filthy rich with 45 percent support than slightly less filthy rich with 55 percent support, and the configuration of American political institutions makes this a perfectly rational strategy. The way to change this calculus is not to convince them of their errors but to render the strategy unviable. That would require a democratization of American political life so that the pursuit of majority support becomes a necessity rather than a luxury.”
    – Daniel Luban, “The Not-So-Strange Death of Right Populism: A string of pseudo-populist conservative movements have reverted to the same agenda of tax cuts and deregulation. Why should we expect anything different?” Dissent Magazine (Winter 2021), https://www.dissentmagazine.org/article/the-not-so-strange-death-of-right-populism

    Liked by 1 person

  22. Apropos of the discussion in the original post about the need for at least four parties in the U.S. (and, admittedly, a diversion from the current discussion of Australia), I found this from law professor G. Michael Parsons, “[S]o long as our system continues to make minoritarian capture of government possible, Republicans will continue being drawn towards the antidemocratic wing of their party.” I haven’t seen that point made as succinctly anywhere else. See


    The whole article is worth reading. It is mostly about gerrymandering, but advocates for PR near the end.

    Liked by 1 person

    • Yes, that quote is a good succinct summary of the emergency.

      Thanks for the Harvard Law citation. Yes, he makes a strong case for PR there. If I were his editor, I would have asked him to say up top that he was going to address that in the piece, because it is so important. In any event, it is a terrific essay that draws important conclusions.


  23. Pingback: A Workable Proposal for Proportional Representation in Rhode Island’s Elections – Rhode Island Liberator

  24. Pingback: Ballot exhaustion, STV edition – voteguy.com

  25. Pingback: Wie man Demokratie repariert: Brechen Sie die demokratische und die republikanische Partei auf, sagen Experten – Letzte Neuigkeiten – Lesen Sie die Welt heute. – Byhil.com

  26. Pingback: Move beyond two-party system to fix American democracy–WaPo | Fruits and Votes

  27. Pingback: How to fix democracy: Move beyond the two-party system, experts say - That One Media

  28. Pingback: Analysis | How to fix democracy: Move beyond the two-party system, experts say - Pehal News

  29. Pingback: How to fix democracy: Move beyond the two-party system, experts say - Worlds's News Site

  30. It seems that discussion of the US electoral system often omits some basic democratic ideas.
    Some important ideas:
    1. DC and PR (and other territories) need voting representation.
    2. The power of the Senate needs to be be reduced ( or follow NZ 70 years ago and abolish it). Few countries have an all-powerful senate like USA.
    3. The constitution is far from democratic and is there no democratic way of changing it.

    It is certainly a challenge to design a PR system with territories of such variable size. Perhaps territories should be grouped to have shared representatives. Or representative should have weighted voting power like shareholders.


    • Simple PR-
      total votes / total members = equal votes to elect each member.

      Pre-election candidate rank order lists of all other candidates.
      surplus votes down.
      Lowes loser votes up.
      ALL votes count.


  31. Pingback: Warnings from the past – voteguy.com

  32. Pingback: Electoral Engineering and the Freedom to Vote | taktik(z) GDI (Government Defense Infrastructure)

  33. Pingback: What is the one-vote system? A Q&A with Jack Santucci – An Alabama News and Information Site

  34. Pingback: “Local PR” | Fruits and Votes

  35. Pingback: Reflections at the end of a wild year – voteguy.com

  36. approval voting is vastly simpler and more politically viable, and would make congress a body of nearly 100% perfect consensus/moderate candidates, rather than warring extremists. i think it would be far more politically viable and beneficial than proportional representation.


  37. A year and a half later, I re-read this thread and Jack Santucci’s related thread at VoteGuy. I understand the broad strokes of the idea of “nomination districts” (candidates get selected in districts the size of the current SMDs but then run on slates larger multi-seat districts). But, even after reading everything twice, I still find the mechanics hard to follow. I have no doubt that this proposal is easier for election officials to implement. But, damn, in comparison with it STV look easy for voters to understand.


    • I’d be happy to have the OLPR proposal ditch nominating districts and a first round/primary. But I do not think that can be sold. And primaries (qualifying rounds, whatever) with M>1 are problematic for all sorts of reasons. So the notion of having nominations look more or less like they are now (single-seat sub-districts) seemed like a good solution. I am open (so to speak!) to others.

      I don’t think it is complicated, actually. Top ‘c’ in each single-seat nominating district qualify for the final round, and they get to decide which candidates from the wider general-election they ally with on a list (and they can run independently if they want, though their chances of winning that way are probably low). In both rounds, voter votes for one candidate. Ballot clearly shows who each candidate’s listmates are. A vote for any one may help one or more of the allies get elected (because it is a list system).


Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s

This site uses Akismet to reduce spam. Learn how your comment data is processed.