Time for something completely different: another thread on @leedrutman's book, "Breaking the Two-Party Doom Loop".

This is just gonna be voting theory, nice and safe. Yeah, you could definitely connect that "doom" to our current more-literal one, but don't worry, I'm not gonna.

The story so far:

I did chapter 8, on how to solve things. Agreed with most of it, but argued that the focus on STV was probably too narrow; there are probably better versions of #ProRep.

...

I did Ch. 9, what's so great about proportional multiparty systems. Again, agreed, with a quibble: most of his points apply just as well to proportional multi-faction systems (ie, more than two subgroups with independent power bases, but still nominally grouped into two parties).

So now, it's time for ch. 10: historical examples of election reform, & what they mean for how to try it now.

Drutman starts by laying out a common framework for when reform happens:
1)Problem is clear
2)Solution is clear
3)Public demands reform
4)Incumbents aren't threatened

I think this is a good framework. Can't think of an example where all 4 points are met, where reform doesn't happen; or of one where they're not all met, where reform does.

Of course, I'm gonna argue later that PLACE is better than STV at being (4,nonthreatening). But first...

...let's look at Drutman's examples—and a couple he didn't use.

#1: US, 1842: Whigs mandate single-seat districts, attempting to preserve power.

Large majoritarian multi-seat districts had swept Whigs into a dominant position, but looked like the next wave election would...

...sweep them back out. To cut their losses, they mandated single-seat districts. This is still a majoritarian, not a proportional, method, but it tends to be a bit more proportional; fairer.

Drutman makes 2 points:
1) Whigs were following selfish advantage
2) US systems change

Example #2: USA, 1912. Appointed senators agree to 17th amendment making them directly elected.

Drutman argues that public pressure was important, but also the fact that this wasn't actually bad for Senators' self-interest. That is...

...by cutting the ties between state and national politics, 17a actually made Senators *more* likely to win multiple terms.

Again, good example, and good points.

Example #3: Western Europe, 1899-1921. #ProRep catches on.

This is obviously a complex story over many countries and multiple decades, but Drutman argues that there are two common threads, as w/Whig example above:

incumbent parties wanted stability;

voters wanted fairness.

Sooner or later, he says, it was inevitable that these two forces would align for PR; backsliding was unlikely, because voters would see it as less fair while incumbents would see it as less safe/stable.

Seems reasonable... but, see my examples below for more discussion.

Example #4: New Zealand, 1992-1993. Popular referendum to switch from choose-one to #ProRep (mixed-member proportional)

Quick technical aside here: note that I call MMP "mixed-member proportional" but am careful to say "single-seat districts" and "multi-seat districts". I find that if you use "member" in both terms, you inevitably trip yourself up. Avoid the confusion; it's "SSD", not "SMD".

So, NZ story.
1978: National Party gets false majority due to SSD, even though Labour got a plurality of votes.
1981: Same thing, again.
1984: Labour finally wins a majority with a plurality. Appoint voting reform commission; recommends MMP #ProRep. But they have power now! So..

...they try to quietly kill it. (Sound familiar, Canada? #ERRE) But the prime minister accidentally promises reform (supposedly, misread his notes).
1989: Labour switches PMs. New one, Palmer, had written a book against choose-one voting/FPTP. But no time before next election.

1990: New election. Since voting reform is now popular with voters, *both* parties Labour&National promise a referendum. National wins, tries to backtrack. But, with a weak hand, they eventually allow a referendum in 1992.

1992: The referendum was set up to try to kill reform—4 options, and if none got an absolute majority, status quo would remain. But MMP actually got 65% over all 3 other options.
1993: Second "are you really sure" referendum also passes
1996: First #ProRep election.

Lesson here: It's easy for a party that's out of power to promise reform, then try to break that promise once it gets power. (Again: surely familiar story to Canadians these days) But NZ suggests that in a democracy, eventually Lucy has to let Charlie Brown kick the ball.

Example 5: Maine, 2018. Single-winner RCV (IRV)

After "spoiled" elections in 2010 & 2014, Maine voters chose a system that tries to guarantee true majorities. My followers know I have issues with IRV—it can lead to center squeeze, and its "guarantees" aren't really—but...

...it's certainly at least somewhat better than what it replaced for Maine.

Anyway: though political insiders tried to quash reform in various ways, IRV is now solidly ensconced in ME.

Lesson here is about people-power more than incumbent politicians. Citizen-led referendum.

Drutman goes on to draw general lessons. But I want to discuss some other examples first.

#6: UK, 1832. Whigs end "rotten boroughs" and establish equal apportionment.

Because parliamentary seats hadn't been systematically reapportioned, well, ever, some had tiny # of voters...

...and this typically favored the Tories on net (does "antidisestablishmentarianism" fit in a tweet?)

Whigs wanted reform. When they won 1831 wave election, they instituted equal-population districts to cement that advantage.

As w/Drutman's egs, it took both voters&incumbents.

To me, this 1832 example may actually be closer to US 2020 than others. One party fighting a headwind of systemic bias; wins nevertheless; institutes reform to reduce bias in the future.

In other words: reform in the US may be more top-down, and more partisan, than Lee suggests.

Two more related examples: Progressive-era (1910s-) Bucklin and STV in US cities.

Bucklin is a single-winner voting reform (modern version: Majority Judgment). STV is proportional, multi-winner. Both help address the two-party (Duverger) bias in choose-one voting.

Both were instituted in numerous US cities in the first half of the 20th century. Both were rolled back almost everywhere (except STV here in Cambridge, MA).

Lessons?
1. Reform comes in waves
2. ...but can also recede the same way.

And also:

3. Nonpartisan voting methods such as Bucklin or STV seem to appeal to US tastes, but may be less "sticky" than those such as MMP which establish explicit roles for institutional parties.

I think that in STV (and Bucklin), the incumbent majority comes to feel that their ox won't be gored by rollback. Meanwhile, it seems voters let that rollback happen. Maybe the complicated ballot & process is harder for voters to get worked up to defend than Euro-style #ProRep?

So, this brings me to Drutman's conclusions. He returns to 4-point schema, and argues all 4 are coming ripe in today's US.
1)Problem is clear? Yep. 2-party system clearly broken.
2)Solution is clear? In general, yes. Proportional=fairer.
3)Public demands reform? More and more.
4)

Well, 4) is, "a majority of elected officials believe they'll survive or even thrive under the new system".

I honestly doubt that will ever be true of STV. Republican incumbents will oppose it unanimously; gerrymandering helps their party on net. And Democrats?...

Sure, I can imagine a majority of D incumbents getting on board with STV. But it takes more than a majority of one party to pass; it takes nearly all of them. I think STV is too disruptive, too scary to incumbents, to get that.

Now, Drutman doesn't take the tack I'm taking here. Voting reform is, in principle, a nonpartisan good-government reform. Democracy should be as uncontroversial as motherhood & apple pie. So his book makes a nonpartisan pitch for reform. That's fair and honest.

But the closer reform gets to reality, the more people are going to inevitably game out its partisan implications. And frankly, there's no way around it: ending gerrymandering with #ProRep would, in US 2020 (or, heck, 2025), be good for Democrats and bad for Republicans on net.

So, I prefer to rip off the nonpartisan band-aid, and make a partisan pitch. This is *still* a matter of basic democracy and in principle good for all voters in all parties. But for incumbents? You gotta pick your team, and I'm picking D.

How does that relate to proportional method design (ie, STV vs PLACE)?

I think PLACE is better than STV for voters of all stripes. Simpler for low-engagement voters; gives more unified negotiating power to high-engagement voter groups. But that's not my point here.

My point here is that, without question, PLACE is going to be more palatable to incumbents worried about their next election than STV is. Unless you owe your seat to gerrymandering, PLACE gives you a fair chance of keeping it; STV is more of a gamble.

I think that that makes PLACE far more likely, in the long run, to get the nearly-unanimous Democratic support it would need to pass as a Federal law. (That is, if it ever becomes more than just my personal hobby horse; which I'll admit, is gonna take me upping my game.)

OK, that's the end of this thread. I'm nearly to the end of @leedrutman's book now; just the brief ch. 11 to cover. But, I'll do that another day.

/end