One thing I’ll note about the Missouri ballot initiative: it doesn’t just get rid of RCV. It also requires primaries only put one candidate forward. The reason that’s important is because St. Louis City has already implemented what is, in my view, a superior alternative to RCV with an approval primary followed by a runoff. The legislature wants to get rid of all of these alternative voting forms because I suppose they like extreme candidates plus they hate whenever a major city does something nice for itself.
I am deeply frustrated with all the money going into RCV, which in its single-winner form just does not do what its advocates want it to do. The big org that promotes it, FairVote, will tell you that RCV gets rid of the spoiler problem. You can vote for your true favorite, and then second-rank your "lesser evil", and everything will be great. This is, generously, only true as long as your true favorite is irrelevant. To be more direct, they know it's false, and I do not understand why they keep saying it. In fact, the spoiler problem in RCV is _worse_ than in our current system.
Suppose you have a town with 100 voters. It's election day, the polls are going to close in five minutes. So far our vote tally looks like this:
35 Right > Center > Left
16 Center > Right > Left
16 Center > Left > Right
31 Left > Center > Right
Right now if you did the RCV tabulation, you'd find that Right has 35 first-choice votes, Center has 32, and Left has 31. So Left is eliminated, those votes go to Center, and Center wins with about 2/3 of the votes. Notably, Center is preferred by ~2/3 of voters over either alternative, so in some sense they're obviously the "correct" winner.
But note also that there are only 98 votes! The final two voters just make it to the polls. They vote with the Left faction. Now what happens? Well, Left has 33 to Center's 32 in the first round. Center is eliminated. Center's votes split equally... leading to a 51-49 victory for Right.
Not only is this stupid because Center was still preferred by 2/3 of voters over either alternative. The people who changed the result were Lefties! They would've been happier if they had just stayed home! This is no way to run an election. At least under Plurality, when you vote third party, you can't actively move the results towards a party you hate.
Center squeeze elections are shockingly common.
Possibly the most famous runoff failure in history (which was not an instant runoff, but would surely have delivered the same result if it were) was the David Duke race. Former Democrat / moderate Republican Buddy Roemer got squeezed out in the first round between the prolifically corrupt Democratic machine politician Edwin Edwards, and the Klansman David Duke.
The recent election of Mary Peltola in Alaska is also obviously a center squeeze. Nick Begich, the anti-Trump Republican, should have won, but was squeezed out between Palin and Peltola. (I'm obviously happier with Peltola, but it's still not the right result.) As noted in Ben's post, there's a movement of Alaskans to try to repeal it.
Burlington, VT adopted RCV, immediately had a Center Squeeze election, and then repealed it. (There are now some rumblings that they might adopt it again, because apparently nobody learned anything from that experience, and getting people to try anything that seems new is incredibly difficult.)
San Francisco also had a very near miss with a squeeze a few years ago, where the progressive Vilaska Nguyen came within a couple dozen votes of knocking out the less-progressive / more-moderate Myrna Melgar, but if she had, the result would've been electing the even-more-moderate candidate (this is San Francisco, there are no real conservatives) Joel Engardio. And just in general, SF politics are insanely polarized and nasty.
Portland, OR, and the small city of Albany, CA, have adopted the _proportional_ version of RCV, which is much better. It technically still has the center-squeeze issue, but it only applies to the final seat selected, and at that point you're talking about very close margins where you might as well have a coin toss anyways. The Australian Senate, IIRC, also uses STV-PR.
But really if you want an election system that reduces polarization; allows for factions of existing parties to develop into actual distinct new parties; and encourages more, collegial campaigning; then you want Approval ballots, where you treat each candidate like a ballot proposition. "Do you like X? Yes/No" This forces every candidate to court support from people whose first choice is some "ideological neighbor". Going negative may hurt your target, but it also turns off people who otherwise might give you a secondary vote. We have documentary evidence that this works, from Fargo and (as you noted) Saint Louis, which have both adopted Approval within the last few years (though still with single-winner elections).
Approval also has the advantage of being far cheaper than a ranked ballot, both in terms of direct printing costs, and voter education. Think about the giant grid you'd have to print if there are 30 candidates for a seat, and folks have to rank them, potentially all the way from first to thirtieth. Additionally, that kind of grid causes much higher ballot spoilage rates -- anywhere from 3x to 7x, if you look at situations where there are ranked and not-ranked races within the same election. Intuitively it's obvious why -- if you mark two candidates as both First, IRV just trashes your whole ballot. This higher spoilage rate is of course concentrated on voters who are less educated, less wealthy, or less English fluent. With Approval, or a Range / Cardinal Ratings system like STAR (Score Then Automatic Runoff) where you rate each candidate independently (zero to five stars, in that case), screwing up a mark on one candidate doesn't void your ratings for the other candidates. (Approval can be thought of as the simplest case of a Range ballot -- you're just rating each candidate as 0 or 1.)
There are also good ways to do proportional tabulation with Approval or Range ballots. In fact the proportional method for Approval was actually invented independently by a French mathematician, D'Hondt, and in the US by Thomas Jefferson. It was used for the initial allocation of House seats among the states.
I really wish some group of rich nerds would invest in giving the Center for Election Science the resources it would need to do national-scale voter education, and run ballot measures in various states.
I actually have been talking with a couple Councilmembers in the city to my immediate south about adopting a city charter that would include using Proportional Approval. And I ran for City Council in my own town (and got 45% against a multi-term incumbent, so not bad for a first try). I will definitely be pursuing that as a priority if I ever get on.
Proportional Representation really is the holy grail here. I've talked with Andrew Yang about this, and while, to my frustration, he has been pushing RCV just because people are more familiar with it and he feels it's an easier improvement to get to quickly, he totally agreed that STV-PR is _way_ better than just RCV for single winner. There's a bill in Congress that would change to STV-PR, on multi-member districts, in states that have more than one House member: https://www.congress.gov/bill/117th-congress/house-bill/3863/text
A PR system would allow sane anti-Trump Republicans to join Yang's Forward Party and make it a real force. It would let Working Families absorb Democrats' DSA wing. Maybe we'd see a functional Green Party. And more generally, it would mean you wouldn't have the thirty percent of "blue tribe" people in rural areas, and the thirty percent of "red tribe" people in cities, feeling completely alienated from the system. No large minority should feel like they have no voice in their government at all. That's a huge factor in the "LOL nothing matters" attitude towards politics that got us Trump.
Approval voting is the way imo, it’s super easy to understand and has the nice property that it’s basically impossible for people to invalidate their ballots by checking the wrong boxes. All you have to tell people is “you can vote for more than one”.
> the result would've been electing the even-more-moderate candidate (this is San Francisco, there are no real conservatives) Joel Engardio
In SF the progressives are literal small-c conservatives, as in they love conserving things. Aaron Peskin runs the city and would do anything to stop anything new from happening.
Yeah, a lot of "progressives" in SF are all style, no substance. I think Dean Preston at least _really believes_ his own rhetoric, he's just constitutionally incapable of understanding that what he's doing isn't helping the people he purports to want to help.
I mean sure your example isn't how it plays in real life but it's just much easier to believe the people you are disagreeing eoth are really just evil liars isn't it?
If you're referring to the Center Squeeze example, I provided multiple examples of where that's happened (or come close to happening) in real-life elections. The toy example is provided just to help understand the math.
I used to live in St. Louis, a beautiful, historic city full of amazing free attractions and crippling race and class segregation. Reminds me of Tennessee taking away Memphis’ right to pursue police reform. Or the many, many Republican states that took away blue cities’ right to implement Covid mitigation measures in schools. Red states won’t be happy until everyone in blue cities are miserable (I live in a blue city in a red state). Luckily it will take a really long time for them to do that, and in the meantime these are some of the only affordable, nice places to live!
I was trying to think of an example of the reverse - a blue state trying to counter the will of a red city, but I had a hard time. I thing that it's possibly due to what you're implying (i.e., red state governments are particular in their willingness to go against the will of blue local jurisdictions), but it's also possibly (more likely, in my opinion) because there really are very few red cities in blue states. My guess is that a blue state government would just as happily overrule the will of a red city, if such a thing existed.
I think it makes more sense to focus on red areas in general, including towns and red counties, rather than cities. Since as you say there just aren't that many red cities.
There aren't as many examples of blue state overruling local *policies* of red areas, but that's largely because, on many issues, Republican voters prefer a smaller government. So often the disagreement shows up in state governments passing regulations red areas don't want, rather than overruling popular local regulations. For example, New York state banning fracking even though it is very popular in red areas of the state.
In the areas where the GOP does see an active role for local government though, blue states are certainly happy to limit their power. Probably the best examples I can think of are laws restricting the ability of local school boards to remove books from libraries, and police reform laws placing various limitations on local law enforcement that are unpopular in red areas.
I mean, this happened all the time in Covid: blue governors telling rural counties their businesses had to shut down. Of course, the cops were almost always against shutdowns, so there wasn’t enforcement, etc, etc, never want to think about that time again.
I think a more equivalent example is Democrats in DC overruling the express preferences of a state. E.g. Limiting oil and gas exploration in Alaska when the people there are strongly for it.
Yeah well to some extent everything is a fiction. Fictions can be good. But for the purpose of how our current system works, states very much exist and have legitimacy in a way that cities do not
I’m pretty sure based on the underlying game theory and simulations that approval achieves the generally desirable traits we want in single winner elections better than RCV does. RCV is actually not that good at being better than plurality voting from what I remember.
You are correct. RCV (also known as Instant Runoff, Single Transferable Vote, or Hare) has some pretty dramatic failure modes, under common circumstances. I've commented on that elsewhere in this comment thread.
My friend Ka-Ping Yee, a couple decades ago, invented a method for visualizing how election systems behave:
RCV is objectively better than plurality voting by pretty much every measure except simplicity of counting.
As for other voting systems - it depends on what you're trying to achieve. Certainly, AV is more likely to return the Condorcet winner in some circumstances, but it appears to have other negative traits. Anyway, I think it's debateable whether we should actually want the Condorcet winner to prevail in all situations.
Fairvote has some good discussion of approval voting (and other systems) disadvantages vs. RCV.
"One reason FairVote prefers ranked choice voting to approval is that with RCV, voters can freely rank back-up candidates without worrying that they will hurt their first choice."
This completely ignores the Center Squeeze phenomenon, where by ranking your true favorite first, you accidentally knock out your second choice, leading to a victory for your last choice. Which is a thing that happens _frequently_ in RCV elections. As I noted in my comment elsewhere in this thread, we had a very near miss in San Francisco a few years back, where if there had been just a couple dozen more voters backing Vilaska Nguyen, she would've knocked out Myrna Melgar, but the Melgar voters broke down in a way that would've thrown the election to Joel Engardio, whom the Nguyen voters liked even less. There are plenty of other examples of this style of run-off failure, both from non-instant runoffs (like Edwin Edwards' voters knocking out Buddy Roemer against David Duke -- luckily in that case, Edwards did win the run-off, but it was not a great situation!) and from RCV races (like Palin voters knocking out Nick Begich and throwing the race to Peltola).
Another example of a non-instant runoff failure was the French Presidential election of 2002, which Lionel Jospin _probably_ should've won, except that the left fragmented so much that the top-two were the conservative Chirac, and the fascist Le Pen. It's possible IRV would've gotten this right, although it's very hard to say -- if the lefties had consistently down-ranked Jospin, IRV still would've failed. In a PR system, what you would've ended up with is the Union for French Democracy, which was more centrist at the time, though it has since moved more to the right because its leftier elements defected, deciding who got to lead a coalition.
No voting system is perfect. And the center squeeze is a real possibility with RCV. But, personally I agree with FairVote's view that it can be a feature, not a bug. I think that a candidate which has so few first-choice supporters that they get knocked out, doesn't deserve to win the election even if they're the Condorcet winner.
Also, even if one sees Center Squeeze as a big problem, I also find their overall argument convincing that approval voting's problems are worse. And - AV has it's own spoiler failure mode too!
The idea that the person squeezed out has "so few" first choice votes is pure hand waving, and puts a massive over-emphasis on the idea of a first choice. The idea that you should get rid of the candidate who would, in a head to head race, absolutely _crush_ every other opponent, because they "only" are the first choice of _just under_ a third of the voters, rather than _just over_ a third, is absurd on its face.
The thing that ticks me off about FairVote is that they're actively dishonest. They just pretend the Center Squeeze doesn't exist. They say, "voters can freely rank back-up candidates without worrying that they will hurt their first choice," which is _obviously false_ and they know it. CES will absolutely acknowledge that of course no system is perfect, and will weigh the pros and cons. (They've come down for Approval rather than Range because of a mix of expense / simplicity / strategy reasons.) But they don't _lie_.
It seems very unlikely that repeal effort will succeed. Approval was adopted by an _overwhelming_ majority (more than two thirds) and it has remained popular since.
In 2022, the St. Louis Board of Aldermen threatened to repeal Prop D, as some felt it threatened their power. The coalition regrouped, even larger this time, to enshrine approval voting into their charter forever. Bucking the desires of the Aldermen, St. Louis voters chose to keep Prop D’s reforms in a measure called Prop R (for “Reform”) - voting 69% in favor.
I haven’t followed the most recent twists, but last I heard there had already been one attempt to repeal, and the voters smacked it down, again with over two thirds of the vote.
ETA: maybe my view would have been clearer if I had put quote-marks around “garbage state,” to indicate that I don’t endorse that characterization. Every state in the union deserves good governance, and Dems should be working for every vote of every eligible American.
(Come to think of it, if I put quote-marks around “garbage state,” that yields “”garbage state””, which isn’t quite right either.)
Both parties social-cultural are cities (D) and then rural (R). Economically though they both represent the suburbs, with R tilting more to the very rich. Suburbs either care/don’t care about cultural stuff depending on who is in power.
How is that better than RCV? Especially for primaries?
It's basically just a less expressive version of RCV where I get to put canidates into only two rankings (I prefer approved to unapproved). Yes, I know it has some nice mathematical features in certain situations but I fear that the natural result in real world non-proportion elections is that eventually people end up campaigning for you to disapprove of their close rivals.
I'm not even sure if approval voting is better than FPTP in general. Though I understand why it might be better in low information voting conditions like local elections.
Re: "It's basically just a less expressive version of RCV where I get to put canidates into only two rankings (I prefer approved to unapproved)": I think you're taking the term "RCV" a bit too narrowly/literally, by only considering the ballot itself. You also have to consider how you take all the ballots and compute a winner; per Arrow's theorem, there's no perfect way to do that. The specific flavor of RCV that's being proposed/adopted in various places has (of course) an imperfect way of doing that, that relies heavily on people's first-choice votes. Approval voting completely sidesteps that, though of course it has its own downsides.
I didn't give a very complete account of my objections but, basically, I don't think the neat math is anywhere near as important as making sure that the substantial (majority?) of voters who take the ballot instructions at face value don't feel badly done by when following the directions doesn't meaningfully effectuate their preferences -- and I think approval voting is particularly bad in that regard. My remark was about the many voters who will take the directions at face value and not strategically.
The various theorems about strategic voting are not only assuming more strategy by voters than actually occurs but neglecting the strategic behavior by the canidates that changes the overall effect not to mention mostly worried about unusual cases. While approval voting fails the voter intent test in the usual case (does approving of canidates I approve of effectuate my preferences reasonably well).
Having said that, given my choice, I'd pick a Condorcet method but I don't think it matters too much versus IRV.
--
Also it's about the worst kind of voting for a multi-position (elect 3) race as 40% of the voters (51% if opposition strategically consolidates) get to pick all 3 of their canidates while the two canidates each approved of by 30% of the voters don't get picked (obviously we want one of the canidates favored by each group). Basically it screws minorities.
You can learn a lot about the arguments for Approval Voting from the Center for Election Science, a non-partisan think-tank that has been involved with the successful campaigns to adopt Approval in Fargo and Saint Louis.
Approval is the simplest case of a ballot style called Rated / Range / Cardinal Ratings.
Arguably if you want maximum expression, a Range system is even better than a Ranked one. Because on a Ranked ballot, you just say "A > B > C". Whereas on a Range ballot, you can distinguish between "A=5, B=4, C=0" and "A=5, B=1, C=0". Those are quite different preferences, but a Ranked system would treat them the same.
Some people have tried to get a Range system enacted -- Eugene, OR recently had a referendum on adopting STAR, "Score Then Automatic Runoff", which would use a 0-5 star rating system, then take the top-two rated candidates, and infer a runoff. (Any ballot that rates A > B gets counted as one vote for A, and vice versa.)
The thing about Range is that it turns out there's a strategic incentive to try to infer who the top "front-runners" are, and then top-rate your favorite front-runner, and any minor candidate you like more. Which means in terms of strategy, Range reduces to its simplest form, Approval.
Approval also _very_ closely resembles local races that people are accustomed to, where you can "vote for up to three" in a Council race or whatever. It's just "vote for everyone you think would be a good choice". It's the cheapest, most minimal reform, and can be implemented with existing equipment. (Basically if your voting machine can collect data in vote-for-N races, you just do the settings to make N equal to the number of candidate running, for every race. Done.)
For races that let people pick multiple canidates, it's really much fairer to use some kind of system that tries not to let people's vote count twice.
For instance, if you have 40% of the population that likes canidates A, B and C and 25% that hates them and loves D and 25% that hates all those and loves E you want the 3 officials to be E, D and one of A, B or C.
Approval voting gives you A, B and C as if the minorities didn't even exist (and if it's 51% who loves A, B and C even strategic voting can't let the other 49% get any form of representation).
You can basically do a version of IRV that does better than approval here but there are better methods (IRV kinda throws out extra votes for a canidate so if you had 95% who all prefer A to B to C to D and 5% like D you should get A, B and C but pure IRV gets A, B and D... but I think some Condorcet methods habdle this kind of situation well).
"For races that let people pick multiple canidates, it's really much fairer to use some kind of system that tries not to let people's vote count twice." Yes, that's why Proportional Approval is great. It conforms to the usual "if you can command the votes of a Droop Quota, you can win a seat" property for PR, and the tabulation can be worked out on the back of a napkin, once you have the totals for each combination of approvals.
n.b.: CES usually says "Proportional Approval" to refer to a simplified tabulation method that in the academic literature is usually called Sequential Proportional Approval:
Basically folks vote, then when you start counting, you pick whoever has the highest approval; then you re-weight ballots at 1/(S+1), where S is the "satisfcation" of the ballot, i.e. how many people the ballot voted for have already been elected. Lather, rinse, repeat, until you've filled all seats.
That site mostly makes relatively unsupported/meaningless claims like "gives voters more power" because you can approve or deny every canidate" huh? That doesn't even make sense. I can approve or deny every canidate by writing it on a slip and throwing in in the trash. What matters is how it affects the actual race and it's not even clear what that means much less if it's true. And it can't decide if it assuming voters will be strategic or won't be.
But I've read elsewhere about the supposed advantages of approval voting -- hence the reference to nice mathematical properties -- but fancy results about strategic voting aren't that important in the real world where voters are more worried about the affiliative meaning of their vote (I supported the 'right' canidate). Yes, some voters will engage in a degree of strategy but only to a relatively limited degree (has to be coordinated by a campaign or online so people feel that the strategic vote is what real support of their canidate looks like) while many voters will take the vote at face value. Also, these results usually don't reflect the fact that canidates themselves are strategic actors.
So I'm not too concerned about the theoretical properties per se, though if I had my pick I'd prefer a Condorcet satisfying alternative to IRV but I don't think it matters in practice.
In practice, what I think matters is preventing the effect of spoiler canidates (by which I mean spoilers because of non-strategic voters not Arrow's theorem style unavoidable but unlikely spoilers) and not decieving/confusing voters about the meaning of their vote.
What I dislike about approval voting is that it does exactly that. Lots of voters go in and simply approve of the canidates they approve of but in practice the choice to approve of another canidate as well knocks out their preferred canidate -- not just in unusual non-transitive preference cases. Range voting has similar problems but adds on an additional issue if how should non-strategic voters even understand what the rating means.
Ultimately, most people understand a vote as a choice, ranking their preferences maps nicely onto that idea but range and approval voting do not.
I honestly don't understand how you could say Range is less appropriate for expressing preferences, when it's obviously _more_ expressive than rankings. I tend to prefer approval because to the extent people express their ratings honestly in a range system, what happens is that they improve the social optimality of the outcome, but at the expense of their personal preferences. In the long run, whoever figures out the strategy of using only top and bottom ratings derives a certain amount of advantage over the folks who are more honest. So just reverting to Approval generally keeps a fairer playing field.
There's no particular evidence from races where people have a Vote-For-N (like with many City Council races), that people have any trouble understanding that there's no point voting for somebody they like but who is woefully behind... Like, everyone votes strategically, already. Approval simplifies the strategy. Look at who seems to be in the lead, vote for the best between them, and then also for everyone you like more.
Regarding expressiveness, you are either concerned with the strategic voters or the non-strategic ones. In the first case they play out the game theory and there isn't really any question of expressiveness only the formal properties about Condorcet criteria etc.
OTOH for non-strategic voters Range voting is less expressive because what does the degree of approval express? There is no clear answer so different people will use it differently and it just ends up adding extra confusion and uncertainty.
And I deeply disagree with the idea that voting should serve as some kind of intelligence test to give the more strategic voters an advantage. Sure, you can't avoid that to some degree but I think one of the most important features of a voting system is that people who show up, read the directions and follow them reasonably effectuate their preferences.
Approval voting and Range voting horribly fail that test.
Basically you've articulated why I prefer Approval to Range.
But I don't see how you can make this an argument for RCV. The problem with RCV is that it has a massive spoiler problem, which _will_ occur with non-trivial frequency, but it's _extremely_ hard to know in advance when you're facing a center squeeze election, and should swallow your pride and just top-rank your lesser evil instead of your true favorite.
Approval, meanwhile, has _exactly the same strategy as Plurality_. You always vote for the best among the front-runners. But then you also get to approve everyone you like better.
So on the statement " deeply disagree with the idea that voting should serve as some kind of intelligence test to give the more strategic voters an advantage" -- yes, but it's RCV (at least for the single winner case) that fails that test, not Approval.
Probably worth mentioning that CA also has a measure on the ballot (Prop 34) to stop AIDS Healthcare Foundation from doing any more ballot measures in the future.
Yep, I link to that in the Politico article! But thanks for flagging here as well, seems like people in CA, activists and political leaders alike really don't like that guy.
I hate him. I think I voted for a few of his ballot measures when I was younger and didn’t know better. But once it started becoming clear how much he was investing in preventing upzoning in Hollywood, and how much his condoms-in-porn initiatives were just designed to transfer porn production to Nevada, I started telling all my friends that if you shop at Out of the Closet then you are supporting keeping housing expensive.
It is a great name for a secondhand clothing store whose profits supposedly support gay causes. It’s a shame that AIDS Healthcare Foundation is really a nimby organization masquerading as a gay health cause.
Maybe some canny politician can add a rider barring the evergreen dialysis clinic measures too, that'd fill two stones with one bird..."brought to you by the Paperwork Reduction Act!"
If it's really written that way -- this one group or guy can't do any more ballot measures -- I fear it might be ruled unconstitutional. Generally the courts have frowned on picking out certain groups - or people - and denying them electoral rights and soliciting ballot measures sounds like that.
Yes, that is a risk to the proposition. The language doesn't specifically mention AHF, but it's written such that AHF is clearly the only organization who it actually applies to. The CA Supreme Court already ruled the measure can go forward, but they reserved the right to look into it more after the election if it wins: https://www.sfchronicle.com/politics/article/aids-healthcare-ballot-measure-19578138.php
Local angle: Minneapolis adopted RCV starting with the 2009 election with all the sorts of promises Ben outlines above being made about how it would fix everything. I think everyone who lives here would have to agree it did not in fact "fix politics" in the city of lakes.
The CA rent control initiative is a great reminder that government by ballot initiative is very much *not* about "giving power to the people" instead it empowers special interest groups, the ballot initiative industry itself, and rich weirdos like Weinstein. Which is a far cry from the idea of representative democracy. I'd go further and say that a lot of California problems at their heart are caused by this system itself. A big part of the reason there's such mass incarceration in CA is the awful "three strikes" law (and here's another one that will make it worse). Likewise a big part of why middle class families are increasingly being forced out of the state and it's becoming a land of haves and have nots is Prop 13 and it's long term impacts.
But hey, you get to vote on ferrets and adult film stars using condoms so there's that I guess.
On the one hand, the risk of pregnancy is very, very low. On the other hand, the risk of freaky zoonotic infections is high enough that I would certainly advise my ferret to wear the condom.
"I would certainly advise my ferret to wear the condom"
And on the whole, he's been a good client for the agency. When he first broke in, he was doing too many low budget quickies -- "Ferrets Gone Wild," "Pop Goes the Weasel," "Two Girls, One Ferret," that sort of thing. Beneath him, really. But he's hoping to go legit some day, and I have persuaded him to do more cinematic stuff, things that show his comedic range, like "Ferret Bueller's Day Off," and even cast him as a traditional romantic lead, like "A Ferret to Remember."
Prop 13 has nothing to do with California's failure to build more housing. Prop 13 is the only reason a middle class exists at all in California, because before it was passed, in the period from 1975-1978, property taxes were doubling every six months, forcing people to sell their homes to pay their property taxes. It constantly blows me away when I see people advocating for returning to a system that taxed people out of their homes as an effective means to "build the middle class."
Counterpoint: prop 13 is bad. There's a whole giant literature about this so I won't get into it but basically it's a Yuge tax subsidy (ie welfare) for largely wealthier, older, white home owners that massively undermined the ability to create functional and equitable local government services for everyone else.
And the "taxed people out of their homes" thing was largely bogus, requiring people to sell a very valuable asset to pay taxes on that asset wasn't that common and actually makes a lot of sense, if you have a 2 million dollar home you're rich! Likewise there are a million better ways to deal with this relatively rare scenario that "Lets make white old home owners richer and gut the public education system at the same time"
So the voters should have voted for political candidates who promised to stop raising property taxes, and then evaluated them on that promise. This is how a functional democracy works. The only system we're advocating for is being a democratic republic
Yeah I think some of the claims RCV supporters make are a bit starry eyed. But at the core, they're right that's it's just an objectively better system, and if rolled out on a national scale it could make a significant positive difference to our politics, especially if combined with proportional representation.
As for ballot initiatives - you can argue three strikes is a bad policy, but I think it pretty clearly passed because it was popular at the time, not because "interest groups" wanted it. Certainly, in some cases a policy will be popular that will lead to unpopular outcomes, which is the argument against ballot initiatives and in favor of representative democracy. The idea being that representatives will try to deliver the outcome the majority of the population wants while reigning in its worst instincts.
However, sometimes representative democracy does fail to represent people on certain issues, mainly because they're voting on others that are more salient. Representative democracy failed to represent the majority of Californians on affirmative action, the majority of Missourians on union policy, the majority of Kansans on abortion, the majority of Floridians on minimum wage, and the majority of Idahoans on Medicaid expansion. The argument in favor of ballot initiatives is that they allow people to have their cake and eat it too - "fine tuning" democracy. They allow the majority to elect representatives on the highest salience issues but still get their way on other issues too.
Personally, I think my ideal state constitution would ban ballot initiatives for issues that could affect housing supply, on the grounds that they'd affect future residents who can't vote, but allow them for everything else.
Yeah lol I considered when writing that it may be a total hack-ish view that I came up with because I thought it would produce the results I wanted. But I do think the rationale of it affecting people who can't vote is a decent one.
RCV is _mostly_ an improvement over Plurality, except for the small-but-nontrivial percentage of the time where it fails so spectacularly that people repeal it and it poisons the well for reform for a generation, like happened in Burlington, VT.
(This paper notes that you can use kind of an opposite method to Instant Runoff, where instead of eliminating from the top, you repeatedly look at an inferred runoff between the _bottom_ two candidates, and eliminate the loser of that comparison. This method works out to be a Condorcet method, where if any candidate would win in every pairwise comparison, they will be elected.)
It's not like other systems have much of a better track record. St Louis, one of the few places which tried approval voting has a majority of aldermen in support of repealing it (if the state ballot initiative doesn't do it first). There are also non-governmental organizations that have tried approval voting and basically abandoned it after it devolved into plurality voting.
By contrast, Australia has been using RCV to elect their for over a century and, while a proportional system would of course be better, they seem to be pretty happy with it. By contrast, trying to find a country of that size, or any size at all, which has actually implemented approval voting, Condorcet, range voting, STAR, or any of the systems that online nerds like? Crickets...
#1 on my list: The Baltimore Baby Bonus, which would provide $1,000 to parents when they have a child. It got the signatures to qualify, but the mayor has filed a lawsuit to remove it from the ballot. It'd be the first baby bonus ballot measure in history--I think also the first cash assistance ballot measure. The Circuit Court will hear the case on August 6th. You can see more at https://baltimorebabybonus.org
Local governments are budget-constrained. This is a small step that can offset some of the upfront costs like strollers, cribs, and diapers. We estimate it would lift 50 babies out of poverty each year, and reduce the severity of poverty for hundreds more, yielding long-term benefits for child development by intervening in the most formative years.
That's.. technically true, but it makes me think worse of you for weasel-wording, frankly. That family will be """out of poverty""" for 10 minutes, then back into poverty again. It's nice to give people a lift, but selling the program with that phrasing is suspicious.
Trying to define something like poverty is always going to seem a bit subjective and all the campaign can really do is use the agreed upon definitions to make their case.
If you read their FAQ page(and Max's other comment) you see they are making the claim that this will lift 50 children out of poverty out of about 7,000 births per year.
They seem to be trying to provide a pretty concrete view of the benefits, in a way that might seem to minimize them as 50 children might seem like a rather small number in the context of an entire city.
If they wanted to be less forthcoming they could have just more nebulously referred to the evidences of the benefits of cash transfers on crime, healthcare costs, CPS visits etc.
I've advised the campaign on policy matters and canvassed a couple times for it, but don't have a formal relationship with them.
I don't think the council was interested in a legislative approach. Yes, cities are constrained on generous transfers, but limiting to babies cuts the cost, in Baltimore's case to about 1/300th of the city budget.
That's at least the efficient way to incentivize having children. Research suggests that very expensive tax breaks do little to incentivize having kids -- and imo are deeply unfair when given to wealthy people (another vacation) with kids as if their life choices were more worthwhile than people who don't have kids -- but money given at birth works well.
Probably makes sense. It's hard to plan too far into the future but there are alot of costs and time involved in having a child that can encourage people to delay.
One hecto-War-and-Piece is therefore about 33 kilo-NYT-articles. We're one hecto-day from the election, so you'd need about 330 NYT-length articles published each day to match this pace. Most US newspapers these days are Gannett shells that just reproduce the national wire, and not all foreign newspapers will be breathlessly attuned to US politics, so many will have coverage a literal fraction of the NYT.
I don't think you can scrounge more than 60 newspaper-equivalents worldwide (citation: just making up numbers using my gut; probably an overestimate). If that's right, then 1 NYT-equivalent would need to publish about 5.5 articles a day on the election to fill out your claim. 5.5 articles is possible, given (say) how breathlessly Clinton's emails got covered, but it still seems a little high. If my estimate of 60 newspaper-equivalents is too high, then each NYT-equivalent would need to produce more articles than that, which is a bridge too far for me.
You're overlooking all the book authors, independent journalists/pundits/bloggers, social media commenters, student essayists, letters-to-the-editor writers, foreign government officials, corporate strategists, and others whose work is included in Matt's passive-voiced "will be written."
Also WTF was the thinking behind using all the money from the cap and trade program to jobs and etc programs to fight climate change?!? The cap and trade program *is* the best thing you could do so take the money and give it back to people with checks to politically insulate it or something else popular. If you think you have political capital to spend on the climate just make the permits more pricey (if you can send people more money that makes it easier).
I can see the policy logic. The more you raise in the short term, the more renewable energy you install, the less you have to spend on permits in the future. But if course that doesn't work if it's repealed.
They had one that looked more like this, in 2016 they proposed a carbon tax and used the proceeds to decrease the sales tax, but the Groups on the left killed it because the proceeds were going to everyone and not to “communities of color” or whatever.
I don't understand how people can be confused by RCV. It's literally, "rank the following x candidates in order from 1-x according to how qualified you think they are to hold the office of (insert office here)."
I do understand why both major political parties would hate any method of voting that increases their opponents' chances of getting elected.
Concerns about comprehensibility are presumably not about how individual voters enter their votes, but about how the scores are aggregated: Borda-count, Dowdall system, tournament-style? There are options, many of which can be called, accurately, “RCV”.
In some of them, the person with the most first-place votes does not always win. I’m not saying that’s a bad thing, but it’s a thing that will need explaining.
Yeah. There are several politicians who have pretended to be against RCV because they believe minority communities don't understand it. I cannot think of anything more racist. Literally rank your choices in order. A 5 year old could understand it.
The majority of voters do not have opinions on multiple political candidates and are unable to distinguish them or their policy platforms from each other. Because the majority of voters do not think about politics, do not care about politics unless it directly impacts them, and are basically voting for who they'd rather have a beer with. They are unable to rank multiple candidates because they have no clue who they are or what the difference is between them. We are in the same country where over 40% of the population doesn't know how many Senators a state has.
I mean, do you love cricket? (Let's assume no). Could you rank the 6 greatest wicket-keepers of all time, in order? Probably not, right? It's not because you can't rank things, but because you have no clue who or what a wicket-keeper is, can't name any of the major players, and basically don't know because you don't care. Right? Well that's quite close to how voters feel about politicians.
You are probably in a bubble of highly-educated political obsessives who could rank multiple candidates. But the average person would think you're a politics-obsessed weirdo, and so cannot. Make sense?
MS Dhoni and Ian Healy for being ferociously quick around the wicket, keeping for tremendous spin bowlers. Sangakkara was pretty special, too.
A guy named Jack Russell (unforgettable name!), for England, who played before my time following cricket, who's still famous for playing up to the wicket for fast bowlers.
If you want to include batting, then you're putting Gilchrist and Quentin de Cock into the mix.
If leadership counts for something, then Ricky Ponting a deserves a place on the list. A cut (but only just) below GOAT with both gloves and bat, but one of the best captains qua captains ever.
Is that six? I've not really ranked them, and don't want to. My top-of-mind list also over-values the last several generations of players, because that's who I've seen. I'm sure there were tremendous keepers back in the days of uncovered wickets that no one ever thinks of anymore.
Nope it's very much the opposite. My experience with this is from actually talking to voters during the last mayoral election in NYC a ranked choice election. Voters have very little trouble understanding the concept of a second choice. Even minorities can figure it out. You're just way over thinking it.
50% of Maine voters only 'ranked' 1 candidate in Maine's last RCV election (which was their 4th BTW). Seems like a more relevant factoid than your anecdote.
You're probably in a bubble of politics-obsessed weirdos, but regular people do not know more than 1 candidate at a time, or what their exact policy platforms are
You keep talking about me being in a bubble but you know nothing about me and I'm telling you about my experience talking to actual voters when canvassing. It seems to me that your ideological priors are the thing that are at issue here. It's really not that hard to rank people. I've had dozens if not hundreds of conversations with regular old normies about this. Who do you like for mayor? Oh who would your second choice be? Not complicated. Everyone gets it. Totally made up concern
People don't understand the complicated weighting of the votes after they are cast, not 'how to fill out the ballot'.
Doubling back to say that I don't think I could do a good job explaining the weighting system to a neighbor, and I am reasonably well educated. Assuming that minority communities, who might be less well educated and less engaged with the minutae of newfangled algebraic voting methods, might not understand it, seems not racist at all and actually a reason to do more work to educate people.
When canvassing in NYC in the last election this rarely came up. Although the New York system isn't algebraic and only takes first place votes so it's relatively easy to understand.
I’ve never understood why YIMBY advocates don’t take more advantage of trying to pass their own statewide ballot measures. Seems perfect for the their coalition: Takes the power away from local NIMBY’s with lots of free time, and gives it anyone in the state who votes.
Have there just been a lot of YIMBY initiatives that failed? Or is it just seen as likely to fail? Or have there been a lot of housing referendum I just don’t know about? I am particularly curious why this hasn’t happened more in CA
It is beyond me how anyone can support RCV after the "stolen election" idiocy of 2020. Whatever it's merits, too many people are incapable of understanding how the counting works. Can you imagine what would have happened if a bunch of swing states had RCV and Trump had been leading after round one? Given that there are alternatives that are easy to understand (eg approval voting), there is no reason to support RCV (yes, I know that RCV has many advantages. But the fact that RCV increases the likelihood that results will been seen as illegitimate trumps all of them).
I have read plausible arguments that RCV doesn't even achieve its goal of leading to more reasonable, moderate, electable (in the case of primaries) candidates. It may encourage voters to vote for their extreme preferences (knowing they can still rank someone more reasonable below that preference) without thinking about electability first. And then the extreme candidates end up winning.
My current rule for voting on ballot measures in CA is to vote no on any tax increase or new bonds to increase spending. This is a very poorly run state that blows money on all kinds of ineffective programs. Some states have no income taxes or no sales taxes but not CA. It wouldn’t hurt for this state to turn a little less blue. Newsom has very little chance of becoming President and he’s making things worse by signing dumb, irresponsible laws that allow schools to stop reporting transitions to parents.
"Newsom has very little chance of becoming President" I think this claim is pretty wrong. The lesson everyone should have learned from 2016 is literally anyone, even a know nothing game show host, can become president if they can win a major party nomination.
If Newsom can do that he'd have a decent shot depending on the fundamentals of the election cycle.
Exactly, if the nominees where Newsom and VP Vance in 2028 and the economy was in recession Newsom would easily win despite people screaming "But what about the French Laundry!?!?!?!"
Cat women aren't particularly Democratic, while dog men are more Republican, men who have cats are more likely to be Democrats than women who have cats, and no-pets women are the most Democratic group
It's not useful without knowing how many cats the individuals own though - the term, "cat lady" is applied to owners of multiple cats, not just one cat. (That said, "men who have cats are more likely to be Democrats than women who have cats" is probably one of the least surprising statistics I've ever seen.)
Sadly the GSS data from 2018 only has data about about 100 women with cats and no children so more there's a lot more to learn here. But yes there are tons of women with cats who vote for Trump/GOP out there so Vance's stereotype of liberal misanthropic cat women is well, dumb.
Don't sleep on the Oklahoma Minimum Wage Increase Initiative which may appear on the November ballot. This deep red state has used I&R to bypass its Republican supermajorities in the state legislature to legalize medical marijuana, achieve criminal justice reform, and to expand Medicaid. Will the voters do it again? I'm hopeful.
One thing I’ll note about the Missouri ballot initiative: it doesn’t just get rid of RCV. It also requires primaries only put one candidate forward. The reason that’s important is because St. Louis City has already implemented what is, in my view, a superior alternative to RCV with an approval primary followed by a runoff. The legislature wants to get rid of all of these alternative voting forms because I suppose they like extreme candidates plus they hate whenever a major city does something nice for itself.
I am deeply frustrated with all the money going into RCV, which in its single-winner form just does not do what its advocates want it to do. The big org that promotes it, FairVote, will tell you that RCV gets rid of the spoiler problem. You can vote for your true favorite, and then second-rank your "lesser evil", and everything will be great. This is, generously, only true as long as your true favorite is irrelevant. To be more direct, they know it's false, and I do not understand why they keep saying it. In fact, the spoiler problem in RCV is _worse_ than in our current system.
Suppose you have a town with 100 voters. It's election day, the polls are going to close in five minutes. So far our vote tally looks like this:
35 Right > Center > Left
16 Center > Right > Left
16 Center > Left > Right
31 Left > Center > Right
Right now if you did the RCV tabulation, you'd find that Right has 35 first-choice votes, Center has 32, and Left has 31. So Left is eliminated, those votes go to Center, and Center wins with about 2/3 of the votes. Notably, Center is preferred by ~2/3 of voters over either alternative, so in some sense they're obviously the "correct" winner.
But note also that there are only 98 votes! The final two voters just make it to the polls. They vote with the Left faction. Now what happens? Well, Left has 33 to Center's 32 in the first round. Center is eliminated. Center's votes split equally... leading to a 51-49 victory for Right.
Not only is this stupid because Center was still preferred by 2/3 of voters over either alternative. The people who changed the result were Lefties! They would've been happier if they had just stayed home! This is no way to run an election. At least under Plurality, when you vote third party, you can't actively move the results towards a party you hate.
Center squeeze elections are shockingly common.
Possibly the most famous runoff failure in history (which was not an instant runoff, but would surely have delivered the same result if it were) was the David Duke race. Former Democrat / moderate Republican Buddy Roemer got squeezed out in the first round between the prolifically corrupt Democratic machine politician Edwin Edwards, and the Klansman David Duke.
The recent election of Mary Peltola in Alaska is also obviously a center squeeze. Nick Begich, the anti-Trump Republican, should have won, but was squeezed out between Palin and Peltola. (I'm obviously happier with Peltola, but it's still not the right result.) As noted in Ben's post, there's a movement of Alaskans to try to repeal it.
Burlington, VT adopted RCV, immediately had a Center Squeeze election, and then repealed it. (There are now some rumblings that they might adopt it again, because apparently nobody learned anything from that experience, and getting people to try anything that seems new is incredibly difficult.)
San Francisco also had a very near miss with a squeeze a few years ago, where the progressive Vilaska Nguyen came within a couple dozen votes of knocking out the less-progressive / more-moderate Myrna Melgar, but if she had, the result would've been electing the even-more-moderate candidate (this is San Francisco, there are no real conservatives) Joel Engardio. And just in general, SF politics are insanely polarized and nasty.
Portland, OR, and the small city of Albany, CA, have adopted the _proportional_ version of RCV, which is much better. It technically still has the center-squeeze issue, but it only applies to the final seat selected, and at that point you're talking about very close margins where you might as well have a coin toss anyways. The Australian Senate, IIRC, also uses STV-PR.
But really if you want an election system that reduces polarization; allows for factions of existing parties to develop into actual distinct new parties; and encourages more, collegial campaigning; then you want Approval ballots, where you treat each candidate like a ballot proposition. "Do you like X? Yes/No" This forces every candidate to court support from people whose first choice is some "ideological neighbor". Going negative may hurt your target, but it also turns off people who otherwise might give you a secondary vote. We have documentary evidence that this works, from Fargo and (as you noted) Saint Louis, which have both adopted Approval within the last few years (though still with single-winner elections).
Approval also has the advantage of being far cheaper than a ranked ballot, both in terms of direct printing costs, and voter education. Think about the giant grid you'd have to print if there are 30 candidates for a seat, and folks have to rank them, potentially all the way from first to thirtieth. Additionally, that kind of grid causes much higher ballot spoilage rates -- anywhere from 3x to 7x, if you look at situations where there are ranked and not-ranked races within the same election. Intuitively it's obvious why -- if you mark two candidates as both First, IRV just trashes your whole ballot. This higher spoilage rate is of course concentrated on voters who are less educated, less wealthy, or less English fluent. With Approval, or a Range / Cardinal Ratings system like STAR (Score Then Automatic Runoff) where you rate each candidate independently (zero to five stars, in that case), screwing up a mark on one candidate doesn't void your ratings for the other candidates. (Approval can be thought of as the simplest case of a Range ballot -- you're just rating each candidate as 0 or 1.)
There are also good ways to do proportional tabulation with Approval or Range ballots. In fact the proportional method for Approval was actually invented independently by a French mathematician, D'Hondt, and in the US by Thomas Jefferson. It was used for the initial allocation of House seats among the states.
I really wish some group of rich nerds would invest in giving the Center for Election Science the resources it would need to do national-scale voter education, and run ballot measures in various states.
I actually have been talking with a couple Councilmembers in the city to my immediate south about adopting a city charter that would include using Proportional Approval. And I ran for City Council in my own town (and got 45% against a multi-term incumbent, so not bad for a first try). I will definitely be pursuing that as a priority if I ever get on.
Proportional Representation really is the holy grail here. I've talked with Andrew Yang about this, and while, to my frustration, he has been pushing RCV just because people are more familiar with it and he feels it's an easier improvement to get to quickly, he totally agreed that STV-PR is _way_ better than just RCV for single winner. There's a bill in Congress that would change to STV-PR, on multi-member districts, in states that have more than one House member: https://www.congress.gov/bill/117th-congress/house-bill/3863/text
A PR system would allow sane anti-Trump Republicans to join Yang's Forward Party and make it a real force. It would let Working Families absorb Democrats' DSA wing. Maybe we'd see a functional Green Party. And more generally, it would mean you wouldn't have the thirty percent of "blue tribe" people in rural areas, and the thirty percent of "red tribe" people in cities, feeling completely alienated from the system. No large minority should feel like they have no voice in their government at all. That's a huge factor in the "LOL nothing matters" attitude towards politics that got us Trump.
Approval voting is the way imo, it’s super easy to understand and has the nice property that it’s basically impossible for people to invalidate their ballots by checking the wrong boxes. All you have to tell people is “you can vote for more than one”.
> the result would've been electing the even-more-moderate candidate (this is San Francisco, there are no real conservatives) Joel Engardio
In SF the progressives are literal small-c conservatives, as in they love conserving things. Aaron Peskin runs the city and would do anything to stop anything new from happening.
Yeah, a lot of "progressives" in SF are all style, no substance. I think Dean Preston at least _really believes_ his own rhetoric, he's just constitutionally incapable of understanding that what he's doing isn't helping the people he purports to want to help.
I mean sure your example isn't how it plays in real life but it's just much easier to believe the people you are disagreeing eoth are really just evil liars isn't it?
If you're referring to the Center Squeeze example, I provided multiple examples of where that's happened (or come close to happening) in real-life elections. The toy example is provided just to help understand the math.
I used to live in St. Louis, a beautiful, historic city full of amazing free attractions and crippling race and class segregation. Reminds me of Tennessee taking away Memphis’ right to pursue police reform. Or the many, many Republican states that took away blue cities’ right to implement Covid mitigation measures in schools. Red states won’t be happy until everyone in blue cities are miserable (I live in a blue city in a red state). Luckily it will take a really long time for them to do that, and in the meantime these are some of the only affordable, nice places to live!
I was trying to think of an example of the reverse - a blue state trying to counter the will of a red city, but I had a hard time. I thing that it's possibly due to what you're implying (i.e., red state governments are particular in their willingness to go against the will of blue local jurisdictions), but it's also possibly (more likely, in my opinion) because there really are very few red cities in blue states. My guess is that a blue state government would just as happily overrule the will of a red city, if such a thing existed.
I think it makes more sense to focus on red areas in general, including towns and red counties, rather than cities. Since as you say there just aren't that many red cities.
There aren't as many examples of blue state overruling local *policies* of red areas, but that's largely because, on many issues, Republican voters prefer a smaller government. So often the disagreement shows up in state governments passing regulations red areas don't want, rather than overruling popular local regulations. For example, New York state banning fracking even though it is very popular in red areas of the state.
In the areas where the GOP does see an active role for local government though, blue states are certainly happy to limit their power. Probably the best examples I can think of are laws restricting the ability of local school boards to remove books from libraries, and police reform laws placing various limitations on local law enforcement that are unpopular in red areas.
Isn’t the classic SB example blue states overruling red suburbs’ anti-building zoning?
Or various gun control laws, see e.g. https://www.npr.org/2019/02/21/696400737/when-sheriffs-wont-enforce-the-law
I mean, this happened all the time in Covid: blue governors telling rural counties their businesses had to shut down. Of course, the cops were almost always against shutdowns, so there wasn’t enforcement, etc, etc, never want to think about that time again.
Isn’t the Washington State climate initiative in the article an example?
I think a more equivalent example is Democrats in DC overruling the express preferences of a state. E.g. Limiting oil and gas exploration in Alaska when the people there are strongly for it.
California has banned some red school districts from enacting their preferred gender policies.
Cities are just fictions that state governments allow to exist. You can't fix that. We just need better state governments.
Yeah well to some extent everything is a fiction. Fictions can be good. But for the purpose of how our current system works, states very much exist and have legitimacy in a way that cities do not
I don't know what you're talking about tbh
It sounded like St Louis already had serious effort underway to repeal approval voting at the local level though?
https://www.stltoday.com/news/local/government-politics/effort-underway-to-repeal-approval-voting-in-st-louis-replace-it-with-new-system/article_2c3bad65-1e46-58b6-8b9f-1d7f49d0aaeb.html
To me approval voting seems like a much more flawed system than RCV, and evidently I'm not the only one, because no actual country uses it.
I’m pretty sure based on the underlying game theory and simulations that approval achieves the generally desirable traits we want in single winner elections better than RCV does. RCV is actually not that good at being better than plurality voting from what I remember.
You are correct. RCV (also known as Instant Runoff, Single Transferable Vote, or Hare) has some pretty dramatic failure modes, under common circumstances. I've commented on that elsewhere in this comment thread.
My friend Ka-Ping Yee, a couple decades ago, invented a method for visualizing how election systems behave:
https://rangevoting.org/IEVS/Pictures.html
https://www.youtube.com/watch?v=7btAd1HYvjU
RCV is _f***ing weird_.
RCV is objectively better than plurality voting by pretty much every measure except simplicity of counting.
As for other voting systems - it depends on what you're trying to achieve. Certainly, AV is more likely to return the Condorcet winner in some circumstances, but it appears to have other negative traits. Anyway, I think it's debateable whether we should actually want the Condorcet winner to prevail in all situations.
Fairvote has some good discussion of approval voting (and other systems) disadvantages vs. RCV.
https://archive.fairvote.org/?page=1920
https://fairvote.org/new_lessons_from_problems_with_approval_voting_in_practice/
https://fairvote.org/resources/electoral-systems/comparing-voting-methods/
FairVote sure does lie a lot.
"One reason FairVote prefers ranked choice voting to approval is that with RCV, voters can freely rank back-up candidates without worrying that they will hurt their first choice."
This completely ignores the Center Squeeze phenomenon, where by ranking your true favorite first, you accidentally knock out your second choice, leading to a victory for your last choice. Which is a thing that happens _frequently_ in RCV elections. As I noted in my comment elsewhere in this thread, we had a very near miss in San Francisco a few years back, where if there had been just a couple dozen more voters backing Vilaska Nguyen, she would've knocked out Myrna Melgar, but the Melgar voters broke down in a way that would've thrown the election to Joel Engardio, whom the Nguyen voters liked even less. There are plenty of other examples of this style of run-off failure, both from non-instant runoffs (like Edwin Edwards' voters knocking out Buddy Roemer against David Duke -- luckily in that case, Edwards did win the run-off, but it was not a great situation!) and from RCV races (like Palin voters knocking out Nick Begich and throwing the race to Peltola).
https://electionscience.org/education/approval-voting-vs-rcv
Another example of a non-instant runoff failure was the French Presidential election of 2002, which Lionel Jospin _probably_ should've won, except that the left fragmented so much that the top-two were the conservative Chirac, and the fascist Le Pen. It's possible IRV would've gotten this right, although it's very hard to say -- if the lefties had consistently down-ranked Jospin, IRV still would've failed. In a PR system, what you would've ended up with is the Union for French Democracy, which was more centrist at the time, though it has since moved more to the right because its leftier elements defected, deciding who got to lead a coalition.
No voting system is perfect. And the center squeeze is a real possibility with RCV. But, personally I agree with FairVote's view that it can be a feature, not a bug. I think that a candidate which has so few first-choice supporters that they get knocked out, doesn't deserve to win the election even if they're the Condorcet winner.
Also, even if one sees Center Squeeze as a big problem, I also find their overall argument convincing that approval voting's problems are worse. And - AV has it's own spoiler failure mode too!
https://archive.fairvote.org/?page=2474#question17
The idea that the person squeezed out has "so few" first choice votes is pure hand waving, and puts a massive over-emphasis on the idea of a first choice. The idea that you should get rid of the candidate who would, in a head to head race, absolutely _crush_ every other opponent, because they "only" are the first choice of _just under_ a third of the voters, rather than _just over_ a third, is absurd on its face.
The thing that ticks me off about FairVote is that they're actively dishonest. They just pretend the Center Squeeze doesn't exist. They say, "voters can freely rank back-up candidates without worrying that they will hurt their first choice," which is _obviously false_ and they know it. CES will absolutely acknowledge that of course no system is perfect, and will weigh the pros and cons. (They've come down for Approval rather than Range because of a mix of expense / simplicity / strategy reasons.) But they don't _lie_.
It seems very unlikely that repeal effort will succeed. Approval was adopted by an _overwhelming_ majority (more than two thirds) and it has remained popular since.
https://ballotpedia.org/St._Louis,_Missouri,_Proposition_D,_Approval_Voting_Initiative_(November_2020)
https://electionscience.org/education/st-louis-success
In 2022, the St. Louis Board of Aldermen threatened to repeal Prop D, as some felt it threatened their power. The coalition regrouped, even larger this time, to enshrine approval voting into their charter forever. Bucking the desires of the Aldermen, St. Louis voters chose to keep Prop D’s reforms in a measure called Prop R (for “Reform”) - voting 69% in favor.
Reports are that at least 9 out of the 14 St Louis alderman supported repealing Approval Voting:
https://www.stltoday.com/news/local/government-politics/effort-underway-to-repeal-approval-voting-in-st-louis-replace-it-with-new-system/article_2c3bad65-1e46-58b6-8b9f-1d7f49d0aaeb.html
That’s paywalled.
I haven’t followed the most recent twists, but last I heard there had already been one attempt to repeal, and the voters smacked it down, again with over two thirds of the vote.
Garbage state
Every garbage state in the union has a lot of good people trapped in it.
Yup. Even New Jersey.
"Yup. Even New Jersey."
You're really pushing the edge-cases, but yes: even New Jersey.
Ken, I think we’re in agreement here?
ETA: maybe my view would have been clearer if I had put quote-marks around “garbage state,” to indicate that I don’t endorse that characterization. Every state in the union deserves good governance, and Dems should be working for every vote of every eligible American.
(Come to think of it, if I put quote-marks around “garbage state,” that yields “”garbage state””, which isn’t quite right either.)
"What's the difference between a shithole country and a garbage state?
One is poor, the other is deplorable.
Nobody's covering themselves in glory, and there are a lot of feedback loops here.
Both parties social-cultural are cities (D) and then rural (R). Economically though they both represent the suburbs, with R tilting more to the very rich. Suburbs either care/don’t care about cultural stuff depending on who is in power.
How is that better than RCV? Especially for primaries?
It's basically just a less expressive version of RCV where I get to put canidates into only two rankings (I prefer approved to unapproved). Yes, I know it has some nice mathematical features in certain situations but I fear that the natural result in real world non-proportion elections is that eventually people end up campaigning for you to disapprove of their close rivals.
I'm not even sure if approval voting is better than FPTP in general. Though I understand why it might be better in low information voting conditions like local elections.
Re: "It's basically just a less expressive version of RCV where I get to put canidates into only two rankings (I prefer approved to unapproved)": I think you're taking the term "RCV" a bit too narrowly/literally, by only considering the ballot itself. You also have to consider how you take all the ballots and compute a winner; per Arrow's theorem, there's no perfect way to do that. The specific flavor of RCV that's being proposed/adopted in various places has (of course) an imperfect way of doing that, that relies heavily on people's first-choice votes. Approval voting completely sidesteps that, though of course it has its own downsides.
I didn't give a very complete account of my objections but, basically, I don't think the neat math is anywhere near as important as making sure that the substantial (majority?) of voters who take the ballot instructions at face value don't feel badly done by when following the directions doesn't meaningfully effectuate their preferences -- and I think approval voting is particularly bad in that regard. My remark was about the many voters who will take the directions at face value and not strategically.
The various theorems about strategic voting are not only assuming more strategy by voters than actually occurs but neglecting the strategic behavior by the canidates that changes the overall effect not to mention mostly worried about unusual cases. While approval voting fails the voter intent test in the usual case (does approving of canidates I approve of effectuate my preferences reasonably well).
Having said that, given my choice, I'd pick a Condorcet method but I don't think it matters too much versus IRV.
--
Also it's about the worst kind of voting for a multi-position (elect 3) race as 40% of the voters (51% if opposition strategically consolidates) get to pick all 3 of their canidates while the two canidates each approved of by 30% of the voters don't get picked (obviously we want one of the canidates favored by each group). Basically it screws minorities.
There are much better systems for that.
You can learn a lot about the arguments for Approval Voting from the Center for Election Science, a non-partisan think-tank that has been involved with the successful campaigns to adopt Approval in Fargo and Saint Louis.
https://electionscience.org/education
Approval is the simplest case of a ballot style called Rated / Range / Cardinal Ratings.
Arguably if you want maximum expression, a Range system is even better than a Ranked one. Because on a Ranked ballot, you just say "A > B > C". Whereas on a Range ballot, you can distinguish between "A=5, B=4, C=0" and "A=5, B=1, C=0". Those are quite different preferences, but a Ranked system would treat them the same.
Some people have tried to get a Range system enacted -- Eugene, OR recently had a referendum on adopting STAR, "Score Then Automatic Runoff", which would use a 0-5 star rating system, then take the top-two rated candidates, and infer a runoff. (Any ballot that rates A > B gets counted as one vote for A, and vice versa.)
The thing about Range is that it turns out there's a strategic incentive to try to infer who the top "front-runners" are, and then top-rate your favorite front-runner, and any minor candidate you like more. Which means in terms of strategy, Range reduces to its simplest form, Approval.
Approval also _very_ closely resembles local races that people are accustomed to, where you can "vote for up to three" in a Council race or whatever. It's just "vote for everyone you think would be a good choice". It's the cheapest, most minimal reform, and can be implemented with existing equipment. (Basically if your voting machine can collect data in vote-for-N races, you just do the settings to make N equal to the number of candidate running, for every race. Done.)
For races that let people pick multiple canidates, it's really much fairer to use some kind of system that tries not to let people's vote count twice.
For instance, if you have 40% of the population that likes canidates A, B and C and 25% that hates them and loves D and 25% that hates all those and loves E you want the 3 officials to be E, D and one of A, B or C.
Approval voting gives you A, B and C as if the minorities didn't even exist (and if it's 51% who loves A, B and C even strategic voting can't let the other 49% get any form of representation).
You can basically do a version of IRV that does better than approval here but there are better methods (IRV kinda throws out extra votes for a canidate so if you had 95% who all prefer A to B to C to D and 5% like D you should get A, B and C but pure IRV gets A, B and D... but I think some Condorcet methods habdle this kind of situation well).
"For races that let people pick multiple canidates, it's really much fairer to use some kind of system that tries not to let people's vote count twice." Yes, that's why Proportional Approval is great. It conforms to the usual "if you can command the votes of a Droop Quota, you can win a seat" property for PR, and the tabulation can be worked out on the back of a napkin, once you have the totals for each combination of approvals.
n.b.: CES usually says "Proportional Approval" to refer to a simplified tabulation method that in the academic literature is usually called Sequential Proportional Approval:
https://en.wikipedia.org/wiki/Sequential_proportional_approval_voting
Basically folks vote, then when you start counting, you pick whoever has the highest approval; then you re-weight ballots at 1/(S+1), where S is the "satisfcation" of the ballot, i.e. how many people the ballot voted for have already been elected. Lather, rinse, repeat, until you've filled all seats.
That site mostly makes relatively unsupported/meaningless claims like "gives voters more power" because you can approve or deny every canidate" huh? That doesn't even make sense. I can approve or deny every canidate by writing it on a slip and throwing in in the trash. What matters is how it affects the actual race and it's not even clear what that means much less if it's true. And it can't decide if it assuming voters will be strategic or won't be.
But I've read elsewhere about the supposed advantages of approval voting -- hence the reference to nice mathematical properties -- but fancy results about strategic voting aren't that important in the real world where voters are more worried about the affiliative meaning of their vote (I supported the 'right' canidate). Yes, some voters will engage in a degree of strategy but only to a relatively limited degree (has to be coordinated by a campaign or online so people feel that the strategic vote is what real support of their canidate looks like) while many voters will take the vote at face value. Also, these results usually don't reflect the fact that canidates themselves are strategic actors.
So I'm not too concerned about the theoretical properties per se, though if I had my pick I'd prefer a Condorcet satisfying alternative to IRV but I don't think it matters in practice.
In practice, what I think matters is preventing the effect of spoiler canidates (by which I mean spoilers because of non-strategic voters not Arrow's theorem style unavoidable but unlikely spoilers) and not decieving/confusing voters about the meaning of their vote.
What I dislike about approval voting is that it does exactly that. Lots of voters go in and simply approve of the canidates they approve of but in practice the choice to approve of another canidate as well knocks out their preferred canidate -- not just in unusual non-transitive preference cases. Range voting has similar problems but adds on an additional issue if how should non-strategic voters even understand what the rating means.
Ultimately, most people understand a vote as a choice, ranking their preferences maps nicely onto that idea but range and approval voting do not.
I honestly don't understand how you could say Range is less appropriate for expressing preferences, when it's obviously _more_ expressive than rankings. I tend to prefer approval because to the extent people express their ratings honestly in a range system, what happens is that they improve the social optimality of the outcome, but at the expense of their personal preferences. In the long run, whoever figures out the strategy of using only top and bottom ratings derives a certain amount of advantage over the folks who are more honest. So just reverting to Approval generally keeps a fairer playing field.
There's no particular evidence from races where people have a Vote-For-N (like with many City Council races), that people have any trouble understanding that there's no point voting for somebody they like but who is woefully behind... Like, everyone votes strategically, already. Approval simplifies the strategy. Look at who seems to be in the lead, vote for the best between them, and then also for everyone you like more.
Regarding expressiveness, you are either concerned with the strategic voters or the non-strategic ones. In the first case they play out the game theory and there isn't really any question of expressiveness only the formal properties about Condorcet criteria etc.
OTOH for non-strategic voters Range voting is less expressive because what does the degree of approval express? There is no clear answer so different people will use it differently and it just ends up adding extra confusion and uncertainty.
And I deeply disagree with the idea that voting should serve as some kind of intelligence test to give the more strategic voters an advantage. Sure, you can't avoid that to some degree but I think one of the most important features of a voting system is that people who show up, read the directions and follow them reasonably effectuate their preferences.
Approval voting and Range voting horribly fail that test.
Basically you've articulated why I prefer Approval to Range.
But I don't see how you can make this an argument for RCV. The problem with RCV is that it has a massive spoiler problem, which _will_ occur with non-trivial frequency, but it's _extremely_ hard to know in advance when you're facing a center squeeze election, and should swallow your pride and just top-rank your lesser evil instead of your true favorite.
Approval, meanwhile, has _exactly the same strategy as Plurality_. You always vote for the best among the front-runners. But then you also get to approve everyone you like better.
So on the statement " deeply disagree with the idea that voting should serve as some kind of intelligence test to give the more strategic voters an advantage" -- yes, but it's RCV (at least for the single winner case) that fails that test, not Approval.
Probably worth mentioning that CA also has a measure on the ballot (Prop 34) to stop AIDS Healthcare Foundation from doing any more ballot measures in the future.
Yep, I link to that in the Politico article! But thanks for flagging here as well, seems like people in CA, activists and political leaders alike really don't like that guy.
I hate him. I think I voted for a few of his ballot measures when I was younger and didn’t know better. But once it started becoming clear how much he was investing in preventing upzoning in Hollywood, and how much his condoms-in-porn initiatives were just designed to transfer porn production to Nevada, I started telling all my friends that if you shop at Out of the Closet then you are supporting keeping housing expensive.
OTOH that seems like a fun name for a gay clothing store
It is a great name for a secondhand clothing store whose profits supposedly support gay causes. It’s a shame that AIDS Healthcare Foundation is really a nimby organization masquerading as a gay health cause.
A fun example of the rich nonprofit NIMBY slumlord, which we certainly don’t have enough of in SoCal: https://www.latimes.com/homeless-housing/story/2020-03-08/homeless-housing-aids-healthcare-foundation-lawsuit-skid-row-tenants
A ballot measure I can support!
Maybe some canny politician can add a rider barring the evergreen dialysis clinic measures too, that'd fill two stones with one bird..."brought to you by the Paperwork Reduction Act!"
If it's really written that way -- this one group or guy can't do any more ballot measures -- I fear it might be ruled unconstitutional. Generally the courts have frowned on picking out certain groups - or people - and denying them electoral rights and soliciting ballot measures sounds like that.
Yes, that is a risk to the proposition. The language doesn't specifically mention AHF, but it's written such that AHF is clearly the only organization who it actually applies to. The CA Supreme Court already ruled the measure can go forward, but they reserved the right to look into it more after the election if it wins: https://www.sfchronicle.com/politics/article/aids-healthcare-ballot-measure-19578138.php
Local angle: Minneapolis adopted RCV starting with the 2009 election with all the sorts of promises Ben outlines above being made about how it would fix everything. I think everyone who lives here would have to agree it did not in fact "fix politics" in the city of lakes.
The CA rent control initiative is a great reminder that government by ballot initiative is very much *not* about "giving power to the people" instead it empowers special interest groups, the ballot initiative industry itself, and rich weirdos like Weinstein. Which is a far cry from the idea of representative democracy. I'd go further and say that a lot of California problems at their heart are caused by this system itself. A big part of the reason there's such mass incarceration in CA is the awful "three strikes" law (and here's another one that will make it worse). Likewise a big part of why middle class families are increasingly being forced out of the state and it's becoming a land of haves and have nots is Prop 13 and it's long term impacts.
But hey, you get to vote on ferrets and adult film stars using condoms so there's that I guess.
“… ferrets and adult film stars using condoms….”
On the one hand, the risk of pregnancy is very, very low. On the other hand, the risk of freaky zoonotic infections is high enough that I would certainly advise my ferret to wear the condom.
"I would certainly advise my ferret to wear the condom"
And on the whole, he's been a good client for the agency. When he first broke in, he was doing too many low budget quickies -- "Ferrets Gone Wild," "Pop Goes the Weasel," "Two Girls, One Ferret," that sort of thing. Beneath him, really. But he's hoping to go legit some day, and I have persuaded him to do more cinematic stuff, things that show his comedic range, like "Ferret Bueller's Day Off," and even cast him as a traditional romantic lead, like "A Ferret to Remember."
Yeah, that's why the park rangers made it illegal: https://www.youtube.com/watch?v=oovqYtMy1BI
Look, lwdlyndale, you cannot drag this negative energy into the tournament!
The condom one was a Weinstein initiative too.
Prop 13 has nothing to do with California's failure to build more housing. Prop 13 is the only reason a middle class exists at all in California, because before it was passed, in the period from 1975-1978, property taxes were doubling every six months, forcing people to sell their homes to pay their property taxes. It constantly blows me away when I see people advocating for returning to a system that taxed people out of their homes as an effective means to "build the middle class."
Counterpoint: prop 13 is bad. There's a whole giant literature about this so I won't get into it but basically it's a Yuge tax subsidy (ie welfare) for largely wealthier, older, white home owners that massively undermined the ability to create functional and equitable local government services for everyone else.
And the "taxed people out of their homes" thing was largely bogus, requiring people to sell a very valuable asset to pay taxes on that asset wasn't that common and actually makes a lot of sense, if you have a 2 million dollar home you're rich! Likewise there are a million better ways to deal with this relatively rare scenario that "Lets make white old home owners richer and gut the public education system at the same time"
Did Prop 13 not apply to non-white homeowners? I must have missed that part.
https://calmatters.org/california-divide/2022/04/california-prop-13-neighborhoods/ "Homeowners in wealthy, white neighborhoods in Oakland received thousands of dollars more in property tax breaks than their counterparts in neighborhoods with large Black, Asian and Latino populations,"
So the voters should have voted for political candidates who promised to stop raising property taxes, and then evaluated them on that promise. This is how a functional democracy works. The only system we're advocating for is being a democratic republic
> forcing people to sell their homes to pay their property taxes
Even if that's true, let's solve it without *also* freezing taxes on commercial and investment properties.
I assume prices and taxes were rising so fast because of limited supply and increasing demand. Limiting the building of housing
Yeah I think some of the claims RCV supporters make are a bit starry eyed. But at the core, they're right that's it's just an objectively better system, and if rolled out on a national scale it could make a significant positive difference to our politics, especially if combined with proportional representation.
As for ballot initiatives - you can argue three strikes is a bad policy, but I think it pretty clearly passed because it was popular at the time, not because "interest groups" wanted it. Certainly, in some cases a policy will be popular that will lead to unpopular outcomes, which is the argument against ballot initiatives and in favor of representative democracy. The idea being that representatives will try to deliver the outcome the majority of the population wants while reigning in its worst instincts.
However, sometimes representative democracy does fail to represent people on certain issues, mainly because they're voting on others that are more salient. Representative democracy failed to represent the majority of Californians on affirmative action, the majority of Missourians on union policy, the majority of Kansans on abortion, the majority of Floridians on minimum wage, and the majority of Idahoans on Medicaid expansion. The argument in favor of ballot initiatives is that they allow people to have their cake and eat it too - "fine tuning" democracy. They allow the majority to elect representatives on the highest salience issues but still get their way on other issues too.
Personally, I think my ideal state constitution would ban ballot initiatives for issues that could affect housing supply, on the grounds that they'd affect future residents who can't vote, but allow them for everything else.
“ my ideal state constitution would ban ballot initiatives for issues that could affect housing supply”
The most Slow Boring comment ever
Yeah lol I considered when writing that it may be a total hack-ish view that I came up with because I thought it would produce the results I wanted. But I do think the rationale of it affecting people who can't vote is a decent one.
In that case we should also ban abortion measures as it affects future births in the state (future voters)?
Well like Jefferson said, the earth belongs in usufruct to the living
RCV is _mostly_ an improvement over Plurality, except for the small-but-nontrivial percentage of the time where it fails so spectacularly that people repeal it and it poisons the well for reform for a generation, like happened in Burlington, VT.
https://ideas.repec.org/a/kap/copoec/v34y2023i3d10.1007_s10602-023-09393-1.html
(This paper notes that you can use kind of an opposite method to Instant Runoff, where instead of eliminating from the top, you repeatedly look at an inferred runoff between the _bottom_ two candidates, and eliminate the loser of that comparison. This method works out to be a Condorcet method, where if any candidate would win in every pairwise comparison, they will be elected.)
It's not like other systems have much of a better track record. St Louis, one of the few places which tried approval voting has a majority of aldermen in support of repealing it (if the state ballot initiative doesn't do it first). There are also non-governmental organizations that have tried approval voting and basically abandoned it after it devolved into plurality voting.
By contrast, Australia has been using RCV to elect their for over a century and, while a proportional system would of course be better, they seem to be pretty happy with it. By contrast, trying to find a country of that size, or any size at all, which has actually implemented approval voting, Condorcet, range voting, STAR, or any of the systems that online nerds like? Crickets...
The Ohio anti-gerrymandering initiative is worth noting as well.
Yeah, but so is Ohio Republicans' history of just ignoring the law on redistricting.
#1 on my list: The Baltimore Baby Bonus, which would provide $1,000 to parents when they have a child. It got the signatures to qualify, but the mayor has filed a lawsuit to remove it from the ballot. It'd be the first baby bonus ballot measure in history--I think also the first cash assistance ballot measure. The Circuit Court will hear the case on August 6th. You can see more at https://baltimorebabybonus.org
Am I taking crazy pills or has no one asked Donald Trump about the Florida abortion ballot measure?
Seems like a no brainier to ask “States should decide for themselves, right? How’re you voting as a Floridian?”
*US history. Australia has a baby bonus. Apparently many other countries do too.
https://en.wikipedia.org/wiki/Baby_bonus
Yes but none has instituted it by ballot measure.
What's the motivation for this? 1k is definitely nice, but doesn't do much to offset the cost of having a kid.
Local governments are budget-constrained. This is a small step that can offset some of the upfront costs like strollers, cribs, and diapers. We estimate it would lift 50 babies out of poverty each year, and reduce the severity of poverty for hundreds more, yielding long-term benefits for child development by intervening in the most formative years.
“ budget-constrained…a small step that can offset some… reduce the severity… yielding long-term benefits”
You’re singing the Slow Boring theme song.
How does $1000 lift anyone out of poverty?
If the difference between a household's current resources and their poverty line is less than $1,000, a $1,000 transfer will lift them out of poverty.
That's.. technically true, but it makes me think worse of you for weasel-wording, frankly. That family will be """out of poverty""" for 10 minutes, then back into poverty again. It's nice to give people a lift, but selling the program with that phrasing is suspicious.
It's the technical definition that all poverty researchers use. Poverty is measured based on resources and thresholds in a given year.
Baltimore has 7,000 births each year. Only 50 are within $1,000 of their poverty line.
Trying to define something like poverty is always going to seem a bit subjective and all the campaign can really do is use the agreed upon definitions to make their case.
If you read their FAQ page(and Max's other comment) you see they are making the claim that this will lift 50 children out of poverty out of about 7,000 births per year.
They seem to be trying to provide a pretty concrete view of the benefits, in a way that might seem to minimize them as 50 children might seem like a rather small number in the context of an entire city.
If they wanted to be less forthcoming they could have just more nebulously referred to the evidences of the benefits of cash transfers on crime, healthcare costs, CPS visits etc.
Are you working with the team on the measure?
It seems like cities would be very constrained on these type of things. Was it just easier to get on the ballot citywide?
I've advised the campaign on policy matters and canvassed a couple times for it, but don't have a formal relationship with them.
I don't think the council was interested in a legislative approach. Yes, cities are constrained on generous transfers, but limiting to babies cuts the cost, in Baltimore's case to about 1/300th of the city budget.
That's at least the efficient way to incentivize having children. Research suggests that very expensive tax breaks do little to incentivize having kids -- and imo are deeply unfair when given to wealthy people (another vacation) with kids as if their life choices were more worthwhile than people who don't have kids -- but money given at birth works well.
Probably makes sense. It's hard to plan too far into the future but there are alot of costs and time involved in having a child that can encourage people to delay.
"enough words to fill War and Peace a hundred times over will be written answering that question."
Not a fact-checker, but this is at the high end of plausibility.
Per Microsoft Word's wordcounter, War and Peace (Aylmer's translation; https://gutenberg.org/cache/epub/2600/pg2600.txt), has 564596 words. According to the anecdata in WaPo's "Why the New York Times is looking to shorten stories" (https://www.washingtonpost.com/opinions/2023/10/10/new-york-times-story-length/), the five front-page NYT stories in October 2023 had 1700 words apiece.
One hecto-War-and-Piece is therefore about 33 kilo-NYT-articles. We're one hecto-day from the election, so you'd need about 330 NYT-length articles published each day to match this pace. Most US newspapers these days are Gannett shells that just reproduce the national wire, and not all foreign newspapers will be breathlessly attuned to US politics, so many will have coverage a literal fraction of the NYT.
I don't think you can scrounge more than 60 newspaper-equivalents worldwide (citation: just making up numbers using my gut; probably an overestimate). If that's right, then 1 NYT-equivalent would need to publish about 5.5 articles a day on the election to fill out your claim. 5.5 articles is possible, given (say) how breathlessly Clinton's emails got covered, but it still seems a little high. If my estimate of 60 newspaper-equivalents is too high, then each NYT-equivalent would need to produce more articles than that, which is a bridge too far for me.
This is what I wanted to see in the comment section!
You're overlooking all the book authors, independent journalists/pundits/bloggers, social media commenters, student essayists, letters-to-the-editor writers, foreign government officials, corporate strategists, and others whose work is included in Matt's passive-voiced "will be written."
Ooh, yeah. Then 1 hecto-War-and-Peace is probably a drastic *under*estimate.
What the fuck is a kilo? Please rewrite this post using bushels and fathoms.
Also WTF was the thinking behind using all the money from the cap and trade program to jobs and etc programs to fight climate change?!? The cap and trade program *is* the best thing you could do so take the money and give it back to people with checks to politically insulate it or something else popular. If you think you have political capital to spend on the climate just make the permits more pricey (if you can send people more money that makes it easier).
I can see the policy logic. The more you raise in the short term, the more renewable energy you install, the less you have to spend on permits in the future. But if course that doesn't work if it's repealed.
They had one that looked more like this, in 2016 they proposed a carbon tax and used the proceeds to decrease the sales tax, but the Groups on the left killed it because the proceeds were going to everyone and not to “communities of color” or whatever.
Yeah, most of the money from the cap seems to be going to the tribes.
Major upvote for this.
Thanks for writing this Ben. Very informative and we need more coverage of state and local issues in our media ecosystem generally.
+1
Also, Bolts (founded by Daniel Nichanian) is the best resource on this.
https://boltsmag.org/
I don't understand how people can be confused by RCV. It's literally, "rank the following x candidates in order from 1-x according to how qualified you think they are to hold the office of (insert office here)."
I do understand why both major political parties would hate any method of voting that increases their opponents' chances of getting elected.
Concerns about comprehensibility are presumably not about how individual voters enter their votes, but about how the scores are aggregated: Borda-count, Dowdall system, tournament-style? There are options, many of which can be called, accurately, “RCV”.
In some of them, the person with the most first-place votes does not always win. I’m not saying that’s a bad thing, but it’s a thing that will need explaining.
Yeah. There are several politicians who have pretended to be against RCV because they believe minority communities don't understand it. I cannot think of anything more racist. Literally rank your choices in order. A 5 year old could understand it.
The majority of voters do not have opinions on multiple political candidates and are unable to distinguish them or their policy platforms from each other. Because the majority of voters do not think about politics, do not care about politics unless it directly impacts them, and are basically voting for who they'd rather have a beer with. They are unable to rank multiple candidates because they have no clue who they are or what the difference is between them. We are in the same country where over 40% of the population doesn't know how many Senators a state has.
I mean, do you love cricket? (Let's assume no). Could you rank the 6 greatest wicket-keepers of all time, in order? Probably not, right? It's not because you can't rank things, but because you have no clue who or what a wicket-keeper is, can't name any of the major players, and basically don't know because you don't care. Right? Well that's quite close to how voters feel about politicians.
You are probably in a bubble of highly-educated political obsessives who could rank multiple candidates. But the average person would think you're a politics-obsessed weirdo, and so cannot. Make sense?
Bwahaha! Since you brought it up:
MS Dhoni and Ian Healy for being ferociously quick around the wicket, keeping for tremendous spin bowlers. Sangakkara was pretty special, too.
A guy named Jack Russell (unforgettable name!), for England, who played before my time following cricket, who's still famous for playing up to the wicket for fast bowlers.
If you want to include batting, then you're putting Gilchrist and Quentin de Cock into the mix.
If leadership counts for something, then Ricky Ponting a deserves a place on the list. A cut (but only just) below GOAT with both gloves and bat, but one of the best captains qua captains ever.
Is that six? I've not really ranked them, and don't want to. My top-of-mind list also over-values the last several generations of players, because that's who I've seen. I'm sure there were tremendous keepers back in the days of uncovered wickets that no one ever thinks of anymore.
Any more SB cricket fans out there?
Nope it's very much the opposite. My experience with this is from actually talking to voters during the last mayoral election in NYC a ranked choice election. Voters have very little trouble understanding the concept of a second choice. Even minorities can figure it out. You're just way over thinking it.
50% of Maine voters only 'ranked' 1 candidate in Maine's last RCV election (which was their 4th BTW). Seems like a more relevant factoid than your anecdote.
You're probably in a bubble of politics-obsessed weirdos, but regular people do not know more than 1 candidate at a time, or what their exact policy platforms are
https://fruitsandvotes.wordpress.com/2023/01/15/the-ranked-choice-voting-elections-of-2022-in-alaska-and-maine/
You keep talking about me being in a bubble but you know nothing about me and I'm telling you about my experience talking to actual voters when canvassing. It seems to me that your ideological priors are the thing that are at issue here. It's really not that hard to rank people. I've had dozens if not hundreds of conversations with regular old normies about this. Who do you like for mayor? Oh who would your second choice be? Not complicated. Everyone gets it. Totally made up concern
People don't understand the complicated weighting of the votes after they are cast, not 'how to fill out the ballot'.
Doubling back to say that I don't think I could do a good job explaining the weighting system to a neighbor, and I am reasonably well educated. Assuming that minority communities, who might be less well educated and less engaged with the minutae of newfangled algebraic voting methods, might not understand it, seems not racist at all and actually a reason to do more work to educate people.
When canvassing in NYC in the last election this rarely came up. Although the New York system isn't algebraic and only takes first place votes so it's relatively easy to understand.
I’ve never understood why YIMBY advocates don’t take more advantage of trying to pass their own statewide ballot measures. Seems perfect for the their coalition: Takes the power away from local NIMBY’s with lots of free time, and gives it anyone in the state who votes.
Have there just been a lot of YIMBY initiatives that failed? Or is it just seen as likely to fail? Or have there been a lot of housing referendum I just don’t know about? I am particularly curious why this hasn’t happened more in CA
I-137 in WA supports social housing. Authorization passed last year with 57%
It is beyond me how anyone can support RCV after the "stolen election" idiocy of 2020. Whatever it's merits, too many people are incapable of understanding how the counting works. Can you imagine what would have happened if a bunch of swing states had RCV and Trump had been leading after round one? Given that there are alternatives that are easy to understand (eg approval voting), there is no reason to support RCV (yes, I know that RCV has many advantages. But the fact that RCV increases the likelihood that results will been seen as illegitimate trumps all of them).
A hot middle take is that maybe it would make sense to use RCV in primaries but not the general.
Approval voting is my number 1 choice and RCV is my number 2 choice, but I would vote for either of them.
Your comment has both a ranked choice and approval in it 😄
Given what I have discussed, why do you think that RCV is preferable to the status quo?
I have read plausible arguments that RCV doesn't even achieve its goal of leading to more reasonable, moderate, electable (in the case of primaries) candidates. It may encourage voters to vote for their extreme preferences (knowing they can still rank someone more reasonable below that preference) without thinking about electability first. And then the extreme candidates end up winning.
Exactly. A run off system is much better.
Very helpful roundup, thanks for doing this
My current rule for voting on ballot measures in CA is to vote no on any tax increase or new bonds to increase spending. This is a very poorly run state that blows money on all kinds of ineffective programs. Some states have no income taxes or no sales taxes but not CA. It wouldn’t hurt for this state to turn a little less blue. Newsom has very little chance of becoming President and he’s making things worse by signing dumb, irresponsible laws that allow schools to stop reporting transitions to parents.
"Newsom has very little chance of becoming President" I think this claim is pretty wrong. The lesson everyone should have learned from 2016 is literally anyone, even a know nothing game show host, can become president if they can win a major party nomination.
If Newsom can do that he'd have a decent shot depending on the fundamentals of the election cycle.
Anyone can, Newsom won’t. He and his brand of deep blue state politics are toxic in swing states that he needs to win.
Adultery. French laundry. Adultery. French laundry. Adultery. French laundry.
Which is why Bill Clinton and Donald Trump could never, ever be elected president.
19 year old girlfriend. Though to be fair he was much younger when this happened, only 39.
I don't really care about this stuff and neither do most voters. If character was an issue, Bill Clinton and Trump would never have been elected.
Exactly, if the nominees where Newsom and VP Vance in 2028 and the economy was in recession Newsom would easily win despite people screaming "But what about the French Laundry!?!?!?!"
Cat Lady Politics Explained: https://x.com/j_wolak_/status/1816854230117822945
Cat women aren't particularly Democratic, while dog men are more Republican, men who have cats are more likely to be Democrats than women who have cats, and no-pets women are the most Democratic group
Democats
It's not useful without knowing how many cats the individuals own though - the term, "cat lady" is applied to owners of multiple cats, not just one cat. (That said, "men who have cats are more likely to be Democrats than women who have cats" is probably one of the least surprising statistics I've ever seen.)
Sadly the GSS data from 2018 only has data about about 100 women with cats and no children so more there's a lot more to learn here. But yes there are tons of women with cats who vote for Trump/GOP out there so Vance's stereotype of liberal misanthropic cat women is well, dumb.
Don't sleep on the Oklahoma Minimum Wage Increase Initiative which may appear on the November ballot. This deep red state has used I&R to bypass its Republican supermajorities in the state legislature to legalize medical marijuana, achieve criminal justice reform, and to expand Medicaid. Will the voters do it again? I'm hopeful.
Good run down - I appreciate bringing attention to these, often ignored, initiatives.