California’s STUPID electoral system, 2022 first round edition

Yesterday was the “primary” that is NOT a primary in California. As I tried to warn the good voters of the great California Republic back in 2010, this “top two” system would be a bad idea. Yesterday offers some further examples of why it is indeed a STUPID ELECTORAL SYSTEM.

My favorite current example is state Senate District 4 (yeah, we do boring district names here).

Source: CATarget on Twitter.

Nearly 56% of this district’s voters voted for candidates branded on the ballot as Republican. Yet, because this is NOT A PRIMARY, but is just a top-two runoff system, the voters will choose in November from two Democrats, whose combined vote total is just 44%. Brilliant!

(For Democrats, it almost looks like a successful contest under two-seat single nontransferable vote (SNTV), with the party coordinating to equalize on two candidates, but I won’t give them that much credit. As for Republicans, well, they just punted away a win for the taking in a body where they regularly struggle to win even one third of the total seats.)

Statewide, we may have an intraparty runoff in one contest, the one for Insurance Commissioner. When the count was at 75% reporting, it looked like this:

Note the close race for the second runoff slot, between a second Democrat and a Republican. In this case, even though I will not waver from my conviction that this is a STUPID ELECTORAL SYSTEM, I will be happy with the result if Levine faces Lara in the runoff. In fact, Levine was one of the few candidates on this whole LONG ballot that I actually voted for. Lara is very expendable, due to being somewhat ethically challenged. In fact, it is certainly not impossible to imagine him losing in November, even to a Republican. Anyway, regardless of how I feel about the specific candidates, the notion that the so-called general election would be between two candidates of the same party is a bug not a feature. That it is for a statewide contest makes it even more so. (It would not be the first time; in fact, twice we’ve had US Senate contests on a November ballot that were between two Democrats.)

In a subsequent update, with “100% reporting” Howell has pulled ahead. However, 100% does not mean the count is finished. Far from it! There is probably still a decent chance Levine will pull back ahead for the second runoff slot. If Levine pulls back into second place, the Republicans will have shot themselves in the foot by their almost perfect vote-equalization “strategy.” If one of the Republicans finishes ahead of Lara, I am going to be mildly upset at Eugene and JJJ for splitting the anti-Lara Democratic vote. Ah, the hazard of SNTV-style competition for two slots in an eventual one-seat contest!

Speaking of US Senate contests in California, we got two of them this time! No, it is not that both seats are open (as was the case here in 1992, or more recently in Arizona and Georgia); both votes were for the same seat. The incumbent, Alex Padilla, was appointed by Governor Gavin Newsom to the vacated seat, and state law requires that there be a special election at the next scheduled election to fill the unexpired portion of the term. So we voted on that, as well as on the new six-year term starting in 2023.

The image on the right is the unexpired term term, and it is on the back of the page that has the full term (image on left). This is confusing! There are also many fewer candidates for the partial term than for the full term. We get 23 of them for the full term!

I am actually not sure whether the rule in the partial term is top-two runoff, or if a majority on this ballot will suffice. [Update: both are same top-two rules; I’ll keep the rest of this paragraph as it was.] It is used to be the case, even before top two was adopted for all formerly partisan elections, that special elections could be over in one round if one of the candidates–with all, regardless of one party, running against each other–won over half. I do not know if that applies here, or if the top two automatically go to a runoff, as in the full term election. If a majority suffices, you technically could have someone sworn in right away to serve only till early January. If the runoff is required anyway, then the person elected for the partial term would serve for only a few weeks. Of course, it is moot. The appointed incumbent is sure to win both contests anyway. But this is another of those poorly thought out provisions of California election law that could produce a strange result (not as bad as the recall/replacement process, about which see what I wrote last year, but bad enough).

Another thing I was watching for was to see just how well Newsom would do. At the moment, he is all the way down to 56.3%, although that percentage could well creep up again. Just last year, NO on the recall got 61.9%. In the 2018 November runoff against a Republican, he won 62.0%. So he may be slipping! Okay, not by much. Even though he will easily win a majority in this round, we get to vote on him yet again in November. What a great democracy–a majority gets to proclaim it wants its governor three times in just over a year!

(Side note: If you add the votes of a few other token self-identified Democrats running for governor to Newsom’s total, you get 58.4%.)

His opponent will be Brian Dahle, a not so well known very far-right and evidently anti-vax Republican from one of the state’s most rural districts, way up on the Oregon and Nevada borders; in fact, as far as most Californians are concerned, it might as well be Idaho. (He is interesting in that he serves in the state senate while his wife, Megan Dahle, serves in the state assembly, in the seat Brian formerly held. They are farmers, so I have kind of followed their careers.)

And then my ballot also included this contest.

That’s right, we have just two candidates. But, of course, this is a top two election, so they both have already qualified for the November ballot just by showing up! WHAT A STUPID ELECTORAL SYSTEM.

As you can guess from the candidates’ indicated occupations–why do we even let candidates list their occupation on the ballot?–and as you would know from the “Fruits” side of this blog, this is very much an agricultural district. Aguiar-Curry makes ag policy a key part of her legislative behavior (so I follow her career, too!). She has been a walnut grower for years. Walnuts are a major crop here, as are wine grapes, as well as almonds, tomatoes, corn, sunflower…. the list is long.

(Funny aside, B. Dahle derides Newsom as a “wine salesman” but here we have a Republican candidate who proudly lists “winemaker” as one of his occupations on the ballot. Wine is big business in this state!)

For this state assembly contest, I was tempted to vote for one now and the other in November. Just because. But instead I decided to vote for neither.

I have never left so many parts of a ballot blank. So many candidates (26 for Governor, 23 for the full term Senate seat), so few I cared enough to vote for. WHAT A STUPID ELECTORAL SYSTEM, and what a disappointing excuse for a democracy the California Republic has.

33 thoughts on “California’s STUPID electoral system, 2022 first round edition

  1. Ranked Choice Voting coming to future California elections? This could improve California elections. California is one of the only Western US states that requires special elections to fill vacant seats in the state assembly and state senate, other states do appointment.

    Liked by 1 person

    • Re RCV, assuming we are talking about no other changes (i.e., single-seat elections) it is questionable whether that would “improve California elections.” If done on the Maine model, maybe. But we would be more likely to double down on nonpartisan elections if we had an RCV movement get a measure to the ballot here. And whether that would be an improvement is a very debatable proposition (so to speak!) indeed.

      Liked by 2 people

  2. This is not exactly on point, but I think it’s interesting that when Marc Levine first ran for his current job in the state Assembly he was a poster child for the alleged advantage of Top Two — a moderate (meaning pro-business) Democrat who advanced to the second round instead of a more progressive Democrat by getting crossover votes from some Republicans. His election is one of the few instances in which Top Two appears to have produced the result desired by its proponents. Fast forward ten years. Now he is running as the progressive champion of consumers against the big bad insurance companies. I’m overstating the contrast, but only slightly.

    If the proponents of Top Two had their heads on straight, they would realize by now that they would get more of what they want, at least in the Legislature, with PR. Instead, they are now promoting Top Four and Top Five as in Alaska. At least part of their purpose in this is to co-opt activists working for ranked choice (instant runoff) voting to be their allies rather than opponents.

    Liked by 1 person

    • Yes, IF. And you are right, these “reformers” keep digging deeper and deeper into bad ideas, rather than going for the obvious fix that is out there.

      Interesting observation on Levine, and the rarity of the outcomes of the type top-two advocates sold the state on.

      Liked by 1 person

      • Also, Bob, thanks for pointing out a really bad typo (or auto-“correct” failure). Given that I have fixed it, and it was indeed embarrassing, I have deleted the comment in which you pointed it out.

        Liked by 1 person

  3. The only thing worse than requiring a referendum to approve changes to the electoral system is allowing citizen-generated referenda to change the electoral system.


    • I assume the motivation for this is to co-opt the supports of Top N into supporting PR (on the theory that what’s sauce for the goose is sauce for the gander). I think that’s a really bad idea. If I were going to propose a compromise with Top N, I think it would be “non-partisan” top N for executive branch offices combined with partisan PR for the legislature. But I’m not sure even that would be acceptable, although the damage could be mitigated by converting some executive branch offices from elected to appointed with confirmation by the legislature..

      The whole point of Top N is to give people with lots of money a way around the political party organizations. I’m rarely in the mood to compromise with that.

      Also, reusing the state Senate districts would result in two-seat Assembly districts rather than three-seat Assembly districts.

      Liked by 1 person

      • Yes, it would simplify the marketing of PR. I agree a jungle primary w/ 7 winners wd be chaotic but if one increased intra-party hierarchy via campaign finance regulations that channeled $speech through party leaders, the party would basically still decide not unlike how it did in IL from 1870-1980 with 3-seat quasi-PR.

        My idea would also entail expanding the number of state reps seats by 50% as well.

        Super-PACs are inefficient. I think it’ll be do-able to end Citizens United in a way that rechannels $speech through party leadership at the state and nat’l levels. That should mitigate the original intent of Top N primaries.
        I like the idea of having he newly elected (by PR) state reps be the final stage for statewide single-winner elections.


      • I gather “three-seat Assembly districts” are what you want. Although I am not a fan of STV for a country with the geographic challenges of Canada, it worked pretty well in Northern Ireland with six-seat districts. They chose six-seat districts because, after watching the Republic of Ireland tinker with STV for decades (shrinking District Magnitude for partisan advantage, like Tasmania), they knew it would be needed to respect Northern Ireland’s political diversity. But then they shrank it to five-seaters recently, and sure enough, no more Greens. That’s the phrase I like to use, after the Quebec Select Committee on the Elections Act considered their government’s model with five-seaters and recommended, on May 31, 2006, “that the principle of increasing the political diversity of the National Assembly be respected.” “The goal of political diversity would not be well served then by a regional compensation formula that set the threshold too high.” (Five was too small.)

        Why would any electoral reformer advocate three-seaters? Any good Irishman will tell you “Three-seaters are s**t.”


        • “Why would any electoral reformer advocate three-seaters?” Because this is the United States, where even advocates as militant as myself often think that anything is better than nothing.

          Liked by 1 person

        • Tasmania is about to revert to magnitude 7. The shrink to magnitude 5 was supposed to exclude the evil Greens, or so the two majors fondly happened. The reality is that while it worked at the first magnitude 5 election in 1998, and Greens fell from 5 in an assembly of 35 to 1 in an assembly of 25, by 2002 they were back to 4 out of 25.

          The expanded assembly may also include indigenous seats on the NZ model.

          Liked by 1 person

      • My feeling is we have no choice but to compromise with the top-x (or N, but I prefer ‘x’ or ‘c’ so as not to confused with number of parties). And that was the spirit of the “nominating districts” component of my OLPR proposal (for which Jack Santucci deserves equal credit and no blame).

        (This comment was meant as a response to Bob at 09/06/2022 at 7:07 am; I thought I had set the blog now to allow another level of nesting comments, but apparently it is not working that way.)

        Liked by 1 person

  4. We know from IL that 3-seat quasi-PR handicapped the rivalry between the two biggest parties and prevented either from dominating the state and both became more dynamic. Also, if they used a Hare quota in the 2nd round, it’d help outsiders potentially win the 3rd seat, increasing interest in the elections.


    • Please do not call cumulative vote “quasi-PR.” It is nothing of the sort.

      Some political scientists call SNTV “semi-proportional” and cumulative vote is not much different from SNTV. I have often asked my political scientist colleagues to ditch this term, semi-proportional, as it is misleading and inaccurate. I think “quasi-” is no better here, and possibly worse.

      Cumulative vote is just another in the family of non-list systems. (So is STV, and I’m willing to call that PR, although I am uneasy with that designation. That’s a topic for another day.) SNTV, cumulative vote, and others in that family are not anywhere near PR formulas.


      • It was in the aggregate nearly proportional between the two biggest parties in its results, as I recall.
        Given rational strategies by parties, I don’t think the sort of options given voters matter much if there are only 3 seats and a droop/d’hondt quota.
        Cumulative voting only makes sense if there is a good reason to give extra votes to some voters than others, like with its use for corporations.

        From a marketing standpoint, I’d rather ignore mentioning the use of cumulative voting and just focus on how it was semi-PR in its results…


        • Even M=1 can be in the aggregate quite proportional, including in the US House! That does not make any form of M=1 a “semi-proportional” system.


        • I agree it’s not well defined as a term. I use it to obscures the specific cumulative voting rule that I think was inconsequential given the strategies of the party leaders. I googled it within google scholar and semi or quasi proportional does have some usage there… What would you suggest? I have no interest in giving attention to SNTV either but I think it’s very important that USAmericans recognize that (quasi-) proportional representation played a critical role in our history, due more credit than is given to it.


      • The “semi” in “semi-proportional method” can be understood in two different ways. (1) The method gives results that are not as proportional as possible given the threshold but are nonetheless more proportional than FPTP. (2) The voting method can give results that are as proportional as possible given the threshold but ONLY if candidates and voters both co-ordinate properly. The cumulative vote is “semi-proportional” in the second sense but not the first. Fully proportional implies that the method does not require such co-ordination in order to work.

        Liked by 1 person

  5. California Republicans elected to the 2nd state representative seat, using PR, might be a good influence on the rest of the party, unlike with Rep. McCarthy.


  6. At the moment, the race for second runoff slot for Insurance Commissioner between the top Republican (Howell) and the second Democrat (Levine) is back down to 0.3%!

    (I am following it on the SacBee’s regularly updated results page.)


    • So what’s the contingency for that w/ top 2 primary? Do both of them go on and the state lists all 6 possible rankings of the 3 candidates, except for the ones that rank both the 2nd or 3rd place over the first place candidate?


      • Under current California law, only the top two go to the second round. If you’re the second place candidate, it makes no difference whether you finish one vote ahead of the third place candidate or 100,000 votes ahead. Under Alaska-style Top X, X candidates participate in a conventional IRV (AV) second round.


        • So, there’s no contingency for when it’s too close? That’s a shame!

          Mshugart, I was just using their term they used. I know the diff…


  7. Pingback: The French thresholds for runoff participation | Fruits and Votes

  8. One more comment. It’s in response to this paragraph above:

    (For Democrats, it almost looks like a successful contest under two-seat single nontransferable vote (SNTV), with the party coordinating to equalize on two candidates, but I won’t give them that much credit. As for Republicans, well, they just punted away a win for the taking in a body where they regularly struggle to win even one third of the total seats.)

    But it wasn’t the Republican Party that made this mistake. It was individual Republican candidates (in this case, four of them). The whole point of top two is to prevent political parties from having any control over which (or even how many) of their members appear on the ballot. Under any constitution that was properly drafted, adopted and interpreted by the courts, this degree of infringement on the right of association would not be allowed. But under the U.S. Constitution it’s just fine.


    • Yes, of course, I understand that. Thats’ why I mentioned that it “looked like” things I see in SNTV data. But actually, even in Japan (and far more so, Colombia), party coordination of candidate entry and behavior was pretty imperfect, even though parties in both countries got generally better and better at it over time.


Leave a Reply

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

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

Twitter picture

You are commenting using your Twitter 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.