Ranked choice voting at the county level in California: Santa Clara’s “unique circumstances”

Posted on by

This post is a follow-on to the previous one about a low-turnout runoff in Santa Clara County, California. Could the county change the election of its officers to ranked-choice voting (via the alternative vote a.k.a. instant runoff)? Yes, it could. However, it is the only county that currently has this authority. Further, it appears it could switch to STV for its Board of Supervisors.

AB 1227 of 2023 was passed by the state legislature and signed by the governor and explicitly grants to Santa Clara County the option to adopt ranked-choice voting (RCV) for any or all elected county officers. It allows that either the county’s Board of Supervisors or a county initiative may enact RCV.

The legislation further says, “The ordinance [to adopt RCV] shall specify which county officers shall be elected by this method and whether they shall be elected at large or by or from district, as applicable.” The reference to “at large” surely enables the country to elect the Board or other multi-person bodies in the county by single transferable vote (STV), sometimes called (in the US) proportional ranked-choice voting (PRCV). The first section of the enacted state law–the “findings” section–references cities in California that have adopted STV, such as Albany and Palm Desert, although it simply refers to their systems as “RCV” without the important detail. (Palm Desert now has two districts, one of which is a single seat using RCV/alternative vote, and the other of which elects four by STV/PRCV.)

Interestingly, the law applies only to Santa Clara County. Moreover, its third of three sections reads, in full: “The Legislature finds and declares that a special statute is necessary and that a general statute cannot be made applicable within the meaning of Section 16 of Article IV of the California Constitution because of the unique circumstances relating to the County of Santa Clara’s interest in having the option to use ranked choice voting in its elections.” (The cited section states, “A local or special statute is invalid in any case if a general statute can be made applicable.”)

Therefore, we might say that we have here a case “electoral system pork barrel”! Okay it is not a project, which is what “pork barrel” implies, but a measure regarding electoral systems that provides a benefit only to the geographical region represented by its sponsors. The originally introduced bill did not have that third section, although it did already state its applicability was only to the one county. The introducing member was Assembly Member Evan Low, who at the time represented Assembly District 26, encompassing parts of the Silicon Valley, which is to say a large portion of Santa Clara County. Assembly Member Alex Lee later joined as a co-author; his district encompasses other parts of the county, including part of the city of San Jose.

Curiously, the final chaptered amendment to the state code says that an election under RCV in Santa Clara would have to take place at the same time as the statewide “primary” (i.e., the first round under California’s two-round not-a-primary electoral system used for state and federal offices). It is very strange that it would not be held at the November “general” election. This provision changed through the bill’s legislative process, and I will address the apparent reasons for this below. The initial version did not specify the timing of a potential RCV election in the county. The first amended version in the state Assembly added this tortured phrase: “If a proposal would elect an officer by ranked choice voting through a single election with no possibility of a runoff, the bill would authorize the county to hold that election at the first statewide general election following the statewide primary election at which the election otherwise would be held in accordance with existing law.” An amended version in the state Senate said in somewhat more straightforward language that the county could hold an RCV election at either the “primary” or the “general” while adding that third section quoted above.

A further round of amendments in the Senate saw two members of that chamber join as further coauthors, Josh Becker and David Cortese, both of whom represent–you guessed it–parts of Santa Clara County. However, before its final passage the language to require that any such RCV election in the county be held at the “primary” was included.

The bill’s Assembly Floor Analysis dated 9/05/2023 (available at LegInfo) indicates that the jurisdictions within the state that have so far adopted RCV are all charter cities, but that it is not clear under current law that charter counties would have this authority, because counties are generally given narrower discretion in their election ordinances.

The county itself had determined that it did not have the authority absent a change in state law. Thus what we have here is a change in state law for this one county for the simple reason that only this county had applied for authorization. It did so in response to a countrywide ballot measure that had indicated the county would adopt RCV if it became possible within the capacity of the vote tabulation machinery used by the county. That measure passed in 1998, but only in December 2022 did the county Registrar of Voters declare that it now had the relevant machinery.

It is not clear from the immediately available resources whether there was any debate on making this legislation a blanket permission for counties to adopt RCV. All versions of the bill as it worked its way through the legislature explicitly were only about Santa Clara County.

As for the question of the date on which an RCV election could take place, it seems this is due to the potential incompatibility with state law if it were set to take place in November. This is implied by the report on the Assembly Third Reading, dated April 24, 2023 (also available at LegInfo). It notes that elections for county officers are required to be held at the statewide “primary” with a second round in November only if there was no majority in the first round. Therefore, a county holding no election for such an office at the first round would be in potential breach of state law, and evidently legal counsel said that even the third section added to the bill would not remedy that problem. This is unfortunate, of course. RCV (in the form of the alternative vote/AV) is by definition a one-round solution to obtaining a definitive result that might otherwise require two rounds to obtain a majority. So logically it should be held at the end of what would otherwise be a two-round process, not at the beginning. For sure turnout is higher in November than at the first round. Not for the first time, logic gets tripped up by the law!

The bill ultimately passed the Assembly 67-0 with 13 abstentions and the Senate 31-7 with 2 abstentions. I do not know if there is currently a strong prospect that the county will go ahead and be a pioneer in using RCV for any of its county offices, but possibly that low-turnout Assessor runoff special election will reenergize the debate.

A very low turnout runoff: Santa Clara County Assessor 2025

Posted on by

On 30 December, voters in Santa Clara County1 voted in a runoff special election for Assessor. Well, some of them did. Turnout was just over 20%. There could be many factors driving such low turnout. The election was on the 30th of December, and most folks were not thinking much about local politics. The office is an Assessor. What does an Assessor do, anyway? (I know, they assess–all sorts of stuff.) The result might have seemed like a foregone conclusion in the first round when the leader (and eventual winner) led by a margin of about 13.7 percentage points. Probably some combination of all of the above. The special election was made necessary by the resignation of the 30-year incumbent (who is 84 years old).

The contest is more interesting than your average special election for a county assessor position. The first round was concurrent with a reasonably high-profile statewide election. Even though that statewide election was itself a special election, it had high stakes. I am referring to Prop. 50, the state’s engagement in the interstate gerrymandering war. 550,315 voters in Santa Clara County turned out for the proposition (393,625 voted yes), while 467,964 voted in the Assessor first round. That is a difference of over 82,00 voters, but as you might image, that nearly 468k would vote for Assessor is a pretty high amount.2 The total turnout was 553,637 (in other words, there actually was a slice of the electorate that voted for Assessor but skipped Prop. 50), which was 51.8% of the registered voters in the county.

Advocates of ranked-choice voting will often point to a contest like this one, alleging that runoffs result in low turnout yet the considerable expense of running an election. On the other hand, the election could be complete in a single round when the turnout might tend to be higher. In fact, it was from just such an advocacy organization that I learned about this runoff. So they have a point. A drop-off in turnout from 51% to 20% is substantial.

If there is resistance to adopting ranked-choice voting (the alternative vote, or AV), as there often is, there are other possibilities. For example, one could elect the office by plurality, at least when it is a special election for a lower-profile office like Assessor. Maybe it is not so bad if such a contest produces a winner with less than 50% of the vote.

In the first round, Neysa Fligor (currently Los Altos Vice Mayor and an employee of the Assessor’s office) had the plurality, which was 37.7%. The runner up, Rishi Kumar (a former city council member for Saratoga), had 24.0%, and two other candidates had 21.3% and 17.0%.3 Yet again, another two-round contest with a closer margin between second and third than between first and second, something that is rather common and potentially a flaw in the two-round system. Maybe the third place candidate could have won the runoff if only he or she had won just a relative few more votes needed to have finished second. In this case, however, that was probably unlikely, as a first-round contender who is 13.7 percentage points ahead of the closest challenger is likely quite hard to beat. In fact, another electoral rule possibility aside from plurality or two-round majority (or AV) is the double complement rule (DCR), first proposed by Shugart and Taagepera (1994). Under the DCR, the first-round leader wins outright if the runner-up has a shortfall from majority that is at least double that of the leader. In this contest, (50–24.0)>2(50–37.7) because 26> 2(12.3). Perhaps 37.7% does not seem like a “decisive” result, but a leader with that strong a lead has a reasonably high probability of being the majority’s preference. And indeed the runoff showed Fligor trouncing Kumar, 65.18–34.81, although we must be mindful that this was in the sharply reduced turnout as already noted. Maybe if turnout had reached 50% or higher in the head-to-head, Kumar would have won. I have to say I find that completely implausible, and most likely the large majority of voters–even those who cast a vote in this first-round contest–either did not care all that much or figured it was a foregone conclusion that Fligor would win.

Perhaps the best reform of all would be to ensure that a runoff is not held right in the middle of the winter holiday season. Another wild idea is that an Assessor really does not need to be an elected position at all, and needs to be one even less when it is an interim replacement for an incumbent who had resigned. In this case, the interim in which the winner of the special election will serve is only about a year, as a regular election will be held in the 2026 cycle.

Or the county could adopt RCV. It has been considered in Santa Clara County before. What do you think? Most of you probably care a good deal less who the Santa Clara County Assessor is than the average Santa Clara County voter did. But the principles of how to ensure effective and accountable government at the local level apply far more widely than this one contest. And a post decided in a runoff with 20% voting hardly seems like good government.

__________

  1. The county includes the city of San Jose. ↩︎
  2. In fact, the last time it was on the ballot was at the first round (“primary”) of 2022, when the incumbent Larry Stone was reelected yet again with evidently token opposition. He got just over two thirds of the vote, so no election was needed that November. In that contest, 302,038 votes were cast. This is obviously a good deal lower than in the first round of the recent special election, but these are not directly comparable given Stone’s easy win vs. an open seat in 2025, in addition to the concurrence of the first round in 2025 with the high-profile statewide referendum. By comparison, the statewide governor first round in 2022 saw 348,647 votes cast in the county. (In November 2022, that would rise to 541,895.) ↩︎
  3. These were current Saratoga Council Member Yan Zhao and East Side Union High School District Board Member Bryan Do. See San Jose Spotlight. ↩︎

Recognition: Somaliland edition

Posted on by

In the past week, the government of Israel became the first to formally recognize Somaliland as a state. This is a good development, and I hope there will be more such declarations forthcoming. Somaliland has existed as a de facto independent entity, very much state-like, since 1991. It even has held regular multiparty elections for over two decades.1 In fact, it uses open-list PR, which was the reason for this blog’s one previous entry on the topic of Somaliland.

It should not properly be seen as a secession from Somalia but rather a reassertion of past legitimate statehood which is now being recognized (albeit by only one other state so far). Somaliland, with its capital Hargeisa, was a British territory that earned its independence on 26 June 1960. Days later it joined the former Italian Trust Territory of Somaliland in a union known as Somalia, with its capital in Mogadishu. The case for Somaliland (Hargeisa) as a rightful separate state today (and de facto over the past three decades) rests on these distinct colonial pasts,2 their separate independence processes, and a claim that the act of union between the two entities was never properly ratified. In fact, the referendum on union was actually rejected by a majority of voters in Somaliland, who were outvoted by the majority in the larger Somalia. (See timeline at Borkena (an Ethiopian publication) or Wikipedia.)

The standard requirements for recognizing this entity as a state would seem to be met, despite the obvious and predictable objections of the government of Somalia. These requirements, based on the Montevideo Convention (see Justia), are generally that the entity in question have “a permanent population and a defined territory,”3 and an “effective government.” Of course, we could say that plenty of entities around the world with existing recognition as states fail one or more of these tests. Including Somalia itself! Sates are rarely de-recognized for falling out of compliance with such conditions and some have gained recognition despite not ever having met them. Yet when an entity clearly passes these tests, as it would seem Somaliland does, it is reasonable to conclude that it merits recognition. Its case seems at least as strong as that of Kosovo, which many (but by no means all) states in the international system recognize, and clearly stronger than that of Palestine (which more have recognized than have recognized Kosovo, and to which I will return below).

It is possible that other states will follow suit and recognize Somaliland in the near future, notwithstanding the many statements of condemnation that immediately (and perhaps predictably) followed Israel’s announcement. Taiwan was an exception, with its Ministry of Foreign Affairs issuing a statement that it “welcomes” Israel’s decision, adding that Taiwan, Israel, and Somaliland are “like-minded democratic partners sharing the values of democracy, freedom, and rule of law.”

Taiwan is an interesting case here, as it surely meets every standard of being a “state” worthy of recognition, and yet it enjoys actual formal recognition from few others. That is, of course, due to the demand of China (Beijing) that no state can have formal relations with both, and the fact that most of the countries of the world want formal relations with the People’s Republic. But Taiwan is in any reasonable sense–at least beyond the world of diplomatic “recognition”–a functioning state, and indeed a democracy.

There is some consternation that the recognition upsets regional balances or the complex internal situation in Somalia. These concerns are surely overstated. Somalia has not controlled Somaliland, as noted, for nearly 35 years. Even the African Union, which normally has strong norms against separation,4 had a Fact Finding Mission in 2005 that Somaliland had undertaken an effective process of state-building, had defined territory, and a constitution that had emerged from grassroots consultation and a proper ratification. It concluded that the AU should find a “special method” of dealing with the situation and that “Somaliland’s search for recognition was historically unique and self-justified in African history,” noting the facts I sketched above about the improperly ratified and malfunctioning union that existed prior to 1991.5

It is also questionable that Israel’s recognition of Somaliland will further destabilize Somalia, as claimed by Lesley Anne Warner in World Politics Review. Somalia is already terribly unstable, with nothing that can be called “effective government” and as Warner notes, there are already internal tensions between the “federation”6 of Somalia and its member units over planned constitutional reforms. I doubt Israeli recognition is so powerful that it can make the internal situation of Somalia measurably worse. In any case, is it worth denying recognition to Somaliland, whose leaders and people have been carrying out a process of quite successful state-building for over three decades, to stoke fantasies of an ineffective leadership in Somalia that it will regain territory over which its past control was already of questionable legitimacy?7

Speaking of stoking fantasies, it is noteworthy that this development came in late 2025, a year when several western democratic governments announced “recognition” (or intent for near-term recognition) of a State of Palestine, despite that entity almost certainly failing any reasonable application of the test of control over “defined territory” or an “effective government.”8 Recognition9 is, of course, a political act. It is one government developing a formal relationship with another government, and actions of governments are political more or less by definition. Nonetheless, the various Palestine recognitions were blatantly political acts, mostly motivated either by domestic politics or by a desire to punish and compel Israel, rather than recognitions of actual conditions for statehood having been met. By the latter standard, Somaliland is the far more deserving of the two. I congratulate Somalilanders on achieving this step, and hope many more states will eventually recognize what they have established.

[Note: I turned off comments, not because I don’t welcome calm and rational discussion, but because–perhaps predictably–this attracted a comment that was rude rather than discussion-enhancing, and thus in violation of the comment policy. It was from someone who, to my knowledge, has never commented on this blog before and will not be welcome to do so in the future.]

_________

  1. National elections are restricted to only three parties, with a periodic process of (re-)qualification. They seem to be generally regarded as quite fair, even though most sources would not regard the country/territory as fully democratic. ↩︎
  2. A brief exchange in the UK House of Commons from 25 July 1960 is interesting. ↩︎
  3. There is a portion of Somaliland territory that is disputed and not fully under the central government’s control but Justia notes that under the Montevideo Convention, “boundary disputes do not necessarily prevent an entity from being considered a state.” ↩︎
  4. Not too surprisingly, given it is a union of… states. Yet it has found ways to recognize Eritrea and South Sudan, though these are clearly different in the sense that the larger unions that broke up consented to the recognition (after lengthy periods of conflict). ↩︎
  5. The text of the report was posted as a series of screenshots on Twitter (X) by Rashid Abdi. It is well worth a read. ↩︎
  6. I doubt it qualifies as a case of federalism in a meaningful sense, but that would be a topic for another time. ↩︎
  7. It not worth seriously entertaining the claims from some quarters that the recognition is a precursor to Israeli military bases in the Horn of Africa or to a plan to resettle Gazans there. I find these allegations patently ridiculous (why would Somaliland agree just to get Israeli recognition?). In any case, if they were true I would not know, and my assessments here of the merits of Somaliland’s justification for statehood would not change in any case. ↩︎
  8. We might say Palestine has had two concurrent governments, neither of them effective. The bigger point is that the Palestinian Authority, which serves as a de facto government of Palestine, has not controlled the Gaza Strip, allegedly part of its territory, in nearly two decades, and under the Oslo Accords has actual control over only portions of the rest of its presumed territory. Normally the “state” should pass these fundamental tests before it is recognized, and the reasons why the conditions are not met should not be relevant to the determination of whether the state exists. ↩︎
  9. Some of them said their recognition was “conditional” on various future developments like internal reform of the Palestinian Authority or disarmament of Hamas (by whom?). On the legal standing of conditional recognition and its problems, see Stanley-Ryan. ↩︎

‘RCV’ in US presidential primaries?

Posted on by

An editorial in MSN, published in late November by Ryan Teague Beckwith, promotes the use of so-called ranked choice voting (RCV) in presidential primaries for the US, specifically the Democratic Party. This is not a new idea, of course. And the piece is frustrating for its lack of specificity about what is being proposed. This lack of clarity explains my quotation marks in the title and “so-called” in the first sentence. To assess any such proposal, it would be crucial to know not only how participants would be asked to vote, but also how the proposed change would affect the allocation of delegates (assuming those remained) and ultimately a nominee. Yes, this gets back to my periodic frustration over the common focus on “voting systems” rather than “electoral systems,” but I will leave that particular pet peeve aside (for now, and mostly).

Beckwith observes that there are “simply too many candidates,” although does not attempt to explain how many is too many. Then he claims that winnowing of the field takes too long (“drags out the process”) when a party uses “an elaborate system of awarding delegates proportionally,” as the Democratic Party currently does (sort of–I will get back to that). Enter RCV as a supposed solution. The idea, of course, is that voters would rank order their preferences over multiple candidates in the field. Then the process of elimination of the trailing candidates and transferring of their votes would go on “until one candidate secures a majority.” In other words, the electoral system he is advocating is the one political scientists would typically call the alternative vote (and is also sometimes called instant runoff voting or IRV).

Perhaps readers already will have spotted the problem here. Although he never makes the implications of favoring RCV clear, evidently Beckwith is proposing to narrow the field down in each state to a single majority winner. That is, unless he actually means to do away with delegates and the state-by-state sequential contests altogether, and use ‘RCV’ in a single-day nationwide primary. That would be a very radical change,1 and there is no hint in the op-ed that this is what is being proposed. Thus I assume the idea is to keep with something like the current schedule of states or groups of states voting on various different days throughout the “season” of sequential primary contests.2

I have to ask, how would it be an advance if each state awarded all of its delegates to one candidate in a process that starts off with numerous candidates–in fact “too many”? Further, how would it be an advance if the process continued to “drag out” as states late in the process vote when a majority of delegates may already have been allocated to one candidate earlier, perhaps to a presumptive nominee who started accumulating state-by-state delegate majorities on a third of the first-preference votes (hypothetically)?

Most of the rest of the piece simply rehashes the longstanding claims of RCV advocates that voters can be entirely sincere under this sort of system (always vote for your first choice) and that candidates will “play nice” because they want second (and lower) preferences from voters who ranked other candidates higher. These claims can be questioned, of course. In a field with “too many” candidates, not only is it likely much harder for voters to identify their true first choice, let alone rank a large field, but they may also find that candidates they thought they were transferring their votes to got eliminated earlier in the count. If they rank all candidates, this is not likely not a problem (the vote should still transfer to some candidate who remains in the running). But if there are a lot of candidates, most voters will not do so,3 and in any case some states may truncate the number of allowed preferences short of the number in the entire field.4

As for the claim about candidates “playing nice,” this has always struck me as wishful thinking. I concede that any given candidate dyad in a multi-candidate field under a system of ranked preferences has a higher potential to be cooperative than under plurality and maybe also relative to two-round majority. However, ultimately these processes all must produce a single winner, there are often intense policy or personal disputes amongst the contenders, and we are talking about winnowing a field down to one of the final contenders for the Presidency of the United States, after all. In any sensible theory of electoral system effects on candidate strategy, the current “proportional” system should be far more conducive to encouraging candidates to “play nice” and yet… they do not.5 To claim that RCV will be notably more conducive to playing nice is quite a stretch.

The current process is a mess and needs reform. We hear various proposals for reform every cycle, and obviously this means that the party continues to fail to come up with a process that is seen as working well. The current “proportional” system actually is not so proportional. It consists of high thresholds and variable and often low district magnitude, with poor information about who is actually in the running for delegates in the voters’ own congressional district. In a multi-candidate field, your first choice may have no chance to get delegates. It is, as I said back in 2020, the strategic voters’ nightmare. While RCV may seem like an obvious solution to difficulties of casting a proper strategic vote (let alone allegedly freeing voters to cast a sincere one), we should be very cautious about claims that such a proposal would either render strategic voting unnecessary (which is hogwash) or simpler for voters.6 The solution to this problem is surely not to have every state award all its delegates to one candidate,7 regardless of the method used to determine the majority winner in each state. The alternative vote form of ‘RCV’ in presidential primaries is a poorly thought out idea, even if well intentioned. It deserves to be dismissed.

_______

  1. Not to say it would not be a good one. I have some doubts, but also admit to at least a bit of sympathy for the idea. Actually implementing it would not be easy. This all would be a topic for another day. ↩︎
  2. The piece links to an article on Axios, which requires one to register to read, and I didn’t. So perhaps the Axios article has these details and I would not know. (Were I an editor at MSN, I would have told the author these highly important details should not be glossed over, and one should not have to follow a link to get them!) Common Dreams also has an article extolling RCV for presidential primaries, and it also avoids mention of the issues I raise in this planting. ↩︎
  3. Beckwith claims that “ballot exhaustion” will not be a problem in a “high profile presidential primary.” I admire his confidence in voters. I will just insert my own anecdote as a reasonably engaged observer of politics that during the 2020 contest I was very glad that I did not have to come up with a ranking of a field that mostly consisted of a bunch of candidates with obvious flaws but not much differentiating them, other than a few that I actively disliked.
    Besides, aș I note elsewhere in the post, many states are sure to restrict the number of preferences allowed to a number that may often be lower than the total number of candidates. In this sense, ballot exhaustion will be a built-in feature of the rules. ↩︎
  4. I am not going to get into the question of whether states even would agree to pass legislation permitting ranked-choice ballots for presidential primaries. This is potentially a very big obstacle. ↩︎
  5. And yet the losers eventually usually endorse the presumptive winner within their party. Of course, the wounds of a bitter campaign may undermine the winner in the general election. This is a general problem with primaries, and no electoral system is likely to do much to dampen this fundamental problem when the ultimate endpoint is only one can win a highly desirable prize. ↩︎
  6. More limited implementations of RCV could be to use it around the threshold. If your first-choice candidate fails to get more than the 15% threshold (or is otherwise too unsupported to win a delegate in your district), the vote transfers so it can be used by a candidate with enough support to win delegates. Some states have done this already in caucuses. (This seems to be what Unite America had in mind for Democrats in its report from 2020.) It might be a better idea, although when applied to primaries it still has the problem mentioned in an earlier footnote whereby state legislatures and governors must be convinced to adopt it. Moreover, states or the state party limited preferences to five, which is well below the number of contesting candidates in many early states. (On these rules, see New America 2021, a report that is actually sensitive to the issues I raise here.) ↩︎
  7. Despite the fact that that one state in the GOP process doing precisely that in 2024 elicited praise from a pro-RCV group at the time. ↩︎

SNTV comes to Kyrgyzstan (and a discussion of how to classify SNTV and to define “PR”…)

Posted on by

According to OSCE PA, Kyrgyzstan has a new electoral system:

Introduction of a majoritarian single non-transferable vote system with 30 new three-member districts, a gender requirement per electoral district, while also revising rules on campaigning, financing, and candidate eligibility.

The linked article notes that the organization will continue to have updates on how voters and candidates adapt to the new rules.

And, yes, we could quibble with calling SNTV “majoritarian” but it is better than calling it “semi-proportional” as many folks do.

(Hat tip to Jack for this one.)

Update: as reflected in the new, parenthetical, part of the title, there is also a lively and interesting discussion of how to classify SNTV, and even what it means for a system to be “PR,” in the comments.

A short bill, climate smart agriculture, Burt Lancaster, Diwali, and top-two elections

Posted on by

In the process of some research on bills in the California legislature related to climate change impacts on the state’s agricultural sector, I ran across this one.

“It is the intent of the Legislature to enact subsequent legislation that would encourage smart climate agriculture.”

That is not the summary. That is the full text of a bill introduced by Assembly Member Ash Kalra.

The bill never got a committee hearing and the bill history indicates “Died at desk.” Members of the California legislature are restricted by chamber rules in how many bills they can introduce in a session, so I wonder what he was planning to accomplish by using up one from his bill allotment with this “intent of the Legislature” bill.

Of course, I then wanted to know more about Kalra. He is a Democrat (not surprising, as “climate smart” is an unlikely keyword in a Republican bill) representing the 25th Assembly District, including a large part of San Jose and a rural swath extending well to the southeast of the city. In fact, he was formerly on the San Jose city council. Per the Wiki page, his career has been a demonstration of the top-two electoral system in all its glory. Recall that, despite official labels (carried over unthinkingly by the Wikipedians), California no longer has primary elections for its state legislature (or anything else, aside from presidential nominating delegates). We do not have contests that advance a single nominee (for single-seat district races) of each recognized and participating political party, which is the plain definition of a primary election. Rather, we have a first round out of which the top two vote-earners–regardless of party affiliation (if any)–advance to a second round many months later.

The first time Kalra ran for State Assembly in 2016 he managed only 19.8% of the first-round vote. This was good enough for second place, so he advanced to face Madison Nguyen, also a Democrat (and also a former San Jose city council member), in the runoff. She won 34.3% in the first round of this assembly contest. The third place finisher was a Republican who got only 14.7% and thus missed out on the runoff. (Three other Democrats split the rest.) The Democrat-on-Democrat second round saw Kalra beat Nguyen, 53.2%–46.8%, in what would prove to be his closest margin to date. I knew nothing about this district before this planting, but it certainly is interesting that Nguyen did so poorly after a wide margin in the first round.

Then things get fun. In 2018, no candidate was on the printed ballot to challenge Kalra in the “primary.” There was, however, a write-in candidate, who with only 285 votes (0.5%) advanced to the runoff. This candidate’s name was G. Burt Lancaster. At least partially a famous name. In the runoff he actually got almost 28,000 votes, apparently just for having that R by his name (and presumably the name itself helped). I wonder where this ranks all time in the records of candidate vote increases from round one to round two. Kalra won reelection with over 90,000 (76.3%). This was not Lancaster’s last try. He and Kalra were the only two candidates in the not-a-primary in 2020. So Kalra got to trounce Lancaster two more times that year–once with 75.7% and then in a squeaker with only 72.6% in the second round.1 Since then Kalra has continued to win easily against a different Republican candidate, 70.0% and 68.4% (getting closer!). In the “primary” of 2024 there was another Democrat, but that will be a mere footnote, though this candidate, Lan Ngo, did manage 22.3% of the vote (51.5% for Kalra and 26.2% for Republican Ted Stroll, so it was not too far off being another D-on-D runoff).

As noted earlier, the bill on climate smart agriculture2 would have stated that there would be future legislation on this topic. I do not see any subsequent effort by Kalra to revive this bill or introduce something else on the topic of climate smart agriculture in his list of bills sponsored. He has, however, authored several other environmental bills, including one that passed regarding migratory bird protection, and one that did not on conservation of the Diablo Range (mountains that are partially in his district). He also sponsored a successful bill adding Diwali to the list of official state holidays (Kalra is “the first Indian American to serve in the California Legislature” and, per Wikipedia, a Hindu) and sponsored various failed bills including (1) to recognize India’s Independence Day and “urge all Californians to join in celebrating” it, (2) to recognize “June 21, 2025, as the 2025 International Day of Yoga in California,” and (3) to rename a portion of Highway 101 “Little Saigon Freeway” (as hinted by the names of some his opponents–although not G. Burt Lancaster–there is a large Vietnamese community in his district). I do find members’ signaling to communities with their bill sponsorship endlessly fascinating!

That is the post. That is the whole post. No profound insights. I just found the very short bill interesting, and I always like pointing out what a stupid electoral system California has, as well as that it has a very active legislature (and a great legislative database!).

  1. California’s top-two election law requires a second round (“general”) even if there was a majority for a candidate in the first round (“primary”). ↩︎
  2. Actually, as one can see from the quoted text of the bill, it says “smart climate agriculture.” I usually see the first two words in the opposite order. So I am sticking with that. ↩︎

Is the fit to the SPM different at different seat products? Netherlands 2025 in perspective

Posted on by

The question in the title is one I do not think I had ever checked before. The occasion for pondering this was the recent election in the Netherlands, which resulted in a unusually high degree of fragmentation, with the leading party winning only 17.3% of the seats. In a comment, one reader suggested that such fragmentation did not seem odd given such a low threshold (0.67% of the vote is sufficient for a seat).1 Without doubt, the Netherlands has one of the most “permissive” electoral systems in current use anywhere, where permissiveness refers to the ease with which the rules facilitate small-party representation and thus lower the barrier to entry of new parties. Such permissive rules tend to reduce the size of the largest party as there is little reason for elites and voters to coordinate on big parties to avoid vote wastage. Few votes will be wasted when every 1/150th of the vote gets a seat.

Nonetheless, I pointed out that such a small leading party put this election far below the expectation for even as permissive a system as the Netherlands has. Still, it seems quite sensible to think that maybe electoral systems this permissive have a greater tendency to have a leading party below expectation than do less permissive systems; that is, extreme permissiveness might render drawing our expectations from the seat product more fraught. So I went to the data to find out.

The Seat Product Model (SPM) for simple (single-tier) electoral systems tells us we should expect the seat share of the largest party (s1) to be the “seat product” raised to the power, –1/8. The expectation is based on deductive logic, as explained by Taagepera (2007) and further elaborated and extensively tested in Shugart and Taagepera (2017). For any given election, then, we can compute something I will call the “s1 ratio,” which is simply the actually observed s1 in the election, divided by the expected s1 given the rules in use. The seat product itself if easily calculated if we know the mean district magnitude and the assembly size: just multiply these two values together. If the electoral system consists of a single nationwide district, as in the Netherlands, the seat product is actually the square of the assembly size. If, at the opposite extreme, the parliament consists of entirely single-seat districts, the seat product is equivalent to the assembly size itself. If an assembly were elected under a districted proportional system in which the assembly had 500 seats and there were fifty districts, that would indicate an average magnitude of ten, and thus a seat product of 500 x 10 = 5000. And so on.

There are two good ways to assess whether the seat product itself has an impact on how well the SPM fits. The first is to take the standard deviation of the s1 ratio across any given steady-state electoral system, and then plot these standard deviations against the seat product. The second is to plot all elections for which we have data, and see if s1 ratio tends to be nearer 1.00 (indicating perfect fit) at some parts of the observed range of the seat product than at others. The following two graphs take up both of these approaches.

First, using the standard deviation of s1 ratio. Here each data point shows one electoral system, defined as (reasonably) steady rules in simple systems over time,2 as long as there were at least ten elections under said rules. The red line is a locally weighted regression (lowess).

As we see, there is not a strong relationship between the seat product and the standard deviation in s1 ratio. However, what relationship we see is downward. That is, as the seat product increases–and the electoral system thus becomes more permissive–the standard deviation tends to be somewhat lower. The tendency is for the very most restrictive systems to have higher standard deviations.3 However, the Netherlands actually has the highest standard deviation in s1 ratio despite having the most permissive electoral system! This system (since 1956) is represented by the data point at the far right of the graph (seat product of 22,500). I did not continue the lowess estimate out to such high seat products because the data are so sparse out there. The Dutch standard deviation is 0.238, and the only other systems higher than 0.2 are two small-assembly FPTP systems (St. Vincent and the Grenadines with a seat product of 13 and Trinidad & Tobago wit a seat product of 36) and two small-assembly districted proportional systems (Costa Rica, with an assembly size of 57 and seat product of 464, and Luxembourg, with an assembly size of 60 and a seat product of 676). So the Netherlands, over the decades this system has been in use, defies the trend by combining electoral-system permissiveness and rather variable s1 ratio, although as noted, it is not a strong trend.

The second graph shows every election in the dataset plotted with its s1 ratio. This graph includes two systems with even higher seat product that had too few elections to calculate a meaningful standard deviation: Serbia (250-seat single district) and Ukraine (450 seat district, but only in 2006 and 2007). In this one, I marked the Dutch elections in a different color (orange, of course). The orange data points at seat product of 10,000 refer to elections in the Netherlands before 1956 when the assembly size was 100.

The first thing to call attention to is just how flat the lowess curves are and how close they are to 1.00! This is the case through seat products from around ten all the way to around 5000. The SPM is quite good! The lowess that covers the entire data range then turns rather significantly upward at very high seat product. There are two plausible explanations for why the lowess curve eventually goes above s1 ratio of 1.5, and they are likely reinforcing: Serbia and Ukraine (2006-07) have tended to have political conditions that support relatively large parties despite their high seat product, and they also employ high legal thresholds. Serbia has required 4% or 5% of the vote to get a seat, and Ukraine required 3%. While a 3% threshold may not seem “high,” it is high relative to that huge seat product, given that if the minimum percentage needed to win a seat were 1/S (S being assembly) size, as in the Netherlands, it would have been 0.22%, not 3%.

For this reason, I also included a lowess that stops with the seat product of 22,500 (dashed green line). It too ticks upward, suggesting that there is some general political resistance to leading parties being as small as a seat product in the high thousands could support. However, the cause of that uptick is not the Netherlands. In that country, the mean s1 ratio over 24 elections in the dataset (which does not include 2025) is 1.01! The SPM explains the Netherlands, on average, almost perfectly! Rather, it is Israel, Kosovo, Slovakia, and Moldova (the only other cases, aside from Serbia and Ukraine with seat products greater than 9000) that drive what is in any case a very modest uptick of the lowess that is truncated at a seat product of 22,500. Some of these systems with high seat products also have moderate or high thresholds, at least in some years.4

Thus we see that the answer to the question posed in the post’s title is mostly no, with minor qualifications. More-permissive electoral systems do not tend to be more variable in their impacts on the largest party (as measured by s1 ratios of systems), but the Netherlands is an exception to the modest trend by which greater variability in s1 ratio is generally associated with more restrictive systems. There is very little tendency of the SPM to be any less effective at accounting for largest-party seat size at any value of the seat product, if we leave the really extreme cases like Serbia and Ukraine (which, as noted, have high thresholds) out of consideration. And the Netherlands itself has almost perfectly conformed to the expectation on average, but also has exhibited the greatest variation over time in its s1 ratio of any long-established electoral system.

______

  1. See the comment by ecstaticba95be5333. ↩︎
  2. If either magnitude or assembly size changes gradually over time (as in Canada or Luxembourg), it is the same “electoral system.” Only abrupt changes count as a new system. This is similar to how Liphart (1994) defines a steady-state electoral system. ↩︎
  3. This is not shocking: these are small-assembly FPTP systems, and thus small shifts in votes in a few districts can be the difference between one party having barely a majority (or less, if there are more than two parties) and one party winning all or nearly all the seats. ↩︎
  4. Moldova has a 7% threshold, and Israel since 2015 has had one of 3.25%, which is moderately high for such an otherwise permissive system. Nonetheless, in most recent elections in Israel, the seat share of the largest party has been roughly as expected from the seat product, or somewhat on the low side. (Israel is not in the first graph, because the frequent changes in threshold raise the question of whether we can consider rules to be “steady state,” even though the seat product itself has not changed.) ↩︎

Netherlands 2025

Posted on by

The general election in the Netherlands on 29 October produced a really stunning level of fragmentation. Two parties tied for the largest in seats, with 26 seats apiece. That’s 17.3%! The parties in question are the Democrats 66 (D66) and the Party for Freedom (PVV). The D66 narrowly can claim to be the first party due to edging out the PVV in votes percentage, 16.9% to 16.7%.

I was looking back at my posts on the 2023 and 2017 elections, in which I went over the high degree of fragmentation in those elections. But in both those (as well as 2021) the largest party got seat percentages in the 20s, not the teens!

As I discussed at length in the 2017 planting, the expected largest seat share given an electoral system with a 150-seat nationwide district is 28.6%, per the seat product model. That would be 43 seats (rounded). So the result is about 60% of expectation. That puts this Dutch election at the fifth (yes, 5th) percentile of all 769 elections in my dataset that were held in “simple” (single-tier) electoral systems. For comparison, the mean ratio of expected largest-party seat share (s1) to the value expected from the SPM for these 769 elections is 1.04, which means the SPM prediction (which comes from a deductive model, not a regression coefficient) is on average very accurate. The standard deviation is 0.285, and its interquartile range (i.e., the range from the 25th to 75th percentile) is 0.849 to 1.21. So this election is truly extraordinary!

The election was called when the unusual coalition led by an independent prime minister following the 2023 election broke up with the right-populist PVV leaving.

In terms of gains in votes and seats, the D66 legitimately can claim to have “won.” Their gain is 17 seats over the last election (and they surged in polls late in the campaign), whereas the PVV showing is an 11-seat decline from 2023. The other party with a substantial pickup in seats is the Christian Democratic Appeal (CDA) at +13 followed by the Conservative Liberals (JA21) at +8. But of course, no party has actually “won” when an election produces a leading party so far from a majority.

The three largest parties other than the PVV would still be short of a majority coalition (D66 26 + VVD 22 + GL-PvdA 20) and would be difficult given the GL-PvdA is a Green-Left alliance (soon to become a single party). Adding the CDA (18) to that group means the four largest other than the PVV would be a majority (86 seats). Several of the many other smaller parties are towards the more extreme right or are niche parties of one flavor or another. None of the other “major” parties wants to hook up with the PVV. Thus the coalition bargaining will take a while. I do not know Dutch parties well enough to offer any informed speculation on what government might eventually form.

Yes on California Prop. 50

Posted on by

California voters have before them now a special election on a statewide measure, Proposition 50, that would put in place temporary districts for US House elections of 2026, 2028, and 2030. The new districts, which are shown in maps sent to all registered voters, would replace the current maps drawn by the state’s Independent Citizens Redistricting Commission following the 2020 census. The measure being voted on stipulates that the Commission will resume redrawing House district maps after the 2030 census. The measure also makes no changes to the current maps for state legislature or other districted bodies in the state.

I am favor of this measure, even though it is a blatant partisan gerrymander, the maps having been drawn at the behest of the Democratic majority state legislature under Assembly Bill 604.1 The measure being voted on itself is Assembly Constitutional Amendment No. 8 (ACA 8), and is explicitly justified (directly in its text) as a response to Republican gerrymandering in Texas and other states. These states are carrying out mid-census redistricting in order to attempt to cling to a House majority even if voters nationwide swing against the Republican Party in the 2026 midterm election. ACA 8 is officially called the “Election Rigging Response Act” in order to highlight that it is effectively fighting Republican gerrymandering fire with Democratic gerrymandering counter-fire.2 I do not like this interstate redistricting war, but I also do not support unilateral disarmament. So, despite my previously indicated reluctance, I say game on. I am voting for this.

The measure also has a provision that I like, in which it explicitly calls upon the US Congress to pass legislation or propose a constitutional amendment that would require the use of “fair, independent, and nonpartisan redistricting commissions nationwide.” Of course, the state “calling” on Congress does not make action by Congress any more likely. But it still sends an important message. As Henry Droop said in 1869 (attributing the remark to a Senator of the day), and I am fond of quoting,

from Maine westward to the Pacific Ocean, in the last ten years, in no state whatever had there been an honest and fair district apportionment bill passed for the election of members of Congress [except] where two branches of a legislature were divided in political opinion, and one checked the other.

We still have not corrected this problem, and it really does require federal action to fix. As long as we remain stuck with single-seat districts, we should at least make them not subject to boundaries decided directly by politicians or by agents with partisan incentives. But we need to make this correction everywhere, not just in California. The current moment, where we are apparently embarking on the interstate gerrymandering war alluded to earlier, shows that “reform in one state” is not an effective way to bring about “fair” districting. An individual state or two may implement relatively fair districting for its own members of the US House–and California clearly has done so with its current Commission–but if states dominated by one party are the only ones playing the fair-districting game, it is a losing game. The field is simply ceded to the party willing to continue and even aggressively expand its own partisan-titled redistricting. I made this point myself on this blog back in 2010, when the state was considering a Prop. 20 to extend the Commission’s remit to House districts. When originally enacted in 2008, the Commission was to apply to the state legislature, but not the US House. In 2010 I noted,

Now, what about extending it to cover US House districts? I believe I will vote NO on 20 as well. Again, I most certainly oppose letting legislators draw district lines. However, I have never been a fan of unilateral disarmament. The federal dimension matters here, and this measure takes California’s legislature (controlled by Democrats) out of the process of determining the boundaries of 53 House districts (12% of the total number of House seats!), with no reciprocal move by Republican-controlled states to “disarm” their legislatures from controlling a like number of districts.

In fact, in the earlier (2008) post about the initial adoption of the Commission (when it was not even proposed to apply to the House), I had suggested,

Thus redistricting reform in the House presumably should be done via constitutional amendment or an interstate compact (on the model of the National Popular Vote for the presidency).

I believe this is still correct, and thus in a sense this year’s Prop. 50 is a chance to undo the arguable mistake of 2010, to unilaterally disarm by instituting Commission-drawn boundaries in this state while other Republican-majority states continue to gerrymander.

To those–including authors of the argument against Prop 50 in the ballot pamphlet–who say this will mean inevitably that California is “permanently” ending independent redistricting, I call hogwash. The measure that would be enacted via Prop. 50 is a constitutional amendment that explicitly sunsets and authorizes the Commission to draw House districts again after the 2030 census. If voters (or those dastardly “power grabbing” politicians) want to end the Commission permanently, they will have to get a measure on the ballot approved by voters to actually do that. That is not what this measure is about. It is about righting an imbalance where one party, with majorities in several other states, is abusing the redistricting process to attempt to invert the popular-vote outcome of the upcoming federal midterm election. For that reason, as well as the others noted here, I am voting yes.

I will not, however, display a yard sign for it. The ones I have seen say to vote yes for “fair elections.” I can’t call a blatant gerrymandering a “fair” electoral process! However, it may have a fairer outcome in the aggregate, if the combination of aggressive Republican gerrymanders in other states and “fair” districts in California would be to enable an inversion (i.e., a Republican House majority despite a Democratic voting plurality), but if California’s defensive counter-gerrymander permitted a national preference for Democrats to be translated into a majority of seats. I do not know if an inversion would happen absent this measure, or if will be prevented by it. But it is a realistic enough chain of events that I believe we have to engage the war that was initiated by our opponents who want to dismantle electoral accountability. War is not fair. But sometimes it is necessary. To the trenches! Or rather the voting box, with a YES on Prop. 50!

______

  1. AB604, interestingly, was introduced by the member who represents my district, and whom I occasionally write about, Cecilia Aguiar-Curry. ↩︎
  2. To clarify, ACA 8, and therefore Prop. 50, explicitly states that it is implementing the districts defined in AB604. ↩︎

That Knesset vote

Posted on by

There is a lot of consternation over a vote in the Knesset on Thursday that approved a bill to apply “sovereignty” to Judea and Samaria (also known as the West Bank). It has led President Trump to declare that if Israel “annexes” the territory in question it would mean the end of US support for Israel and VP Vance said he was personally “insulted.” (This must be some major cognitive dissonance for those on the Israeli or Diaspora-Zionist right who thought there would be no “daylight” between Israel and the US under Trump!)

The Israeli opposition obviously does not favor annexation. Yet the Israeli Prime Minister has said the vote on the bill was a stunt of the opposition. No one seems to want to claim credit! Well, of course, the two far-right parties in Netanyahu’s cabinet–Otzma Yehudit (OY) and Religious Zionism (RZ)– voted for it and are proud of it. So it was not exactly an “opposition” stunt, although in a very technical sense it was an opposition bill. It was put forward by MK Avi Maoz, the leader of the one-man Noam Party (you can look it up). This was a member of the coalition, elected on the joint slate of OY and RZ, but Maoz withdrew from the government some months ago because it was much too liberal and pacifist for his taste.

The funny thing about it all is the vote by which it passed. It was 25-24. Yes, 25 votes in favor in an assembly of 120 members. A reminder that the Knesset does not have a quorum requirement. It was also a first reading and would have to go through three more before it could become law. In other words, it is going nowhere. But it did cause a firestorm of criticism and embarrassment. Probably no one is more embarrassed than the Prime Minister and Likud leader, Benjamin Netanyahu himself.

Likud’s official line was to boycott the vote. But one member of the party turned up and voted in favor. That would be Yuli Edelstein, the regular thorn in the PM’s side who was recently stripped of his chairmanship of the Foreign Affairs and Defense committee over his reluctance to push through a bill that would codify the exemptions Haredi Jews obtain from the military draft. Edelstein is proud for standing by his values and voting for this sovereignty bill and being the decisive vote that advanced it on first reading. He also is, as of later the very same day, no longer a member of the committee. (He called his dismissal a “moral badge of honor.”) Knesset committee members (like legislative committee members pretty much anywhere) serve at the pleasure of their party. And the party is not too pleased with MK Edelstein. Likud MK Tally Gotliv, an alternate member of the committee, said “Ultimately, the ruling party has coalition discipline… He wanted to win political capital, and there is a sanction for that. The sanction is that he will not be on the Foreign Affairs and Defense Committee.” So party discipline has prevailed, albeit not until after an embarrassment to the party just as a parade of US officials was passing through Jerusalem to try to salvage the ceasefire brokered by the Trump Administration.

As for the actual opposition, it could have provided more than 24 votes against if it had taken seriously the possibility that the bill might advance, and wanted to prevent that. But, of course, embarrassment of the government is not all bad when you are the opposition. Why lend it a hand to defeat a bill when the PM’s own party is not whipping members to vote against a bill it supposedly opposes. What a day in the Knesset!

Legislatures are always fascinating, the Knesset especially so.

Wales: A Labour member introduces bill for STV

Posted on by

Just as Wales is approaching an election that will be its first under a new closed-list proportional representation system, a Labour member of the Senedd is proposing that Wales adopt the single transferable vote (STV). This is, of course, not a new idea. Since getting its devolved parliament (Senedd), Wales has used a mixed-member proportional (MMP) system.1 In 2017 an Expert Panel on Assembly Electoral Reform recommended a change to STV (see here, in PDF). However, in 2024, the Senedd adopted a measure to shift to CLPR instead. This new system will be used for the first time in May, 2026 (see my analysis of likely impacts on the party system).

The bill, introduced by Alun Davies, Member of Senedd (MS) for Blaenau Gwent, to change to STV likely has no significant chance of passing, but is interesting nonetheless for reflecting the ongoing discontent over the rejection of STV, and the decision to adopt CLPR. It has some interesting quotes from critics of, respectively, CLPR and STV. Labour MS Lee Waters, explains how some of the objection to STV came about:

[former Cardiff West MP] Kevin Brennan, who was Mark Drakeford’s close constituency colleague and very influential with him, comes from an Irish family and has an experience of STV promoting parochialism, getting internal candidates to be competing with each other with voters rather than campaigning for the party as a whole, and really did not like the system and thought it would be a retrograde step for Wales. And I think that was very influential on Mark. Mark himself would have preferred STV – but what could he get through?

Parochialism, or locally centered “friends and family” voting, as well as the potential undermining of parties, are all frequently expressed criticisms of STV. It is nonetheless interesting that at the end of the excerpt, MS Drakeford is said to have preferred STV, with CLPR being what was politically viable.

The article also quotes Professor Laura McAllister, who chaired the aforementioned Expert Panel,

Closed list PR was one of the electoral systems that we rejected in the very early stage of our analysis of different electoral systems… Hardly any of those [values against which the Panel assessed proposals] are carried forward with closed list PR, and the most significant thing to say is that it reduces voter choice and gives more authority to the parties in terms of candidates. One of the traditions of voting systems, traditionally and historically in Britain, has been that people go out and vote for a candidate that represents a party, rather than the other way round.

[Of course, she means electoral systems, not “voting systems.”]

Jess Blair, director of the Electoral Reform Society (a pro-STV organization) in Wales says

Closed lists will deliver a proportional result but will not deliver accountability, with voters only being able to back political parties rather than the individual candidates on the ballot paper that will actually represent them in the Senedd…

We warned from the start that closed lists risked being a lose-lose compromise and that we needed a voting system that works for the Welsh people. This draft Bill demonstrates how easy the change from the closed list system to STV would be, if the circumstances after the May election meant a change would be possible.

[Of course, Blair, too, means electoral system.2]

While I would question the claims that CLPR means no accountability of individual members,3 it is quite possible that, in the political context of Wales, the CLPR system was a sort of lose-lose option. Even if the bill to adopt STV does not prosper in the next Senedd,4 the issue probably will not go away.

It seems that the experts like those who were on the Panel, or legislators and others in Wales addressing this issue going forward, might consider taking a closer look at open-list proportional representation (OLPR). The Expert Panel report barely mentioned this option other than in passing, with the most prominent mention being in a rather dismissive footnote:

An Open List proportional system might also have met our principles. However Open List systems are very close to STV in their effects. On the basis that STV is already used within the UK, we decided to focus our attention on STV rather than Open List proportional representation.

No justification is offered for the claim that OLPR has effects “very close to STV.” The claim is easily rebutted, and in fact the other in-passing mentions to OLPR in the report show the experts are certainly aware of the rebuttal, but elected not to entertain it. That rebuttal would be that there is a substantial difference between an electoral system that makes party proportionality its first criterion in seat allocation (as any list system does, for a given district magnitude and assembly size) and one that is entirely candidate-centered and has no party criterion in seat allocation (as is STV, in its standard forms5).

The 2017 Expert report does have an entire section devoted to flexible lists, entertaining a proposal that is modeled roughly on the Swedish list rules. In fact their overall recommendation states it is for either STV or flexible list, although it does conclude by calling STV “our preferred option,” while flexible lists are called “a viable alternative” (p. 213). On the other hand, I still think that perhaps a better option than any of these would be the reformed (and more proportional) MMP that I suggested near the end of an earlier post.

______

  1. It is commonly known in the UK as the “additional member system” (AMS) but this is a poor term for the system, given it is a compensatory system and therefore a case of what electoral-system experts generally call MMP. See Lundberg (2018; abstract only unless you have library/subscription access). ↩︎
  2. Yes, I will concede that the voting is different under STV, due to the ranking. So it is in a sense a “voting system” difference. However, all this talk about how seats are allocated, how proportional the result might be, and how accountability would work are references to the collective process of allocating political power, which is what an electoral system does. ↩︎
  3. Most such claims are not based on actual evidence. It is not my sense that individual members are simply party tools with no public accountability under actual CLPR systems. The process works differently of course, but accountability still happens. And all electoral systems involve tradeoffs. ↩︎
  4. The article indicates it would need a two-thirds majority, which Labour and Plaid Cymru had between them in the current Senedd, but likely will not in the next one. It does not address the likely preferences of other parties. ↩︎
  5. Unless there is an “above the line” option, that is a list vote that results in a process of pooling votes among associated candidates. The electoral system then ceases to be STV, as generally understood. ↩︎

Japan: Upcoming change in PM

Posted on by

The following is a post that started as a comment elsewhere on this blog by Yibo. With permission, I am elevating it to the center row of the virtual orchard so it will be more widely seen. The background is that the recent change of LDP leader, which necessitates the change in Prime Minister, comes in a no-majority parliament due to the outcome of the 2024 election.

_______

Japan’s Liberal Democratic Party (LDP) recently elected Sanae Takaichi, a woman from the far-right faction, to become party leader. Komeito decided to tear up the coalition deal with LDP after some discussion. Now, there are talks amongst Komeito, CDP, Isshin and DPFP to nominate DPFP leader Yuichiro Tamaki to become PM in the upcoming PM election that’s happening on or after October 20th when parliament reopens. This is incredible considering DPFP only got 28 seats in the last election while CDP got 148 seats. However, it is clear that Tamaki has more popularity among young people and swing voters than CDP leader Yoshihiko Noda, a former PM. A CDP-Isshin-DPFP-Komeito gov would be barely a min-winning coalition of 234 seats out of 465. Current seat distribution can be seen from the parliament’s official website.

Last thing I [Yibo] want to point out is that two far/radical left parties, Reiwa and the communists, have 17 seats. Currently, the anti-LDP coalition that is being talked about emphasises its centrist credentials, so it seems like the centrist parties would do everything to dissociate themselves from the radical left. However, when it comes to the PM election in the lower house, the 17 left-wing MPs’ votes can be crucial in the runoff if one of the minor centrist parties abstain from the 1st round. The constitutional procedure to elect the PM in Japan is for the lower house to first hold an open vote in which a candidate needs an absolute majority of the total membership of the chamber to get elected (233/465). If this doesn’t happen, a 2nd round runoff occurs between the top two vote-getters.

Anything can happen in the next few days.

Leveling vs. compensatory vs. top-up

Posted on by

In discussions on this blog about electoral systems of Norway and Sweden, some of you use the term, leveling seats, to refer to the seats that are allocated based on nationwide results to “level” or smooth out disproportionality that arises from the otherwise district-based allocation of seats. I would tend to refer to these as compensatory seats. And then there are others who use the term, top-up seats.

Are these just synonyms? Or do you understand any of them to refer to something different from any of the others?

I suppose we could add adjustment seats to the mix, and perhaps other terms that may (or may not) refer to the same mechanism.

French cabinet loses confidence vote

Posted on by

The National Assembly of France today voted against a confidence measure in the cabinet of Premier François Bayrou. As a result, per Art. 50 of the constitution, the premier must resign and the government is ousted. President Emmanuel Macron had appointed Bayrou last December after the previous premier, Michel Barnier, had been ousted in a no-confidence vote in December 2024. The deeper political problem goes back to the July 2024 election, which Macron called under his constitutional discretion, but which predictably went badly for him.

Under the premier-presidential system of France, the premier must maintain the confidence of the assembly majority. However, under the French variant, the assembly has no formal role in the initial appointment, not even a requirement that it confirm a nominee. The president constitutionally can appoint under his discretion. But a selection who is unable to maintain the confidence of a majority will not last in office, as the case of Bayrou–like Barnier before him–has demonstrated.

In this case the vote the government lost was a vote of confidence, unlike the vote of no confidence that toppled Barnier. The difference is that the premier himself (rather than the opposition) called the vote, knowing he must resign if he loses. Normally a vote of confidence is called to reinforce the position of the government, that is, in expectation that the government calling the vote will win it. I believe this is the first time in the Fifth Republic that a cabinet has fallen under this specific procedure. Again, the underlying phenomenon here is the lack of political support in the assembly. Bayrou likely had no chance of surviving in office, but also no realistic chance of passing a budget. The vote was 194-364, so somewhat short of a cliffhanger. Commenting on the vote, Bayrou said “The greatest risk was to not take one, to let things go on without changing anything, to go on doing politics as usual.”