Voting Systems

Most elections in the US and around the world use the first-past-the-post voting system. In this system, everyone gets one vote, and the candidate elected is the one that gets the most votes.

But this is not the only way of electing candidates! There are many other possible voting systems, most of which involve each voter ranking or rating multiple candidates according to their preferences, and then somehow turning all these rankings or ratings into a single elected candidate. (If you’ve heard of “ranked choice voting”, that refers to a specific ranked voting system, not a general term encompassing all ranked voting systems.)

This is a proposal for a new1 voting system called Ranked Head-to-Head Voting.

The Problem: Compromise Candidates are Ignored

The problem this system addresses is that first-past-the-post has no way of electing a compromise candidate: a candidate who is not many people’s first choice, but is a choice that many people could get behind.2

Imagine an election with three candidates: Alice, Bob and Carol. Alice is the top choice of 42% of people, Bob is the top choice of 40% of people, and Carol is the top choice of 18% of people. It might seem like Carol should be written off, but all Bob voters actually prefer Carol over Alice, and all Alice voters prefer Carol over Bob. In this way, Carol is a compromise candidate.

Notice how Carol would actually win a head-to-head matchup against either Alice or Bob, because all Bob voters would choose Carol over Alice, and all Alice voters voters would choose Carol over Bob:

But under first-past-the-post, Carol is never going to win. In fact, Carol becomes a “spoiler” candidate: the Carol votes are wasted, and Carol voters do not get a say in the Alice-vs.-Bob matchup.

Ranked Head-to-Head Voting is a system designed to pick out Carol here.

We need more elected Carols. The US is increasingly splitting into two groups that have almost totally opposed political views, a trend taking US politics to its breaking point. This polarization is exacerbated by the standard first-past-the-post system, which forces people to band into two groups to have a chance of amassing enough votes. Electing compromise candidates would go a long way toward reversing this trend.

Ranked Head-to-Head Voting

Ranked Head-to-Head Voting is a two-round system. The first round narrows the field down to at most 4 candidates, and the second round chooses from them the head-to-head winner: the candidate that would beat any other candidate in a head-to-head election.3

The first round works like first-past-the-post, with everyone voting for a single candidate. The 4 candidates with the most votes make it to the next round. The one exception is that any candidate with less than 5% of the vote is always eliminated, so the second round can have fewer than 4 candidates if there aren’t 4 with at least 5% vote share. See the Q&A section for the rationale behind all this.

In the second round, each voter ranks the remaining candidates. A voter has to rank all remaining candidates for their vote to count. In this sample ballot, Dave is added to the example from before:

You then tally up how each candidate would do in a head-to-head election against each other candidate. Each voter’s choice in a head-to-head matchup is determined by their ranking: if they rank Dave above Carol, they count as a vote for Dave in a head-to-head against Carol:

So if 60% of voters rank Carol above Bob, and the other 40% rank Bob above Carol, then Carol would win 60-40 in a head-to-head election. You can visualize all these matchups easily with a matchup table:

The winner is the candidate that wins a head-to-head matchup against every other candidate, shown by their row in the matchup table being all green. In this case that’s Carol.

Carol here was a compromise candidate - she wasn’t many people’s first choice:

Ranked Head-to-Head doesn’t always elect a candidate with low first-choice support like this, though. If a candidate has high first-choice support and is also popular enough to beat any other candidate in a head-to-head election, then they’ll win under Ranked Head-to-Head as well:

Ranked Head-to-Head just helps in the case that things are more polarized, where the candidate with larger first-choice support might not be the head-to-head winner.

“Compromise” also doesn’t necessarily mean “moderate” or “incrementalist”. A compromise candidate could be an ambitious outsider that wants to take things in an entirely different direction, maybe someone like Ross Perot in 1992. Some people have this image of a compromise candidate being a moderate who wants to preserve the status quo, but this is not always the case!

Tiebreaker: Highest Floor

Most of the time, that’s it. Occasionally, you can have no candidate winning all of their head-to-head matchups, because there is a cycle: Irene beats Jack, Jack beats Kira, and Kira beats Irene. In that case, you look at each candidate’s head-to-head “floor”, the vote share they get in their worst matchup, and pick the candidate with the highest floor4.

Jack here is the Ranked Head-to-Head winner because his floor, 45% in his head-to-head with Irene, is higher than Irene’s floor of 20% and Kira’s floor of 35%.

And that’s really it! The system is not overly complicated.

Practical Use

In the US, the electoral college probably makes this system impossible to implement for presidential elections. The intended use cases are congressional elections (especially Senate) and lower-level executive elections (governor, county executive, mayor).

Ranked Head-to-Head Voting is recommended in the US as a replacement for primaries, so all candidates from all parties run on the first round ballot, in addition to any independents. This two-round, top-4 system replacing primaries is already being used by Alaska (though not with head-to-head voting for the second round), so it has precedent. Using Ranked Head-to-Head in addition to primaries, while possible, forces people to show up to vote many times in what are often already low-turnout congressional elections.

For lower-level legislative elections (state assembly, city council, etc.), and for parliamentary democracies outside the US, a different approach called proportional representation is recommended instead. The US House is a special case where some form of proportional representation is potentially workable but probably hard to implement5, making proportional representation and ranked head-to-head both defensible systems to advocate for.

Q&A

Why Not Ranked Choice Voting?

Ranked Choice Voting (RCV) is the most popular alternative voting system to first-past-the-post. While Ranked Choice helps with spoilers, it does not elect compromise candidates in the same way as Ranked Head-to-Head.

Take the case from before:

The point of Ranked Choice Voting is that, after Carol is eliminated during the first step because of her lower vote total, all of Carol’s votes get redistributed to the Carol voters’ second choices instead of being wasted:

But Ranked Choice Voting still doesn’t lead to Carol getting elected! It helps with the spoiler aspect, but does not elect the compromise candidate.

Why not [other voting system]?

Read this post for in-depth discussion of the different voting systems (including top-2, score and approval voting, and STAR) and the advantages of ranked head-to-head.

Why Mandatory Ranking?

Without mandatory ranking, in practice many people rank only a top choice and do not bother with the full ranking. This type of partial ranking is called bullet voting.

You can see this in Alaska and Maine’s ranked choice voting results. Since 2022, Alaska has used a two-round top-4 system like the one described here, with the second round using ranked choice voting. Below is a graph showing the amount of bullet voting in Alaska. It plots the percent of each candidate’s first choice voters that did bullet voting, for all candidates that did not become one of the two finalists6. As you can see, Alaska sees in most cases upwards of about 20% of these candidates’ voters doing bullet voting:

Since 2018, Maine has used ranked choice voting for federal elections without a two-round system, though it’s only been relevant in their second congressional district. There too upwards of 20% of people voting for non-finalists did bullet voting:

Bullet voting would be a big problem in ranked head-to-head. Ranked Head-to-Head depends on all of the cells in the matchup table being accurate: if a large fraction of people rank neither Alice nor Bob, that could lead to an inaccurate picture of the head-to-head matchup between them.

Bullet voting also leads to bad incentives for candidates. As in ranked choice voting, it can cause candidates to drop out of the race so they don’t become spoilers, though the exact meaning of the word “spoiler” is more complicated in ranked head-to-head.

Why two rounds?

Mandatory ranking also reveals why there are two rounds. With only one round, there can be too many candidates for a voter to realistically make an informed ranking.

In Australia for example, they have ranked choice voting for their House of Representatives, and voters are forced to rank all candidates for their vote to count. Voters there see an average of about 8 candidates7 on their ballot. Voters forced to rank 8 candidates are bound to give uninformed choices for many spots on their ranking, leading to a worse and less democratic process overall.

One consequence of this two-round system is that voters may be pushed to vote strategically in the first round rather than putting their genuine top choice. If their top choice has no chance of making the second round, their best strategy would be uniting behind one of the candidates that do have a chance. Unfortunately, there is no easy way around this. You need voters to unite around a few candidates at some point, and the second round still provides plenty of opportunity for true preferences to shine through.8

Why the 5% cutoff?

The 5% cutoff is used for similar reasons as the top-4 part. A candidate with less than 5% of the vote is very unlikely to win the second round anyway, and forcing voters to consider them just puts more burden on voters, leading to more uninformed choices.

Why the “highest floor” tiebreaker?

There are many possible tiebreakers to use. Highest floor is chosen as a simple approach that fits well with the “compromise candidate” rationale of the system. A candidate with a high floor doesn’t lose too badly against anyone, and so should be someone most people can tolerate. Tiebreaks should be rare anyway - they require a very specific configuration of voter rankings unlikely in real elections - and highest floor generally doesn’t encourage strategic voting that could lead to tiebreak situations.

I wrote this sentence a few weeks ago, when mid-decade redistricting was mostly a wash. I’m guessing fewer people today would make the claim that gerrymandering’s partisan effect is not likely to be large. Because I can see ahead more than a few days, I had already written in this piece that I was skeptical, that I expect Republicans to benefit at least a little bit, and potentially a lot.

That is still my position in light of everything that has happened since. The basic rationale is that Republicans are more ruthless and less respectful of democratic norms. Democrats are starting to become more ruthless too, but still I expect them to, for instance, be more likely to create maps that abide by the majority-minority district guidelines that the VRA formerly required even after SCOTUS has gutted it. There is already starting to be some basically cope spreading around1 about how Democrats don’t need to violate these former requirements to do more gerrymandering. This is not to mention how Democrats have to dig themselves out of the hole they put themselves in by adopting independent redistricting commissions in many blue states.

The second claim, that gerrymandering is going to lead to fewer competitive districts, is definitely true if sometimes a bit overstated. Sometimes it is in a party’s best interest to gerrymander by creating more competitive seats rather than maximizing the number of safe seats, as Democrats in Oregon have done. If Republicans are generally more willing to gerrymander, then it is probably the right play for the Democrats that are willing to gerrymander to do so by creating more competitive districts in some areas, in order to maximize the number of winnable districts. (They probably largely won’t do that anyway, just cus it’s a little counterintuitive, but they might).

I think most interesting is the core claim, that more competitive seats is better.

The Ideological Effect of Competitive Seats

The primary intuition people have is that more competitive districts is likely to lead to more moderate representatives. If you’re a fan of consensus democracy and think democracy should represent the median view in society better rather than fluctuating between two extremes, as I am, then this is good.

More precisely, usually what we care about is not that there are a lot of moderates elected, but that the median vote in the legislature - the one that determines whether bills pass - is moderate. Note that this logic does not apply the same way when there are supermajority requirements, like the Senate’s 60-vote filibuster, because there you actually also care about how moderate the 40th or 60th senator is. Let’s set that aside for now and just focus on the median vote in the House.

Should we expect more competitive districts to actually lead toward a more moderate median vote? I think it depends.

The general dynamics

The first thing to observe is that competitive districts actually make the results of an election less proportional. If there are 100 seats, 45 safe for each side and 10 swing districts, and a party wins 55% of the vote, they are likely to get exactly 55 seats. If there are 30 swing districts, that party will probably win all 30 and get 65 seats. This extremely disproportionality is an inherent feature of first-past-the-post in single-member-districts, and actually gets worse with more competitive districts.

This is weird and counterintuitive. Many of the people who want moderates elected and hate gerrymandering because of the statement at the top are the same people have actually called for proportional representation to remedy it. Not what you would do if you were concerned about competitive districts!

Bigger majorities because of more competitive districts often lead to a less moderate median representative. As a simple example, suppose 10% of Republicans are moderates, then if Republicans win 51% of the seats you’d expect the median vote to be moderate, while if they win 60% of the seats you wouldn’t.'

In other cases, more competitive districts can lead to a more moderate median representative. Suppose moderates have an electoral edge, then competitive districts will be more likely to elect moderate candidates. This is both because of selection effects - voters select for moderate candidates in the general election - and because of intentional moderation - either representatives strategically moderate to win the general election, or parties are more likely to nominate genuinely moderate representatives in swing districts.

I vibecoded a little applet to understand these dynamics a little better.

More competitive districts leads to a more moderate median representative when there are a lot of moderates in swing districts, and not a ton of very moderate candidates in opposite-party seats:

If there are a lot of moderates in swing districts, but also a decent number of very moderate candidates in opposite-party districts (10+% to opposite party) capable of pulling off upsets - as has generally been the case historically - then this effect disappears or even reverses. This is because reps in opposite-party seats are likely to be the ideological median under extreme gerrymandering.

When there aren’t a ton of moderates in swing districts, but the overall level of moderation is quite high (and varies a lot from representative to representative), then the smaller majorities under extreme gerrymandering lead to a strong effect in the opposite direction:

A priori, I’d think the exact scenario where you get the intuitive effect is somewhat unlikely - when you get to a point where swing-district candidates are moderating a lot, you’d think you’d also see reps in opposite-party safe districts moderating a lot. Notably though, I do think you can make a case that we were in a mild version of this scenario in in 2022 and 24, where Democrats in swing districts substantially outperformed others but without any Democrats winning in deep red districts:

Of course, these are all just toy models, there’s a lot of complexity not captured here, and even here there’s more complexity when you add in asymmetry. (I am pretty confident these are the general dynamics though). Overall, I think it’s hard to predict whether more extreme gerrymandering will lead to a more or less extreme median representative going forward. I suspect the actual effect will be relatively small. And regardless, this is not really why gerrymandering is bad.

Gerrymandering’s primary negative consequence is not its effect on competitive seats

Gerrymandering is bad primarily because it creates a partisan bias, leading to one party taking power even when it loses the popular vote. This is undemocratic.

If the baseline map is slightly uneven, more (naive2) gerrymandering in equal proportion from both sides amplifies the partisan bias substantially. The best way to measure this is by looking at the median partisan lean of the districts, also called the mean-median gap, which is a rough measure of how much each party has to win by in order to get a majority. So a mean-median gap of R+X% means Democrats have to win the popular vote by about X% to win a majority.

This is because if there are many swing districts, a bit of Republican gerrymandering (moving a few swing or blue districts to be solidly Republican) will move the median district from an EVEN district to one of the next middle-most districts, which is maybe like R+1. If there are very few swing districts, then a bit of Republican gerrymandering will move the median district from an EVEN one to the one of the next middle-most districts, which is likely to be far more Republican:

Secondarily, gerrymandering is bad because it nationalizes state-level elections, forcing voters to consider the effect of their vote on their state’s congressional districts, rather than just considering who would do a better job governing. This makes voters less likely to throw out the party that dominates the state even when they do a terrible job governing.

(You might question the extent to which voters actually consider the effects of state-level elections on redistricting. But I do think gerrymandering at least creates a vague feeling that these elections have national consequences, even if voters aren’t making the specific connection in their mind.)

Holding partisan bias constant, gerrymandering’s effect on competitive districts - if a problem at all - comes strictly third.

Aims of Gerrymandering

The key to gerrymandering is to make the opposing side’s margins in the districts they win as large as possible while making your side’s margins as small as possible (while still being comfortable). Winning a seat narrowly is just as valuable as winning it by a landslide, so doing so allows you to win more seats. The general strategy to achieve this is “packing” most of the opposition voters into a few districts and “cracking” the rest of the opposition voters among many districts won by your side.

Seat-Vote Curve

The best way to think about the political distribution of districts is to consider the seat-vote curve: what percent of seats a party can expect to win at different levels of vote share. A totally proportional system, for instance, is just a straight line with slope 1:

In practice, single-member districts always induce disproportionality, so the seat-vote curve will look more like this:

The goal of gerrymandering if you’re Republicans is to push the line upward, like this:

Generally, this reduces to two more specific goals:

Mean-Median Gap

The chief goal of gerrymandering is to change the mean-median gap. This is the gap between what percent of the vote a party gets overall (which is almost the same as how well it does in the mean district, except for different districts having different turnout) and what percent of the vote it gets in the median district.

Since different districts tend to move in lockstep from election to election, the partisanship of the median district will generally determine which party wins a majority. That is, the mean-median gap will correspond to what I’ll call the “majority partisan bias”, the difference between 50% vote share and the vote share at which a party wins 50% of the seats:

The other thing to note is that candidate quality can mean the difference between winning and losing a race, particularly in a less polarized environment but also on the margins in a polarized environment like we have today. For this reason, the best way to measure the mean-median gap is at the presidential level, ie measuring the overall popular vote margin in the presidential election and then the margin in the median district (by presidential margin). Political margins on the congressional level have in recent elections closely correlated with presidential results, and because of the much higher media coverage that presidential elections get relative to congressional elections, we can be pretty sure that the presidential margin is causing the congressional margin far more than vice versa.

(Another piece of evidence for this direction of causality is that the partisanship of congressional districts tends to lag presidential elections; for example, the best predictor of the House margin in 2026 is something like 2/3*(presidential margin in 2024) + 1/3*(presidential margin in 2020). Sources: FiveThirtyEight and some David Shor tweets that I can’t seem to dig up)

Of course, districts don’t move totally in lockstep. A more sophisticated analysis would consider not just the median district but all the close districts and how likely it is that, due to variance in candidate quality and in future seat trends, a party will win >50% of the seats at different vote shares. Like the seat-vote curve, the probability that a party wins a seat goes up much faster than the percent of the presidential vote they got in that seat. In fact it’s even more stark than the seat-vote curve:

You might also object that the variance in future seat trends in different seats are not independent. If a Republican does better in one heavily Hispanic seat next election, they’re likely to do better in other hispanic seats as well. If Democrats have gerrymandered many districts to be narrow victories off the backs of Hispanics, this might turn into a “dummymander”, where all of those seats flip at once.

All this said, in practice the mean-median gap is very predictive of the majority partisan bias, particularly if you consider not just the one median district but the few middle-most districts.

Proportionality

The secondary and strictly less important goal of gerrymandering is to win larger majorities, which corresponds to shifting the seat-vote curve upwards. The bigger your majority, the more defections your side can handle from moderate or otherwise rebellious legislators when passing legislation.

If every state elects legislators proportional to their partisanship, the overall results will be proportional. In practice though, this is very difficult: it’s usually impossible to draw a blue district in a 70+% Republican state, for instance, unless you have lopsided margins in a specific region. You’re going to have to expect disproportional results in lopsided states. The most you can hope for is that things will be proportional on the national level.

As illustrated by the above graphs, single-member districts in practice always lead to disproportionality; ie, a party that wins 52% of the vote will win more than 52% of the seats. In the 2024 Presidential Election, for example, Trump won 50.7% of the two-party vote and 52.9% of seats.

If you have totally proportional results, you’re guaranteed to have zero mean-median gap. If you have small mean-median gap however, proportionality may or may not be a good thing. Proportionality in that case basically means fewer competitive seats, which has interesting and potentially counterintuitive effects on politician ideology.

Credit to these Eharding posts for helping me think about this stuff more clearly.

Non-Political Objectives

What would it mean to draw lines without a political objectives? What other criteria of objective redistricting are there?

The main logic behind the different objectives that people point to is that districts should be homogenous. If a representative is elected by people with the exact same political views, that representative will be totally loyal to these views. Once again, if you buy this logic, you should really just adopt proportional representation, which satisfies this logic precisely: a party will represent all people with views closest to them.

With single-member districts, this logic becomes less obviously appealing. For starters, the way people think about this is pretty backward. A political group is often actually more powerful if it is split up between multiple districts than if it gets a district all to itself. Take a case where wealthy suburbanites are the swing voters and the rest of the state is split 50/50. These wealthy suburbanites will be most powerful if they are spread out among multiple districts, as they will determine which party gets elected in each of those districts, as opposed to if they are packed into a few districts and only determine which party those reps come from.

But ok, maybe you actually want homogeneity because you don’t want swing groups to be so powerful. Even then though, I don’t think this works as you might expect. The issue is that swing voters make up a relatively small percent of the swingy groups, and these swingy groups are not 50/50. Asian voters swung dramatically toward Trump between 2020 and 2024, but still probably by less than 10%. Because of this, even if you put all the Asian voters into one district, it’s unlikely that district will be a swing district, because Asians still vote for Democrats overall. In practice, the real way you get swing districts is by putting a heterogenous group of people together, and then it will just be whichever swingy group happens to make up a decent percent of that district gets all the power.

Where I do think homogeneity maybe makes sense is at the primary level: you want to allow smaller interest groups to win primaries by clustering them together. This is what the VRA does did for minority groups, as I’ll get to in a second. Again a proportional system would accomplish this better, but nevermind that. This is a legitimate reason to want homogeneity, although homogeneity of primary electorate is different from homogeneity of general electorate.

Regardless of if you think the metrics I’m about to introduce have much objective merit, they are kind of “Schelling” metrics, compromises that people with different political objectives would naturally come to because what else are you going to optimize for. Optimizing for these metrics would cut down on gerrymandering just because you are not optimizing for political goals.

Compactness

There are a few different metrics of compactness, but generally what it means is that you want districts to look like this:

Rather than like this:

If you care about homogeneity, the right metric is probably the average distance between a voter and the center of population of the district, the center of population being the average x and y value of voters in the district. Smaller numbers are more compact. This basically measures how close to each other people live, with the assumption being that people closer to each other are more similar.

If you don’t care about homogeneity, you could measure it differently:

• The average distance between a voter and the center of geography of the district, ie the average x and y value of land in the district

• The area that a circumscribed circle would have.

• The diameter of the district, the max distance between any two points in the district.

• The ratio of the perimeter of the district to its area

Splitting Locality Lines

This one is simple: how many counties, or cities, or precincts, do not sit entirely within one district (1 for every district the locality does not fit inside). Perhaps you could weight this by the population of those localities to account for gerrymandering being more effective when you split more populous localities. For homogeneity, the assumption is that people in the same locality tend to have more similar views.

Communities of Interest

You don’t just have to stop at localities: you can look at the geographic distribution of ethnic, racial, religious, etc. (for simplicity I’ll just say demographic) groups by precinct, and treat areas with lots of precincts each supermajority-demographic as quasi-localities whose lines should not be split.

I don’t love this metric because it gets pretty tricky and subjective:

First, you have to decide which demographics are of interest.

Next, you have to decide on what qualifies as a supermajority of a precinct for communities of interest purposes.

Next, you have to decide what to do with communities of interest when they are not totally contiguous, like there is a neighborhood separating two black regions that is not majority black. This is often relevant in the Black Belt in the south, which is pretty scattered. The darker areas in this map have higher black population:

Finally, the demographic distribution of a region often changes, so communities of interest can become outdated.

All this said, if you specify in law how to deal with the above issues, it might be ok. The Voting Rights Act in fact enforces enforced communities of interest for racial minorities.

Geographic Advantages

Gerrymandering means intentionally drawing lines to get a leg up. A party may however have an inherent advantage due to its geography. That is, lines drawn without political intent, perhaps drawn to optimize the above metrics, would still give one party an advantage.

Overall, Republicans have a geographic advantage. The general rule is that Republicans have a geographic advantage in the Deep South and Midwest/Rust Belt, while Democrats have a geographic advantage in the West Coast, New England, and Texas. There are exceptions of course: Georgia, New Hampshire, Maine are all pretty neutral, for example.

This great tool from FiveThirtyEight from 2018 (preserved via internet archive) is a good way to visualize this. This is their attempt at drawing districts matching objective criteria:

Some asterisks with this map:

• This map does not comply with the Voting Rights Act, which helps Democrats overall

• This map is outdated, Republicans have less of an advantage these days.

That said, it is possible to draw federal maps that score well on mean-median, proportionality, compactness, and splitting few county lines, you just have to try to do it.

Solutions

With all that out of the way, let’s get on to solutions.

The first thing you might think of is to have an independent commission drawing lines. This can be difficult. New Jersey and California have independent commissions but which ended up drawing democratic gerrymanders. They were quite mild gerrymanders by politician standards, to the tune of 2 (not super-safe) seats out of 12 in New Jersey and 5 out of 52 in California, but gerrymanders nonetheless. In both states, the commission had an equal number of Republican and Democratic appointees, but had tiebreaking procedures that involved people who leaned left2.

This isn’t always the case though - in Colorado, Arizona and Washington, the districts drawn by their independent commissions were genuinely fair. Certainly an independent commission results in less gerrymandering than politicians drawing it, but still not amazing. I also think it may not be possible to mandate the type of procedures that states have in place to make their independent commissions non-partisan - on the state level they mandate that members of the minority caucus in the legislature get to make appointees, which doesn’t feel like the type of thing you can mandate nationally (national election law is pretty party-blind), but maybe.

Is there a better metric-based alternative?

The thing that makes assessing partisan gerrymandering in the US tough is that you can’t use the mean-median gap. The mean-median gap only makes sense at the federal level, as most states are well to the left or right of 50-50, and redistricting is done at the state level. You cannot possibly tell a state "redistrict in a way that minimizes federal mean-median bias". That sentence doesn't make sense, the state map that minimizes federal mean-median bias will depend on other states' maps.

(Maybe you could have a federal redistricting commission that approves states’ maps and is given a goal of minimizing federal mean-median bias? I’m skeptical of this idea though because it doesn’t seem like the commission can provide a meaningful reason to reject states’ maps if they don’t lead to small federal mean-median bias. I think the commission would have to tell a state something like “draw another Democratic district so that things federally are evened out”, which is not great.)

So instead you have to create other criteria to judge state maps on. Once you have some criteria to judge a map on, you still have to figure out how bad a map is relative to other maps. It might just be impossible to draw maps that are compact or that have high proportionality.

The approach taken by the plaintiffs in the Supreme Court case Rucho v. Common Cause was to generate a bunch of maps with an algorithm that optimizes for the objective criteria (compactness, splitting lines), and then assess how bad the adopted map relative to the other maps the algorithm spit out on partisan fairness. This is one solid approach if you can trust an algorithm enough.

Probably the partisan fairness metric you want to use is proportionality of presidential election results at various points close to 50-50. If you want to be extra mathematical about it, you can model elections with a normal distribution centered at 50-50, with the standard deviation measured historically3.

Another approach is to just use the algorithm. Have an algorithm defined in federal law that states are forced to adopt. This may or may not be constitutional - states are after all given the power to redistrict, federal law may regulate that but not override it. You also have to really trust the algorithm; if you trust the algorithm to optimize the objective criteria a good amount but not entirely, you might want to go with the above method.

Here’s an alternative idea I thought up. Allow a court to strike down a law if someone sues the state with a map that scores better on objective criteria and/or partisan fairness. Or instead of allowing people to sue with new maps after the fact, you can force states to take map submissions beforehand, and then courts can strike down a map if one of the maps submitted does better than the map chosen. (This is to prevent someone from coming along years later with some new map that forces redistricting.) Alternatively, you can simply force states to choose the submitted map that best satisfies the criteria, although this may not be constitutional as in that case it is the submitters rather than the state drawing the map.

If you choose this non-algorithmic approach, you need to either strongly select for objective criteria - strike down any map if there is another map that does even slightly better than it on objective criteria - or select for objective criteria + proportionality. If you only weakly select for objective criteria, this gives a lot of freedom to states to draw gerrymanders that happen to match objective criteria pretty well. That’s still better than nothing, but not great.

This post is about the ins and outs of the legislature - what it looks like and how it turns the many different views of the legislators into concrete policies to pass. First, an intriguing question: how big should you make your legislature?

Optimal Legislature Size

The pros of a large legislature are similar to the pros of the legislature itself that I laid out in my post on the legislature:

• When the legislature is bigger, there’s less randomness in the views of legislators due to larger sample size.

• Random events like deaths are less impactful when the legislature is large.

  • These two points are more important than you’d think because the legislature is often closely divided.

  • You can partially get around this second point by allowing each legislator to appoint a “vice legislator” who takes over if they die, and maybe can take over temporarily if they get sick. This could be the legislator’s chief of staff. Or alternatively you can let the party of the legislator appoint a replacement.

• You have more talent in the legislature when it is larger, and hopefully the most talented people in the legislature have outsized impact by introducing good bills or working their way up to party leadership.

• You can represent the views of the public with more granularity when a legislature is larger; smaller groups are more likely to have proper representation in the legislature.

  • If you use proportional representation (which you should), you can always get rid of this granularity using a threshold, which you might want to do if you’d prefer small groups form coalitions instead of getting their own representative

• It’s more expensive to bribe more people when the legislature is larger relative to the country’s GDP

  • The thing is, it’s also harder to bribe legislators when their public salary is higher. I’m not sure if it takes more money to bribe 100 legislators paid $100,000/year each or 10 legislators paid $1 million/year each.

• The legislature can get more work done when there are more legislators

  • Though legislators can also just hire more staff in a larger legislature, so I’m not sure how big a deal this really is

The cons of a large legislature are:

• On the flip side of the point above about talent, the average talent of a legislature is lower when the legislature is larger

  • This one is a bit confusing. I don’t think it should be hard to find a few hundred really talented people out of a country of 10 million+ people. But I think in practice you do see quite a few really inexperienced people in the legislature. I’m not sure how much this is dependent on legislature size though. It definitely depends more on how the party selects its candidates.

• It’s physically more expensive to run a larger legislature.

  • As stated in the point about bribery, you might want to pay legislators in a small legislature more to prevent bribes, so the salary differences might not be huge, but there’s also costs of a larger building to accommodate more people, pay for travel, etc.

• When a legislature is larger, more of the power is held by party leadership; these are largely the people doing the serious negotiations. This can make things more polarized, because the views of party leadership are likely more polarized than a lot of the legislators

  • Whether this makes things more or less democratic is debatable: you could argue it makes it less democratic because most of the elected representatives are doing less of the work, or you could argue that it makes it more democratic because people’s votes are largely de facto votes for party leadership anyway

  • But I think this is unavoidable until you get to a really small legislature, like 20 people. Still, it’s probably more prevalent in larger legislatures.

• If you have districts, once the district gets too small public awareness about the legislators can get lower because there isn’t as much media coverage on them.

I think a legislature of at least size ~500 on the national level is probably optimal; that’s the typical size around the globe. I think the first point about randomness is really important: you get a majority of 5 if a coalition gets 51% of the vote, anything smaller than that is really just asking for trouble. All the other points in favor of large legislatures are a nice bonus.

When the population gets smaller, the cons of a large legislature start to become more apparent, so you want to lower the number. I’m not exactly sure what size this is; maybe like 5 million?

An alleged empirical observation is that the size of a legislature goes up proportionally to the cube root of the population. I’m a little skeptical of this observation, I mean look at the fit here, but I guess you can judge for yourself:

Is the cube root a good formula below 5 million? IDK.

Procedure

The general procedure for passing a bill in almost all countries works like this:

1. A bill is introduced

2. (In some countries) there is a first round of voting on the bill itself and maybe on amendments among the whole legislature

3. It heads to a committee, where they vote on amendments and the bill itself.

4. It heads back to the legislature as a whole where they vote on amendments and final passage

For clarity I’ll use the American convention and call actions that take place among the whole legislature “floor” actions.

Leader Control

If you didn’t have a leader, how you would have a legislature function is make it so every bill that has majority support among the legislature passes. This is how I think legislation not subject to leader control should work, if you set out such categories of law. The policy that a legislature would agree to on these laws would be the Condorcet optimal policy (if one exists), the one that a majority of the legislature prefers to any other policy.

Once you have a leader though, that leader becomes the figure of accountability, so you want to give them power to control the agenda. As I described in my piece on the executive and the leader, there are two main ways of doing this:

• Give them veto power that can be overridden by supermajority

• Give them some number of votes in the legislature

You can implement these via “leader’s representatives”, people appointed by the leader to vote on their behalf who can be replaced at will.

I like veto power more on the national level. But whichever power you choose, I think you should make this power extend to all actions in the legislature, not just final passage. So if you have veto power, after a bill is introduced, the leader controls whether it goes to committee, then whether it gets back on the floor, then on every amendment before it passes. Each time, the legislature can override their veto only by supermajority.

The key step is that first one, when a bill is introduced. I want the legislature to work only on the bills that the leader wants them to work on, so the leader gets to set the agenda. I would even consider making the executive the only entity allowed to introduce bills in the legislature, which is the mostly the same in substance but different in appearance. The point of the rest of the steps is really to prevent the legislature from being hackish, amending the bill that the leader introduced to focus on something different.

Allowing the leader to veto amendments is also important for avoiding a bill simply reaching a “Condorcet final form”, and instead forcing it to look along the lines that the leader wants. If anyone can propose amendments and it takes just a majority to pass them, this gives all the leverage to the swing votes in the legislature. If there is a right-wing majority for instance and a right-wing bill, the left-wing minority legislators can offer amendments to make the bill less and less right-wing, and as long as each of these amendments is supported by the swing legislators they will get through.

Giving the leader power to veto at several different stages of legislating is not actually substantively different from just giving the leader veto power over the final bill. A legislature could do all of the negotiation before a bill is introduced, then immediately propose an amendment that changes the bill to what they want it to look like, and vetoing this amendment is basically the same as vetoing the bill. It is different in appearance though, which I think makes it less likely that this will happen. The whole committee and amendment process is also just a useful one that legislators might not want to skip.

Speaker

In addition to the leader of the country, there is also the leader of the legislature itself, which I’ll call the speaker for clarity as that’s what it’s called in the US and UK. They’re responsible for selecting which bills and amendments to bring up for a vote, deciding on the schedule of the legislature, sending bills to their appropriate committees, etc.

In the US, the speaker is a member of the majority party, and in fact broadly has the powers that I described the leader having above. In the UK, the speaker is a nonpartisan figure. In both cases, the speaker acts in accordance with legislative rules and norms, importantly those giving voice to the minority party(ies). I think it’s better to have the speaker be a nonpartisan figure to make it more likely that they follow norms. You’d select them the same way you’d select heads of non-partisan entities. The system I proposed in my other piece is to have a supermajority requirement and if the legislature fails to select one then all legislators (or some random fraction of them) become ineligible for re-election.

Committees

After a bill is introduced, and after a potential first round of voting/amendment, it goes to committee. In theory, committee is where the hard negotiation on a bill can take place. In practice, the real negotiation takes place only among members of the majority bloc (the only ones whose votes are actually needed to pass legislation) and in informal settings, so that they can be in secret, have staff present, discuss in smaller groups, etc.

The real function of committees is mostly twofold.

Inspection

The first is to ensure that somebody thoroughly inspects a bill before it passes. They may not be the only ones inspecting a bill, but at least they are one set of eyes on it. Members on a committee should hopefully be more invested in a bill during a debate on it than most legislators are during floor debate.

In most countries, the committee votes on amendments to the bill and then has to approve final text to send to the legislature as a whole. This is fine, although I also think it’s ok if the committee just gives a nonbinding recommendation on the bill and recommends amendments. In the US, there are subcommittees within these committees, and this is how these subcommittees work: they give a recommendation before a bill travels to the committee at large, recommendations which can be ignored by the committee as a whole.

Committees should have the same structure as the legislature as a whole. So have one leader of the committee, called the chair, who schedules and organizes debate. You can have them appointed by the speaker or elected by the members of the committee in a similar way to the speaker. It might be harder to find a nonpartisan figure as committee chair, so maybe I’d make it a partisan figure and just allow members of the minority to send complaints to the speaker if they feel they aren’t getting adequate say. Alongside the chair, I think you should have the same leader input as in the legislature as a whole, meaning a leader’s representative with veto power or a number of votes. These are important again for giving the leader agenda control and preventing the descent into Condorcetism.

I think you should make the debates in a committee secret, in order to maximize the amount of real legislating that takes place at this stage. These debates are usually conducted using some variant of Robert’s Rules of Order, a manual with rules on conducting debate. The general idea is that people can make motions, and if they get a second they go to debate; then during this debate people can propose amendments to the motion, which are then also debated on and can be amended. These seem fine.

Oversight

The other major role of a committee is holding hearings, in which both the majority and minority bloc legislators can ask questions of the executive or of other members of the public. The primary goal of these hearings is to give the legislators information that they can use to better fulfill other parts of their job like crafting laws and approving minister appointments. When it comes to hearings of the executive specifically, it’s important to separate out the utility for the members of the majority and the minority. For the majority bloc, these hearings allow them to keep tabs on what the government is doing so they can pressure them and if necessary call a vote of no confidence in a parliamentary democracy or (to a lesser extent) pursue impeachment in a presidential democracy. For the minority bloc, these hearings allows them to expose the majority’s actions to the public so they could hopefully win the next election.

On the question of whether these hearings should be private or public, I think a good mechanism might be that hearings can be private if a supermajority of the committee agrees to it. This will probably be if the members of the minority want to allow the majority to be more candid in their questioning of the same-party executive. If one of the majority or minority wants the hearing to be public, it should be though, so as to inform the public of the testimony as well. (Probably with an exception for classified information).

Specialization

In most countries, committees that inspect bills are standing (i.e., permanent) committees, inspecting all the bills that come in on a certain topic. The UK is unique for having committees formed with the specific purpose of analyzing a certain bill. I think standing committees generally make more sense to allow legislators to become more knowledgeable in a specific area. The UK does have standing committees for the purpose of oversight, and I think it makes sense that these committees use the knowledge gained from oversight to craft legislation as well.

The choice of which committee to send a bill to should be up to the speaker or other nonpartisan figures they appoint. I would also consider sending a bill to multiple committees in sequence if they touch on multiple areas.

Committees should in my view be about ~15 people. This means sacrificing quite a bit of representativeness of the legislature as a whole (which is why I would rather not give committees the ability to veto legislation), but ultimately I think any bigger and you start to lose the advantages of committees over the legislature as a whole.

In the US, committees are of size 30+ and what I described as a committee is the size of a subcommittee. But honestly I think these subcommittees should just be the real committees and you should skip the step where the committee as a whole debates. If you want to hold hearings where more than 15 people get to ask questions, you can have joint hearings between multiple committees.

Position Selection

You need some way of selecting members of each committee, as well as other party leadership figures like whip (the person who is responsible for counting and ensuring votes from their party).

I think party leadership figures should all be directly elected so that the legislative elite in a party is not all dominated by a single person. Secret ballot is fine, this isn’t something the public should be concerned with.

Committee selection is more interesting. I think every committee should be represented proportionally by party. For small parties (too small for one seat on a committee, also maybe not enough legislators to fit on every committee) I would err on the side of letting them on committees; the one thing is you need to make sure the majority coalition parties have a majority on each committee.

Then you have to decide which legislators from each party get each committee spot. What’s usually done in practice is that party leadership selects everything. I like this for its tendency to reward better, more active legislators, but I really dislike how this rewards party loyalty. I think there is already enough incentive for legislators to be loyal to their party through party candidate selection, and legislators should already be pretty aligned with the party. If you do decide to go with leadership selection, which I don’t think is outrageous, the one thing is I wouldn’t let leadership remove people from their committee posts if they’re disloyal.

The alternative is probably to come up with some ranking for legislators, and then have the legislators sequentially pick their committee spots: the first one on the ranking gets first pick, then the second one, etc. This allows everyone to pick the committee they’re most interested in, while also rewarding people higher on the ranking. If you want people to be on multiple committees, you could have two rounds, the first time everyone gets to pick their first committee, and the second time everyone gets to pick their second committee.

There are a few options for this ranking:

• If you have list proportional representation, you can make it the list ranking. See my post on party candidate selection on how to choose that list

• Order of seniority (how many terms a legislator has served for)

• Random

I think any of these are defensible. I would also consider allowing party leadership to determine the very top of the ranking so they can choose who gets to be on the powerful committees, so that hopefully more competent legislators take those spots.

To promote continuity, you can also make it so that people are guaranteed a spot on eg 1 of the committees they’re currently serving on, so that they aren’t constantly shuffled between committees.

I also think two people, not necessarily from the same party, should be able to swap committee appointments. If you have party leadership select positions and want to get really fancy, you can turn this into a stable matching problem. Party leadership lists a ranking for each committee of who they think should get the spot, every legislator lists a ranking of which committees they want to be on, and then you calculate which legislator belongs on which committee to make it so that there are no remaining intra-party swaps. There is a nice algorithm to calculate this.

Floor Debate

Various thoughts on floor debate:

• Most legislatures also abide by some variant of Robert’s Rules of Order for floor debate. The one part that I am a bit wary of for floor debate is the idea of unanimity and seconds - as a legislature grows in size, you are less and less likely to see unanimity and more likely to have ideas be seconded. Maybe I’d make it so that a constant 1% of the legislature has to object for a motion not to be unanimous or for it to be considered seconded.

• Generally what you want out of floor debate is that people from both sides can give a bunch of speeches on the bill before you pass it. My hope is that these speeches are primarily meant to convince other legislators of their position. In practice, though, these speeches tend to be more targeted at the public, and I am skeptical that they actually accomplish much because how many people are watching legislators give speeches on C-SPAN. The one purpose I can see public-facing speeches actually serving is as a signal of the views of the legislator giving the speech.

• The speaker should select all bills and amendments that seem likely to pass (i.e. have the leader’s blessing + majority support, or have supermajority support). One fun mechanism here that bypasses the speaker is to force a piece of legislation (or amendment) to the floor if it gets the signature of a majority of the legislature (without getting the veto signature of the leader)

• The full text of a bill should be released like a week before it is debated for smaller bills or a month before for larger bills, so that people have time to read it before deciding what amendments to propose and whether to vote for it. This makes it harder to pass bills because people will have more time to develop objections, which I think is valid: people shouldn’t be voting for things they object to. You also get much more polished bills when you give people time to read them. (I’d make an exception to this scheduling constraint for bills that get 2/3 support; in general I think a 2/3 majority should be able to do whatever it wants, bypassing any and all legislative rules.)

• I think floor debate should be public. I don’t expect much serious negotiation to occur in the form of legislators giving speeches on the floor anyway, and I do think it’s important for the public to see the amendments that are proposed and who votes for them.

Factions

In the US, there are “caucuses”, groups of legislators (usually from one party) that share common views. This is helpful in allowing them to vote as a bloc to get more of what they want. These are good and I’d encourage any legislature to have them. It’s good when legislators can use their leverage to demand changes.

This is the second part of my post on electoral systems, this one covering multi-member electoral systems for electing a legislature. You don’t have to have read the first part on single-member electoral systems, but it is helpful.

Don’t Use A Non-Proportional System

A common multi-member electoral system is to have single-member districts and use one of the previous electoral systems to elect them.

There might be some argument for using this method with an electoral system that elects a moderate/Condorcet winner. Even there though, I think if you want a moderate/Condorcet winner, you should just go with a proportional voting system instead, so the moderate party ends up being the swing vote in the legislature.

Usually, however, people argue for non-proportional voting system using one of the two-party electoral systems because they don’t want their politics devolving into a bunch of small parties that make governing and therefore accountability democracy harder. But there is a better way.

You can force a proportional electoral system to elect larger parties by adding a threshold, a % of the vote a party must cross to get any seats. If you add a threshold, please use ranked voting so that people’s votes are not wasted if their preferred party doesn’t make the threshold. This is called Spare Vote, and works the same way as ranked choice voting (RCV) does for single-seat elections, but instead of stopping at 1 candidate you stop once every remaining party is over the threshold1. It comes with the same downsides of RCV—for instance you should probably have two rounds if you set the threshold really high—but the downside to throwing out votes that don’t make the threshold is even bigger than the downside to using first-past-the-post. This is because a lot of people are going to vote third-party in a proportional system.

High thresholds push a country away from consensus democracy and toward accountability democracy. High thresholds also force voters rather than politicians to form coalitions. I generally prefer low thresholds, but it’s nice to have a single parameter to adjust. If you want a two-party system, for instance, you can raise the threshold to 1/3 (or more accurately you’d just seat the top two parties so that a party with >2/3 support doesn’t get all the seats).

You might still favor single-member districts because you want seat share to go up faster-than-proportionally with vote share, for more stable majorities. Even then, though, you shouldn’t use single-member districts. The relationship between seat share and vote share in them is dependent on the weird geographic property of the partisan distribution of districts—the more swing districts there are, the larger majorities will be. (Also you have to deal with gerrymandering).

The better way to implement this greater-than-proportional relationship is to have some formula to give a boost to the winning party, often called a plurality bonus. For example, you can adopt a two-party proportional system (ie only seat the top 2 parties), and then adopt the following formula: a party gets an extra 1% seat share for every percent above 50% they get in vote share, up to 55%. This yields a nice predictable relationship between vote share and seat share.

(I don’t think you should use plurality bonuses outside of a two-party electoral system, because it can give a minority bloc a majority if it happens to have the largest party. For example a right-wing party gaining 40% of the vote can win a majority if the opposition left parties have 30 and 30% respectively. But it is an option.)

Now, onto the forms of proportional representation.

List PR

Closed List

Closed-list Proportional Representation (PR) is the simplest. Each party issues a list of candidates before the election, they’re allocated seats proportionally to how many votes they get in the election, and the top x from their list get seated (where x is the number of seats they were allocated).

The upside is that this system is simple and immune to gerrymandering or any other weirdness. There are two main downsides:

1. Poor candidates can’t be punished as precisely. People can’t vote against bad, maybe scandal-ridden candidates directly, they can only vote against their party as a whole. By the same logic, good candidates can’t be rewarded as precisely.

2. There are no representatives from swing districts to feel the heat.

   • An important reason that legislatures act moderately is that politicians from swing districts know that if they don’t vote moderately, they will lose their seats. You might see this phenomenon in party list PR with politicians low down on their party’s list, who know that if their party passes extreme stuff it will lose popularity and they will lose their seat. However, they don’t feel the pressure from voters examining their specific voting record and choosing parties based on that.

One thing I don’t think is a downside is the lack of geographic districts.

If the point of geographic districts is that people can call their representative, national politicians can just, like, hire more staff, eg have a staff member “representing” every region. I don’t really see what having politicians who represent local districts really accomplishes here, legislators have to hire a lot of staff for the 500,000-person districts we have in the US anyway.

If the point of geographic districts is that local communities have someone voting on their behalf, I say that’s stupid and a community shouldn’t have more power if it’s clustered geographically. If a community is big enough it should get national representation, either with a party of its own or through concessions from larger parties.

District-Level List PR

You can mitigate closed list PR’s drawbacks (particularly the first one, about precision) by doing list-PR on a district level, so that voters have more knowledge of who they’re voting for. Note that you’d still implement thresholds on the national level2, and you’d still combine this with spare vote.

The issue is the number of seats (ie, legislators elected) per district. The more you have, the lower voter knowledge of individual candidates will be, so it’s easy to end up back at national-level closed list PR. The fewer you have, the more you sacrifice proportionality, and the more vulnerable you are to gerrymandering.

The optimal seats-per-district count also depends on how many parties you expect to have. If you have a lot of parties, you want more seats per district. This is for two reasons. The first is that proportionality becomes more important, particularly for smaller parties. A party with 5% of the vote may not get any seats in district-level list PR.

The second more subtle reason is that when there are few parties but a lot of seats per district, there are more legislators elected per party in each district, so public knowledge of the candidates they are voting for is going to be lower. When there are lots of different parties, even if there are many seats per district, each party may only be vying for 1 seat or so. A prospective voter can therefore mostly just examine the top candidate for the 2 or 3 parties they’re considering voting for.

Let’s use Ireland as a case study for how this looks in practice. They have a system similar to district-level PR that I’ll talk about in a bit, and they have 3-5 reps per district with no threshold. Their elections look like this, with 3 main parties and a few smaller parties:

So that’s about a 1-2% difference on average between a party’s seat share and vote share. It’s pretty common for elections to be decided by 3% of the vote, so I think this level of disproportionality is uncomfortably high. Look at 2016: the center-right parties Fina Gael and Fianna Fail combined for 49.9% of the vote but got 56.6% of the vote, enough for a clear coalition that can handle a few defections.

The other thing to note is that the geographic distribution turns out not to punish small parties here. I think this is partially due to small parties not running candidates in districts they have no chance of winning. I do still think this is an issue, but it’s good to see that it’s not as big as you might think.

In Ireland, in the 4- or 5-seat district, it’s very rare for parties to elect more than 2 legislators per district. In the 3-seat districts, it’s rare for any party to get more than 1 seat. These 5-seat districts are right on the edge of the level of knowledge that I think you could reasonably expect a lot of voters to have. Any more and it’ll basically be just be list PR.

I’d want to try this out in more places before rendering final judgment, but based on Ireland I think 5 reps per district is probably the magic number. This should be a bit more proportional than the Irish election, while still being somewhat manageable. Unfortunately you are still sacrificing quite a bit of both proportionality and public knowledge, but I think those are bullets you have to bite.

I also wouldn’t bother forcing a two-party system using thresholds if you adopt district-level list PR. There would be like 4 candidates per party in each district in that case, so good luck getting voters to consider all of them. You are bound to essentially eviscerate public knowledge.

Open List PR

Open list PR is very similar to closed list PR, except voters get a bit more control over the ranking. The most common method is that people can vote for a party and then put a “check” for specific candidates, and then a candidate guarantees a seat in their party by getting a certain threshold of checks (either in their district or nationally), regardless of their spot on the list. You can also do this in a negative way, where voters can “X” candidates they don’t want, and then a candidate is denied a spot if enough people X them. (You can also consider checks minus Xs as the deciding number.)

This functions as a form of pseudo-primaries. I generally think there’s not much point - it gives voters a very small degree of control - but it’s also not very harmful so I guess it’s fine. It does more in district level PR, though at that point you might also consider STV, a system I’ll get to in a second.

Direct Multi-Member Systems

This bucket is of electoral systems that directly elect multiple members in a proportional way. Usually these methods are extensions of single-member electoral systems: they have proportional results if everyone votes strictly by party line, and if not they function similarly to their single-vote counterparts. They are implemented in multi-member districts, which has the same size considerations as district-level PR: I would probably go with 5 votes per district.

The most notable of these systems is Single-Transferable-Vote (STV), which is the extension of RCV. Among the other systems are Schulze STV and CPO-STV, which are extensions of Condorcet ranked methods, and Proportional Approval vote, which is the extension of approval voting. You can read about those methods if you want to understand them, I don’t think it’s worth my time explaining them (especially since I don’t understand how Schulze STV or CPO-STV work).

These are the only systems that can handle a country where polarization is very low and people care more about individuals than parties. Honestly though if a country gets to that point, voters are probably voting for people that don’t represent them very well on the issues, so I don’t think this is a good thing to encourage.

The other big thing about these systems is that they’re extremely complicated, to explain/implement and to some extent for voters to use. Strategic voters seems like a nightmare to figure out in these systems.

One final practical note about these systems: I think they’re the only ones that are consistent with the US Constitution for House elections, which means that they’re the only ones even worth discussing in the US.

STV

Let me dig further into STV, the only one that I really understand. Probably a lot of what I say applies to these other systems as well.

Feel free to read the full description here, but you can think of STV as an extension of district-level closed-list PR. (STV is the actual system that Ireland uses). There are three main differences from closed list.

First, voters of a party can choose the order of the candidates in their party. I see this as mostly a bad thing: if you want this behavior you can always just implement it in separate primaries, and there are good reasons you might not want it. Second, swing voters can split their vote up, which can cause a less popular candidate of a party to lose a seat to a different party. This probably doesn’t help in electing moderates (they would be less popular among base voters, who are the main ones deciding which of the party’s candidates get seated), but it does help in punishing scandal-ridden candidates, since party voters probably have them further down their ranking and swing voters can just not vote for them. Third, independent candidates can actually win seats.

A note: you can allow parties to give a recommended ordering of their candidates. You can even let voters “check” a party list they want to use it instead of having to manually rank candidates in the same way. If most of the party’s voters stick with that ordering, it functions a lot like open list PR, where most of the candidates come from a list but a candidate low down might make it up if voters really like them.

Proportional Multi-Vote Representatives

I would also be remiss not to take some time to talk about Proportional Multi-Vote Representatives (PMR), a proposal I wrote about a long time ago, which also fits into the category of direct multi-member systems. That system does away with the idea that every representative gets equal vote in the legislature. Instead, each legislator has some number of “legislator-votes”, and votes on bills in the legislature are tallied by counting the legislator votes rather than by counting people. Each district has some number of legislator-votes to give away, and every representative gets legislator-votes proportional to how many public votes they got in their district (and are seated if they get 1 legislator-vote or more). You can have as few as 5 legislator-votes per district, but this system can handle a lot more. You don’t even really need districts, you could technically have the entire country be one district (which functions a lot like closed list-PR). The downsides of having many votes per district is that the legislature size can become enormous if votes are spread out among multiple politicians, or if they’re concentrated among few politicians you get a small legislature and thus start to lose the advantages of a legislature to begin with. Probably I’d have districts of like 10 votes each.

As with closed list PR, you don’t have representatives from swing districts who “feel the heat” as much because (as long as they aren’t at 1 legislator vote) they aren’t going to lose their jobs if they do poorly. Although you can still give reps with fewer votes fewer privileges in the legislature, like being in committees, to add some extra incentive.

The other potential downside is that there is probably significantly less turnover in this system than in other systems if you have like 10 votes per district. I tend to actually like this for allowing legislators to develop experience serving, but you might not.

I really like this system. It’s a bit speculative, obviously untested, but it seems really good.

Mixed Systems

Mixed systems are systems that have some district-level elections, but also some mechanism for ensuring that a party’s seat share is directly proportional to their vote share.

There are two methods I know of. There’s mixed-member proportional (MMP), where there are district seats usually elected in normal FPTP elections, as well as list seats given to make the party vote proportional and which are selected from a party list. And then there’s dual-member proportional (DMP), where each district elects two legislators, one by normal FPTP, and the other chosen to make the national results proportional.

In MMP, instead of FPTP, you can have each district elect multiple reps with a different system, such as district-level list PR. This fixes the disproportionality of district-level list PR, while still keeping the number of list seats much lower than normal MMP. I really like this system; the main drawback is just that it adds some complexity.

A lot of the properties of these systems depend on implementation details. So let me make some broad comments about these systems:

• To the extent that a mixed system relies on list seats, it functions similarly to closed-list PR.

• To the extent that a mixed system sees candidates from different parties compete against each other in each district, it has the usual nice properties of geographic districts: scandals are directly precisely, politicians in swing districts are incentivized to moderate.

• To the extent that a mixed system sees candidates from the same party compete against each other in different districts (vying to get more votes in their respective districts), candidates have the incentive to moderate.

• To the extent that a mixed system sees candidates from the same party compete against each other in the same district, these systems function like primaries, which as I stated before I think is mostly bad. You can just implement separate primaries if that’s what you want.

• If you have a separate party vote and person vote (like in some variants of MMP), combining it with Spare Vote is sort of annoying, because you probably want ranked voting both in the party vote and the person vote. This is why I prefer single-vote systems. (Also they’re just simpler). The downside of single-vote systems is that they mess up strategy a bit: if you vote for a party because you like their candidate in your district, but then their candidate doesn’t get the seat, your vote goes toward their list seat candidate (whom you might not like).

• MMP is kind of complicated, and DMP is even moreso. MMP has been adopted in a few places, notable Germany and New Zealand.

Recommendations

My favorite system is my novel PMR system regardless of threshold. But if you want a more tested system, my favorite with low/no thresholds is closed-list PR, and with a high threshold it’s MMP. If you have high thresholds but not a two-party system (I’m thinking like 10-15%), I would go with MMP but using district-level list PR for the geographic seats, districts of size 3. It has all the properties you could possibly want, and since you have high thresholds there should be very few list seats. It’s a little complicated, but I think once you know how thresholds work it’s really not difficult to see how the list seat part of MMP works.

I’ve previously covered the topic of electoral systems, different ways to decide which party wins spots as leader or in the legislature. But this is only half the problem of how politicians are elected. The other half, less discussed but in my view equally important, is how you decide which candidates the party puts forward. That’s what this post is about. It’s a long one, but I think its length is warranted for how important and complex a topic it is, and there wasn’t a super natural place to split the piece in two.

Why This Piece Exists

In most countries, candidate selection is left over to the parties themselves. So that’s it, just let the parties decide - piece over, right? Not so fast.

Even if you let the parties themselves decide, it’s still worth thinking about what the “best” decisions are. Many people use appeals to democracy to justify primaries or appeals to organization to justify elite selection, and it’s worth understanding how to evaluate these claims in light of broader constitutional design considerations.

There are also useful resources that the state can lend to parties for candidate selection, even if the parties decide whether they want to use them. The state can help facilitate primaries, for instance, instead of forcing the parties to figure out how to ensure secret ballot, accessible voting, and minimal fraud. The state can provide financing to parties to help them select candidates.

Or maybe you don’t want to let the parties decide how to select candidates! You might think that candidate selection is a problem of such importance that it’s worth the state mandating a process to ensure that it’s good or fair.

I’ll talk about how much you’d want the state to intervene at the end of this piece, but I thought I’d preface the piece with this as it’s many people’s gut reaction to this question.

The Party

A political party is an organization with two main functions. For low-information voters, it lets them easily assess a candidate’s political views without having to look into them further. For high-info voters, parties allow those with common political views to coalesce around certain candidates for a general election. Parties are a natural phenomenon; if they were banned, they would re-emerge informally in the form of elite endorsements.

On the one hand, you want party unity, so that people know what the political views of the party they’re voting for are. On the other hand, you don’t want too much unity, or else you lose many of the advantages of legislatures that I wrote about (namely, that they’re pluralist and moderate). It’s particularly important that a party is not totally beholden to its leader, as this increases the randomness of a legislature’s views and can lead to dictatorship.

As per usual in this series, what I’d like are dials you can adjust based on your preference for diversity vs. uniformity. There are two good ones. The first is changing who selects the candidate, which I’ll call the “selectorate” based on this paper (though I use it slightly differently here). When you make the selectorate more exclusive, more open just to those who are engaged or connected enough in the party, you’ll tend to get much more uniform views. The second is the electoral system - how proportional vs. majoritarian it is. This second one obviously doesn’t apply to selecting candidates for a single leader.

The Selectorate

Public

These are primaries. The advantages of them are:

• They’re democratic, which means the views of politicians more accurately reflect public will.

  • As I discussed in my piece on direct democracy, you probably don’t want politicians to be able to freeze out the public view if there’s bipartisan elite consensus on an unpopular position, and primaries are a way of preventing this.

• They help avoid “backroom politics” in deciding who gets to be a candidate, which can select for corrupt and otherwise ruthless politicians

• They’re a good way to ensure party diversity in a legislature, particularly if you pick a good electoral system.

The main drawback is that direct primaries select for candidates who are popular among the primary electorate, rather than those who are better at governing in good and popular ways. In this way, they can undermine accountability democracy, as well as just generally leading to bad governance.

One other potential drawback of direct primaries from the party’s perspective is that primary winners are likely worse at campaigning for a general election. When elites choose candidates, particularly a candidate for leader, they are very concerned about winning the general election; the public not as much. Whether this is a drawback from a constitutional design perspective kind of depends on how you view campaigning.

The options for public selectorate differ primarily in their level of exclusivity, how much they select for engagement. As mentioned earlier, a more exclusive selectorate will lead to more uniformity. It’ll also tend to lead to more extreme candidates (see here) and candidates that appeal more to people of high socioeconomic status. Let’s go over the different options.

Party Voters

The most inclusive possibility for the public selectorate that is still defensible (IMO) is everyone who voted for the party.

The straightforward way of doing this is to host primaries where people who voted for the party last election, or maybe in the last few elections, can vote. With closed-list PR or with presidential democracy, you can even get fancy and host the general election before the primary, so first you decide how many seats each party gets / which party gets the president spot and then the parties decide which candidates takes those spots.

The hard part is checking who voted for the party. There is a good way of doing this that preserves secret ballot - hand voters a usb stick or barcode which contains an encoded version of who they voted for, and let them use it as they vote in the primary. But you (and the public) have to trust that the government will do this correctly, and if not it can make people uncomfortable and lead to social pressure to vote for certain candidates in the general election.

The other way to implement primaries for all party voters is just to use an electoral system that allows people to vote for not just a party but candidates within that party; see my electoral systems post for more about that.

Party Members

If you find the logistical issues of party voters too difficult, or if you want a more exclusive system, you can opt for party membership as the requirement for getting a vote in candidate selection.

You want to make sure that party members are actual supporters of the party. Otherwise, nobody has any incentive to vote in their own party’s primary: they should strategically vote in whatever primary would be most beneficial. Voting in a primary for a party you don’t support is called crossover voting. I’m impressed by how rare it is in the US, but its existence is obviously a design flaw, as it means the parties wouldn’t necessarily have internally-unified beliefs.

There are a number of different requirements for becoming a party member to ensure this:

• Fees

• Length of party affiliation

• Sponsorship by a party member (ie, someone else attests that you are a committed party member)

• Party activity (attending local party meetings or campaign events)

Each of these selects for party engagement to a different extent. Most also select for other attributes that I think are undesirable: fees select for wealth, sponsorship selects for sociability, party activity selects for outgoingness as well as for availability (which tends to correlate with age) and living in an area with lots of events (which tends to be urban areas, area where the party has lots of support, swing region with lots of campaign events).

The most commonly used one and the one I think you should probably opt for is fees. By paying a fee, you’re hopefully helping the party win elections, so you shouldn’t do it unless you’re actually a supporter. I also think a fee creates a sort of coordination problem that discourages crossover voting. In order for it to be worth it to pay a fee to the other party and vote in their primary, you have to be assured that your vote will actually matter, else that’s just money you’re giving up. It becomes more worth it if the primaries are done with a proportional system (which I think they should be, as I’ll get to in a bit), but even there the effect remains at least somewhat.

(Even in a winner-take-all environment, the effect isn’t perfect; crossover voting becomes more worthwhile when the party you’re considering voting in is closely divided and you just have to help push a candidate over the edge, or when you have a bunch of legacy party members who are now more crossover voters but haven’t yet ceased their membership).

To give you a sense of how fees look like in practice, in the countries I looked at, the base fee ranged from $20 to $120 annually, often with lower fees for certain groups (eg young people) and higher for higher income people. In most places, the fee is the only requirement for becoming a party member (beyond non-binding stuff like alleging that you support the party’s principles). In places where they charge a fee, party membership is somewhat rare. Here is a chart I could find for membership in the late 1990s, and note also that party membership has been in steady decline:

Source

There is of course the issue of poor people not being able to pay the fee. You can get around this by having the government facilitate the financing: charge people for party membership commensurate to their income (ie more if they’re richer), and then pay parties with government money proportionally the number of members they have. I quite like this mechanism.

There is also another issue, which is that fees may not actually help a party win elections that much, so it might be worth it to pay a fee and vote in a primary for a party you don’t actually support. You can try to get around this a bit by adopting tight restrictions on the ability of parties to raise money by other means, making the value of party membership money higher, though as I talked about in this piece that might be hard to do.

Other than fees, I think length of party affiliation is a no-brainer, with some exception for young people who cannot have been affiliated with a party for a long time. I think sponsorship by party members or party activity are options to explore as alternative routes to becoming a party member alongside fees, rather than extra requirements in addition to fees, as I’d guess their selection effects are really bad.

Voting at Party Meetings

Instead of making attending local party meetings a prerequisite for voting in the primary, you can have the voting take place at the local party meetings themselves. In the US these are called caucuses.

This has the same selection effects as making attendance a prerequisite for party membership. It also has the downside that you don’t have secret ballot, which means people can be socially pressured to vote for certain candidates. The main upside I can see compared to a party membership prerequisite is that it opens up potential flexibility to pursue different voting systems like single-transferable vote (STV), as you can explain the system to members before they vote and people can discuss all the candidates with each other.

I should also mention open conventions, which are like massive party meetings. Like party meetings, they have the draw of potentially opening up more advanced voting systems. They’re used mainly because they’re just much logistically easier than a whole caucus or primary system, but I think the massive selection effects make it just a bad option unless you logistically can’t do anything else, perhaps for really tiny parties.

Party Elites

The alternative to letting the public select candidates is giving that power to party elites.

When a party is not new, elite selection is mostly a form of continuity. You expect the candidates last time to be mostly the same as the candidates this time. This can be a good way of ensuring a central long-lasting party ideology. Of course these same people can change their minds on an issue, particularly if there’s incentive to do so to win the general election, but it’ll generally be the same people. The only way to break this continuity is to let the public in.

With a new party, the elites are basically whoever started the party. When Nigel Farage started the party Reform UK, he and his close friends get to choose who the candidates are.

Now let’s go through the different options for which elites exactly could get candidate election power.

Party Leader

Giving power to the party leader is the single biggest way of promoting uniformity within a party. One person mostly has a single group of policy stances.

You might object that because of randomness, it could be that the leader has a less “pure” version of the ideology that the party espouses, and thus it’s not the best way of ensuring a uniform vision that people can vote for. But I think people generally associate the views of the party with the views of its leader, so the leader’s ideology actually becomes the pure ideology. The leader also likely comes from the biggest faction of the party and so is more likely to be representative of the party’s majority view.

National Politicians

This is the purest form of a feedback loop that promotes continuity - politicians select candidates that become the party’s next politicians. (Likely with a level of indirection in between, which I’ll touch on in a bit).

Legislative Party Leadership

This is mostly the same as giving power to the party leader, just for a few people rather than just one person. The main difference is that the party leader can be chosen by the public (more on that in a bit), breaking the continuity, while legislative party leadership never is.

Local Politicians

Having local politicians select candidates is a way of making politicians more in touch with local issues, which can be a pro or a con.

How it affects big-picture factional disputes depends on how these local politicians are chosen, which opens up this same conversation but on the local level.

Indirection

Instead of having the selectorate choose candidates directly, you can have them select a candidate committee that then chooses the candidate. This is what’s done in most countries. (You can have even more layers of indirection, like having this committee choose party leadership that chooses candidates, but I think generally there’s not much point).

For elites, this is useful so they don’t have to take the time to learn about every single candidate. For the public, this is primarily useful when things aren’t split up neatly into single member districts where everyone focuses just on the candidate in their own district, so you need some central committee to select a slate of national candidates. Even if you do have single-member districts though, standard primaries in geographic districts don’t allow for any ideological diversity unless certain factions are clustered geographically, so I think a candidate committee is a must-have with them as well.

New Parties

New parties are where I think candidate selection restrictions becomes tricky.

You want to have restrictions to prevent a party leader from totally dominating their party. The issue is that a popular party leader can always start a new party if they feel like the restrictions are making their party’s candidates too disloyal. Then, they can appoint loyal elites, and will likely have a loyal base that follows them to their new party as party members - thereby essentially bypassing restrictions. This is especially a problem if a leader has been around for a long time, giving them a better idea of which elites are loyal to them and which are not.

The most direct way of addressing this loophole is to put a cap on the number of seats that a new party can win. This is like a threshold but with a max rather than a min; as with min thresholds you use ranked voting to make sure votes above the cap aren’t wasted. This forces a party to grow over time rather than immediately get a majority. In practice, this shouldn’t matter that much: growing over time is what a new party does most of the time anyway, it’s rare for a party to immediately get a majority.

(There are some more technical details to work out to fully close the loophole through this mechanism; read this footnote1 for more.)

Besides this direct mechanism, I think to some extent you just have to rely on the public and elites not being endlessly loyal to a new party leader2. As long as there is some minority of the party that shows real independence, that can be enough to add important diversity.

Recommendations

For electing a single party leader, I think you should opt either for public primaries for all party members or selection by legislators - I don’t think anything else makes sense. You can even do a mix, where legislators narrow it down to a few people and party members select among them, or vice versa, more on that in the next part.

For legislators, I think the cleanest way of approaching candidate selection is to have a candidate selection committee of like 30 people. These can be selected by whatever means you want - appointed by the leader, by legislators, by party members, etc. You can also combine delegates chosen by different means, some by party members and some by elites, which is what parties in most countries do.

A good system would maybe be that 30% of delegates are chosen by the leader, another 30% by current legislators, and the other 40% chosen proportionally by party legislators.3 You could also easily make this multiple regional candidate committees instead, where the percentages hold for each individual committee.

Electoral Systems

Everything up until now has been about the “who” of party candidate selection - who makes up the selectorate that chooses the candidates. This (shorter) part is about the “how” - how you translate the selectorate’s preferences into candidate choices. I’m going to make my posts on (inter-party) for a single person and for a legislature required reading for this part; I’m going to reference them a lot.

Single Leader

There are two parts to candidate selection: narrowing the field down to a few people, and then deciding among them. What I’ll say here applies to both if you have direct primaries for the leader or if you let the party’s legislators choose them. You can even mix and match and have the first part done by party leaders and the second by legislators, or even vice versa. If you have the second part done by the public, in parliamentary democracy you can even implement it during the general election if you want the selectorate to be all of a party’s voters: people choose which party they would like to see take office, and simultaneously vote for a leader of that party. This is kind of a weird thing to do - since you don’t know who the leader is going to be until the election, people don’t know exactly who they’re voting for, and you also can’t give the leader extra power in candidate selection - but it’s possible.

The way to implement the first round is to use FPTP and have the top X candidates from one round, or maybe all candidates that get above Y% of the vote, make it to the next round. Maybe this system could be good: a first round using FPTP with a 5% threshold to advance, and then a potential second round to whittle it down to 4 candidates (if more than 4 advance from round 1).

The hope is that once you’re down to a few people, the voting system you use for the second part won’t matter that much. Things should be a lot less factional at the party level than in a general election, so the plurality winner, the condorcet winner, the utilitarian winner (not used in an ethical sense but in a voter satisfaction sense, see my post on electoral systems), etc. should hopefully be the same person.

Unfortunately, sometimes things are quite factional. Sometimes, for example, there is one really polarizing figure that emerges as a frontrunner - Trump 2016, for example - and the public splits for and against them. In these cases, I don’t think it’s clear what you should do; I could see a case for any of the single-member systems I talked about in my post on electoral systems. I think the “ideal” for party selection selection is probably to choose the utilitarian winner - this is the person who I think can best said to “unify” the party. For this reason I’d go with score voting, which hopefully should lead to the utilitarian winner if people are honest, and if people are strategic it’ll lead to the something between the condorcet and plurality winners, neither of which are terrible.

Committee Selection

By The Public

As stated previously, I’d like for there to be publicly elected members of the central candidate selection committee in charge of selecting legislative candidates. I’d like this to be proportional, with the majoritarianism part introduced by the other leader-selected delegates.

I would have these members chosen by a direct multi-member electoral system like single transferable vote (STV) or proportional approval vote (PAV). If you have more than 5 members, you can have this take place in many different districts. Alternatively, if you want more granularity, you can have the public elect a body of 100+ delegates in districts, and then have these delegates meet at a convention to select candidate committee members. In that case I really like my proportional multi-vote representatives (PMR) with districts of 10ish member votes for this.

Exact proportionality is less important on the party level (because the stakes of who gets a majority are lower), so I wouldn’t mind going down to 3 members per district if using STV or PAV in order to reduce exhausted ballots (ballots that go to waste because a voter didn’t rank or rate enough candidates). That said I also think 5 is ok for letting smaller factions take power.

I do worry that elected committee members aren’t going to be super representative of the public because people don’t know who these candidates are. There aren’t as many “issues” for delegates to run on because they don’t do as much. I am hopeful though that voters will be able to be able to assess things like whether a delegate is a “party insider” or “leader’s pet” or “random businessman”, which should be indicative of important factional disputes.

By Legislators

How you have legislators elect members of the committee is pretty flexible.

You could have them select members by STV or PAV in 1 big election. Alternatively, you could split the legislators up into groups and then have each group elect committee members by STV or PAV. Those groups perhaps being geographic regions if legislators are elected in districts, or if legislators are elected in a party list you could cut up the list somehow (eg spots 1,6,11,16 are 1 group, spots 2,7,12,17 are another,…). Or another thing you could do is just let the legislators manually sort themselves into groups and have each group unanimously agree on delegates proportional to how big it is.

I would use secret ballot for these elections.

Legislative Candidate Selection

Regardless of which inter-party system you use, you want to generate a ranking of legislative candidates: which candidates will be seated if a party gets 20% of the vote, which will be seated if it gets 40% of the vote, and which will be seated if they get 60% of the vote. For closed list PR, this is just the party list. For MMP, this is which candidates are in safe districts, which are in swing districts, and which are in long shot districts. For multi-member districts like STV, this ranking is handled by the voters themselves, by which candidates get more of the vote.

I’m ok with giving the leader’s committee members all the top spots on that ranking. (if you do that, maybe I’d reduce the number of delegates the leader gets, maybe 15% instead of 30%). But for the rest of them, I think they should be selected by the candidate committee proportionally. This preserves the proportionality of the elected delegates, which means we can neatly control the amount of majoritarianism through the amount of leader-selected delegates. More precisely, what I want is that not just that the overall ranking will be proportional, but that every contiguous sublist will be proportional. That way, no matter how many seats a party gets, the candidates will be selected proportionally to the delegates.

The easy way of doing this is to let each committee members select an equal number of candidates, and then interlace the candidate spots evenly. So if you have 30 committee members and a 60 candidate list, member 1 can get spots 1 and 31, member 2 can get spots 2 and 32, …, with some ranking for the committee members. If you give the leader’s committee members all the top spots, you would use this same system for deciding on committee spots between them. Once a committee member gets a spot in the ranking, they get to control which candidate is chosen for that spot. They might consult with other committee members, or with local party branches, but they are ultimately in charge of choosing that spot.

So that’s the method for when you have closed list PR as the inter-party electoral system. For inter-party mixed member proportional (MMP) or district-level closed-list PR, you essentially replicate this by letting the first place in the ranking choose any district (and get the top spot on that district’s list, for district level PR) first, then the second place gets to choose any district (and get the top remaining spot on that district’s list), then the third place, etc. People should choose the seats that are the safest first, and potentially the seats close geographically to them. If you want delegates to take the seats closest to them geographically, you can have this process take place in multiple regional candidate committees. Unfortunately, unless you have big regions - probably regions that are expected to elect 5 candidates *from that party* - you aren’t going to get a level of proportionality I think is acceptable. The only time I think this is potentially feasible is if you have a two-party inter-party system. (This is the only time I’d consider MMP anyway, so maybe then you could do it).

For inter-party multi-member systems like STV, you run into an issue. Since the voters themselves pick who the top spot on the list is, it’s likely that (if a district elects fewer candidates than the party nominates) the biggest faction of a party takes all the top spots. The only way around this is to let each committee member choose *all* the candidates in each district, which defeats some of the point of multi-member systems. Maybe you’re ok with sacrificing intra-party proportionality in these systems, but I’m not, which is why I generally dislike these systems.

State Intervention

The final question is how much of these processes the state should mandate.

If you have a proportional system with no/low threholds, I’m ok with keeping things a bit looser. But if you have high thresholds or plurality bonuses, I’m in favor of mandating a lot. In that case I think you should mandate at least:

• That candidates are selected by one or more committees

• That these committees are composed at least 70% of members elected by a party’s current legislators and/or the public.

• That these committees choose candidates proportionally, satisfying the contiguous sublist proportionality I described, except at max 15% of candidates.

• That the members selected by legislators are selected proportionally.

• That publicly elected members are selected by a proportional system

• That this “public” includes people in any geography (ie not a single convention, or if you have local party meetings you need them throughout the country).

I’m ok with mandating a lot more though - the proportional system you use, the exact percentages, etc.

I think if you don’t mandate, there are going to be different pressures that push candidate selection in different directions - elites will want to have more elite say, the public is going to want more public say, the largest faction will want a less proportional system and the smaller factions will want more - and there’s no guarantee you’ll get a good system out of it. This can seriously undermine democracy.

Voting Rights

Let’s start with voting rights. There are a number of decisions to make on this front:

• What should the age threshold for voting be?

• Should prison inmates get to vote?

• Should non-citizens get to vote?

• Should citizens abroad get to vote?

• Or some even crazier ideas that have been brought up:

  • Should there be an age cap for voting?

  • Should people who don’t pay taxes get to vote?

  • Should people without kids get to vote?

To tackle these questions, I think you should have a good model of how democracy works (or is supposed to work, at least). As I wrote in this piece on politics as rational decision making, people’s political motivations consist of three things:

• their political views, the policy they would push for to make society better

• their personal interest, the policy that would benefit them personally

• norms, such as the norm of punishing scandal-ridden politicians. (This is not super relevant for this piece).

I think personal interest would dictate that people of any age, inmates, non-citizens, people who don’t pay taxes, and people without kids should get to vote. All these groups are just as human as anyone else. When people in the above groups don’t get a vote, politicians have the incentive to create policy that sacrifices their welfare in favor of those that can vote. The only group listed above whose case is not particularly strengthened by personal interest is citizens living abroad, who are much less affected by national policy, although they do still often have to pay taxes and such.

Political views are where it gets dicey. I think there’s a legitimate case to be made that many of these groups are likely to have worse political opinions than other voters, because they are less informed, not as mentally fit, or just plain the type of people you might think to have bad political views (eg inmates). For the case of children, you might also think they’re so heavily reliant on others that they are likely to essentially function as another vote for their caretaker’s political views.

But as I discussed in the “politics as rational decision-making” post, you have to ask yourself: these groups’ political views are worse… according to whom? If you were an inmate, you wouldn’t feel that inmates’ political views are worse. White southerners in the Jim Crow era probably felt that black people’s political views were worse, and they probably had some arguments to back that up. I think core to the idea of democracy is that everyone’s voice counts equally. Once you stray too far from that you can get seriously horrific results - Jim Crow being a good example.

Recommendations

My ideal is probably that everyone living in the US, 16 or older, gets to vote. So this includes inmates, old people, non-citizens, people who don’t pay taxes, etc. Everyone there has legitimate claims to being affected by policy and are independent enough thinkers for their vote to be legitimate. Another advantage of making the voting age 16 is that young people often move out of their parents’ house at 18, which makes registering to vote trickier.

I would probably lean toward citizens abroad being able to vote too - they are still affected by national policy quite a bit and they are still people who can form informed political opinions. That being said, I could see the argument for them not getting a vote.

More generally, I think the danger of too few people voting (those non-voters being treated poorly by the government) is greater than the danger of too many (legitimate votes being diluted a little bit). Just let people vote.

Mandatory Voting

Here’s a different question: should there be a fine for not voting?

The considerations here are very similar to those surrounding voting rights. Personal interest dictates yes, otherwise politicians will prioritize voters over non-voters when making policy. There’s some case to be made that people do not vote when there is less personally at stake for them in the election, but I think this correlation is fairly weak.

Political views probably dictate no - you’d expect people who don’t vote to be less informed and have worse political views than people who vote.

Overall, I could go either way. Maybe it’s better to give would-be non-voters more political power so they are treated better, maybe it’s better to leave voting to more politically informed citizens, I’m not sure. The one thing I will say is that the effects would not necessarily be what you think - although non-voters in the US are disproportionately low income and nonwhite, they probably lean right relative to voters these days:

Secret Ballot

This one is simple. Who people voted for should be secret. People shouldn’t be able to be pressured to vote for a certain candidate.

If you voted is actually public in the US. That’s bad - it should be private as well. You don’t want people to be able to be bribed to vote.

Voting Methods

I didn’t initially include this part because it’s a very logistical thing and not a great fit for this more theoretical series. But I have some interesting things to say, so let’s do it.

Let me go through the different decisions there are and address them one by one.

• Should you have to register to vote? How should that work?

Let’s back up for a second. Why do you even need voter registration? Why can’t the government just keep a list of all people in the country at all times?

The main reason is that it matters where you vote, so you should be registered to vote in a certain place. I still don’t think it’s mandatory - I think it’s pretty unlikely that everyone would drive to a swing district to vote instead of voting where they live - but it’s a legitimate reason.

Still, you can just automatically register people in the place they have listed as their last address or whatever and then let people manually change locations when they move. This is what’s done in a bunch of places (Israel, Spain, Italy).

Maybe there is some other logistical reason for making voter registration manual, but I don’t see it.

• Should you have to bring an ID to vote?

The consideration here is that voting with an ID makes voting more secure (you don’t want people claiming to be people they’re not), but can leave people out who don’t have government IDs.

The obvious solution is to mandate government IDs to vote but then make them very easy to obtain and ~universal.

• Should you be able to vote by mail? If so, when should mail-in ballots count - if they arrive before election day, if they’re sent by election day, or sometime else?

Again the tradeoff with mail voting is security vs. inclusion. Some people might not be able to vote or want to vote in person. But mail in ballots are quite insecure - it’s very easy just to sign as someone else and vote on their behalf.

I’m honestly unsure. Voter fraud seems to be extremely rare in practice, but mail fraud seems like such a gaping flaw - you can just vote on behalf of anyone you know that doesn’t vote. You can notify a person when they vote so they can dispute if it wasn’t them, but still.

(Maybe one day we will be able to sign documents biometrically to make things secure. Ok I know the privacy people are yelling at me).

If you do choose to allow mail voting, you should probably require ballots come in by election day. For some reason, whenever elections take too long to count, people get outraged and suspicious. I guess the logic is supposed to be that those in power could decide to rig an election after the votes begin to be counted and they realize they’re going to lose? But why would they not just rig it from the start to be safe and so that there’s no suspicion raised from early counted ballots?

Honestly I’ve never really understood this logic, but given that people feel this way, you should count ballots by election night, and that means no mail-in ballots that come in past election day. You just have to make sure those in power don’t intentionally sabotage the mail service (as Trump’s pick DeJoy was alleged to have done in 2020, though I’m skeptical) in order to nullify votes.

• Should you be able to vote early? How early?

Early voting is fine. Sure, it could happen that there’s some big news that drastically changes people’s vote sometime before election day, such as a change in candidates, but this is rare. People should be able to vote a month+ in advance.

You can also allow people to change their votes up to election day, eg to handle the case of a candidate switch. This does raise the security risk that someone could pretend to be another person and change their vote, so it’s probably not worth it if you don’t require ID to vote, but might be worth it with IDs.

Term Lengths

When should you hold elections?

Too frequently and you don’t give those in power time to do things. Big policy takes a certain amount of time to set up - negotiation, implementation, refinement - so if you have elections too frequently you aren’t going to allow big policy to be implemented properly. This is especially problematic because the public is going to judge you based on half-baked policy.

Another drawback of holding elections too frequently is that you won’t be able to see the results of new policy, which means that voters won’t be able to tell if it’s working. In 2010 we had barely yet seen the effects of Obama’s economic stimulus, leading to a red wave, while by 2012 voters had warmed up to the Obama presidency.

On the flip side, the less frequently you host elections, the less democratic your system is - the less the views of those in power will reflect the views of the public.

4 year terms is probably my preference. I think it’s long enough to get stuff done, but short enough to lead to pretty democratic results. Also, do not hold midterm elections - they screw things up in the same way that short terms do, and divided government is a bad state to be in in general.

Rolling Elections

Here’s a fun idea: what if instead of the entire legislature being elected once every 4 years, you have 25% of the legislature elected every year.

A bunch of things to say about it. First, the pro: these gradual elections makes the legislature more responsive to current positions. And unlike frequent total elections, they don’t penalize politicians as much for pursuing actions that make them initially unpopular but later make them popular. And it can help prevent the opposite problem; it’s pretty common for parties to act more radically early in their term only to moderate by the time the election rolls around, which is ultimately just kind of undemocratic.

The main con I can see is if a country is closely divided and frequent elections lead to power changing back and forth between the left and the right. This is particularly a problem if you have a two-party electoral system, although it could still apply in a proportional system, either one that’s very divided (like Israel with its pro-Netanyahu and anti-Netanyahu camps) or between a left/right-wing and a centrist government. This frequent alternation doesn’t give either side enough time to implement stable policy, which hurts accountability democracy and generally good governance.

If you do choose to implement rolling elections, you could have each 25% be a representative sample or you could split the country into 4 big contiguous regions - like northeast, southwest, southeast, and northwest.

Having the 25% be representative is somewhat challenging. You can of course just totally randomly do it like based on your Social Security number, but that seems hard to implement and also makes it likely that people won’t know if they’re scheduled to vote. The better way to do this is probably to randomly divide the country in 4, maybe by district, perhaps after each census.

The advantage to making it totally representative is that there will be no geographic skews in the election. You could imagine, for instance, a world like the 2014-2020 period in US politics, where the Midwest is trending rightward and the southwest is trending leftward. If the Midwest and Southwest take turn holding elections, you could have power alternating back and forth between the left and right.

The advantage of splitting it into contiguous regions is that it will do a better job of making the election very salient in the region that is currently holding it. I think this salience is a key to how elections work in the real world: people are exposed to lots of campaigning that allows them to make the best decision for themselves. It also makes it more likely that people will know about elections and actually vote when they’re supposed to.

I actually do really like this rolling elections idea and would probably implement it in my ideal system. My preferred way things is probably to split the country into regions where there is likely a lot of information transmitted between the different people in it, and so this salience thing is more impactful. US media markets might be a good model:

One last thing: one other downside of rolling elections is that they make it harder to tell when a system is straight up undemocratic, like the US senate, because you can’t easily say “the left got x% of the vote with y% vote share”. For this reason I might not be in favor of rolling elections unless you’re really sure that your electoral system is good.

Term Limits

The rationale for term limits really depend on who you’re applying them to. The downsides on the other hand are pretty universal:

• Once you stumble upon a good popular politician, you should be able to keep them around, instead of having to find a new one

• A politician develops experience upon serving

• A politician not eligible for re-election doesn’t have the incentive to act in popular ways

For Leaders

Term limits for the leader are there to ensure the leader doesn’t become a dictator. If a leader serves too many terms, they can consolidate power by ousting any politician who doesn’t show them complete loyalty. This then allows them to destroy democracy.

The usual downsides of term limits all apply to the leader. The one thing I’ll note in particular is that while a term-limited leader doesn’t have the same incentive to get themselves re-elected, they are still incentivized to make their party more popular so that its next leader (who likely shares most of their political views) can win election. Still, it’s not quite the same, and the next leader can always distance themselves from the previous leader if they are too unpopular.

The one clear mechanism I think you should implement is non-consecutive term limits. If a leader loses re-election, you no longer have to give them the incentive to act in popular ways. In fact, giving them the ability to run for re-election might give them less incentive to act popularly as leader, because they know if they lose they can always run again. Sure, non-consecutive term limits mean that chosen leaders will have less experience, but I think the threat of a dictator makes them worth it. This one mechanism would do a lot to stop would-be dictators - stopping Trump, Netanyahu, Orban.

Consecutive term limits are trickier. A limit of 1 or 2, which is likely to actually prevent democratic decline, leads to constant term-limited leaders - something like a third of US presidencies have been term limited. I’d probably go with 2, but I can see the argument for 1, 3+, or no consecutive term limits.

(I don’t support term limits for leaders on the local level though, since the risk of democratic decline is much lower)

For Party Leaders

If you don’t want a leader to have dictatorial control over their own party, you can make the party leader spot term limited as well. In a parliamentary democracy, this party leader is well defined, it’s the one vying for the prime minister spot. In presidential democracy, the party leader probably encompasses both the party’s presidential candidates and their legislative leadership; you could have one timer that counts toward the number of times a person can serve as either a presidential candidate or as a legislative leader.

In formalizing the party leader position(s), you want to try to prevent there from being some sort of “shadow leader” figure in a party, who dominates the party but only takes the leader spot when they think they have a decent chance of winning.. This is what Nigel Farage did with his Reform party, for instance.1

Probably you’d want the party leader position to encompass anyone who:

• is a party’s candidate for president (in presidential democracy)

• is prime minister

  • You can try to do a thing where the only people eligible to be prime minister in the eg year following an election are the party leaders declared during an election. This unfortunately has the downside of making it ruling out some potential compromise prime ministers that could unite an ideologically diverse majority coalition, but this might be worth it.

• has predominant say in a party’s candidate selection

• has predominant say in the legislative leadership of a party, eg the committee chairs

These are the powers that I think would allow someone to dominate a party in a way that would merit term limits.

I don’t think you should make term limits for party leaders apply to small parties - parties with less than eg 15% of the seats or 15% of the vote last presidential election. With very small parties you may just not have a ton of legislators to choose from for legislative party leader, and it isn’t so important that they don’t dominate their own party.

For larger parties though, I think it’s worth considering. It’s not without the usual downsides of term limits, but it has the important upside of ensuring that a leader doesn’t dominate their party. Maybe I’d make it a term limit of 3 terms as party leader (either in the legislature or as a party’s presidential candidate), and if they ever become president/prime minister the regular leader term limit kicks in instead.

For Legislators

I think these are generally a bad idea.

These have the all the same downsides of term limits. I think the inability to develop experience when serving is particularly bad for legislators; a leader can always rely on those around them for support, while legislators are all you have.

The upside is supposed to be that long-time serving legislators are often corrupt. I don’t think this logic is super strong though - sure long-time legislators might have closer connections to lobbyists, but newly elected officials have just the same incentive to take bribes. People can also just vote out longtime leaders if they feel they are corrupt.

And actually, because term limits for legislators cause them to have less experience, they can increase the power of lobbyists, because the lobbyists are the only ones who have really been around long enough to know the ins and outs of policy.

All that said, you might still want term limits for legislators because you think incumbency in party candidate selection is just too large. Perhaps new legislators are more likely to try to act in popular ways where long-time legislators become complacent. I would set the limit to be high to minimize the downside, maybe four or five terms of 4 years each.

An important part of democracy is that people know what their politicians are doing. Still, you probably don’t want the public to know literally everything that happens in government. This piece is about finding that balance.

Key Distinctions

In his paper Output Transparency vs. Input Transparency, Cass Sunstein draws a distinction between outputs and inputs in government. He presents two main examples of outputs: law - regulations, statutory law, etc. - and facts produced by the government - studies. Then with inputs, he talks mainly about discussion among different members of the government that goes into producing outputs. He also discusses conclusions drawn from facts (which go into the writing of law) as sort of an in between category.

I don’t love the input/output binary and think you should think of it as an axis from facts to discussion to law.

But actually, I don’t like the names here. Instead of law, I want to call it actions. This is what the government is doing. If you take my definition of law as any direction given to a government employee, law should be a complete account of what the government is doing, but I think actions captures more of what I really care about.

Instead of input, I’d like to use the word formulation - it is the record of how the actions came to be.

Facts is fine - that term can stay.

Now the axis makes a bit more sense: politicians look at facts, formulate actions to respond to them, and then take those actions. Some examples of things on this axis:

Action transparency underlies democracy

It is pretty easy to see why government actions should be transparent. Democracy requires that voters know what the politicians in power are doing so that they can re-elect them or not.

Maybe you could argue this is less important for lower salience issues that people don’t vote on anyway, or where we’d expect politicians to have better views than the public, but as I discussed in this piece I think it’s hard to make a principled arguments about in what cases you should shield politicians from public opinion.

Maybe you could also argue against transparency on the basis of accountability democracy - that voters should assess the job of those in power based on their results, not their actions. I think that this is quite an extreme and hard-to-defend position for high level actions (statutory laws, executive orders, etc.), although it might make some sense for lower-level actions.

You might also question the utility of transparency for non-partisan entities like the courts, who are not subject to election and supposed to be shielded from public opinion. Still I think these entities are supposed to be subject to some sort of pressure - maybe the pressure of their professional peers - to ensure they are actually doing their job in a principled manner. This requires knowing what they are doing.

So unless you have some more specific reason about why a certain action should be private, you should default to making it transparent.

Formulation transparency is about motivation

Formulation is more complicated. The main thing that formulation reveals is motivation. As I wrote in this post, I think politicians’ motivation comes down to three things:

• their political views, ie what actions they would take to further society

• their personal interest, their desire to enrich themselves (corruption) or stay in office (potentially by destroying democracy)

• norms they feel they have to adhere to

Formulation can reveal all three in important ways.

On political views, I think it’s important to know if Trump is enacting tariffs because he likes tariffs or if he’s enacting tariffs because he wants leverage to make trade deals with different countries. These revealed political views serve as a good indicator of what types of actions the politician is likely to do in the future - in this case, whether Trump is likely to pursue trade deals. They also can help reveal when the government is half-assing something. It’s often hard to tell how thoroughly the government is following through with claimed actions, and revealed motivation helps a lot here.

On personal interest, I think it’s important to know if a regulation was slashed because the minister’s nephew works at an affected company or as a result of principled cost-benefit analysis. Or if a change to an electoral system was made after carefully considered debate or because politicians want to stay in office.

On norms, I think it’s important to know if a certain ruling was made based on serious analysis of the law or just willy nilly to adhere to the judge’s political views.

But sometimes motivation is overrated

In his piece Against Transparency, Matthew Yglesias compiled a good list of reasons to prefer the formulation of government action stay private.

I’d like to bring up three in particular:

• Policy decisions need to be made with political sustainability in mind, but part of making a politically sustainable policy decision is you don’t come out and say you made the decision with politics in mind.

• If a previously embarked-upon course of action isn’t working, you may want to quietly change course rather than publicly admit failure.

• Healthy brainstorming processes often involve tossing out bad or half-baked ideas in order to stimulate thought and elevate better ones.

The main point underlying these three reasons is that sometimes the public makes too much of the motivation behind certain actions - what ultimately matters at the end of the day are the government’s actions. The politicians in power might be doing something for political gain, and that can be ok. Politicians might throw out bad ideas during brainstorming that offend people, but as long as they don’t end up pursuing these ideas, it’s fine.

So it’s a tricky balance. You should aim to make formulation transparent if you think it’s really important for understanding current or future actions of politicians, and not just going to be used as a gotcha if the intentions of politicians aren’t totally pure.

Facts are important for assessing results

Sunstein brings up two main reasons for making facts transparent.

The first is that facts can just be useful to the public. He brings up the example of the GPS, which is a tool created by the government that has served very useful in allowing people to navigate around.

The second, more relevant to constitutional design, is that facts allow the public to assess the state of the country. For various reasons, the government is often the entity in society that collects the most information about how things are going. It collects mortality statistics, crime statistics, economic statistics, etc. These facts are very useful to the public in being able to assess how well those in power are running the country.

Again I think it’s hard to come up with principled reasons for when shielding politicians from consequences by hiding facts from the public would be a good idea. Maybe you think certain facts are overrated by the public, but I think democracy dictates that the public ought to decide for themselves. So as with actions unless you have specific reasons to keep certain facts private, make them public.

Specific Considerations

Let me go some more of the specific considerations for making information private.

National security concerns

Countries have enemies, which means it’s important to keep certain information safe from them. This can apply to facts (eg conditions on the ground), formulation (eg debate over military action), or actions (eg military strikes).

One problem might be that you need someone to declare information classified, and if it’s the people in power they might have reason to hide information by calling it classified. So you need norms or legal rules on what to call classified.

I would allow the legislature, both the majority and the opposition, to see some classified information though so that they can more thoroughly scrutinize the executive. This is a difficult balance - you probably can’t let them see everything, if you have 500 legislators it can be easy for things to get out. Maybe you could limit certain classified information to very senior or high-ranking legislators.

Maintaining good relations

If national security concerns are about enemies, maintaining good relations is about allies. This applies both to other countries and to different actors within a country, even within a government, that those in power want to retain good relationships with. As Yglesias puts it:

• A realistic survey of options may require a blunt assessment of the strengths and weaknesses of different members of the team or of outside groups that would be insulting if publicized.

This mostly applies to formulation.

Panic

Yglesias writes:

• Someone may want to describe an actual or potential problem in vivid terms to spur action, without wanting to provoke public panic or hysteria through public discussion.

I’m skeptical. I can imagine an argument that panic is bad, but I’m not exactly sure what qualifies as panic and when it’s actually a problem. Did the public panic about COVID in March 2020? If so, were they wrong to?

I really don’t think panic is a good reason to keep things private. I think it’s more an excuse than anything else.

High-Level Law

I want to take some time to talk specifically about high level law - things like statutory/constitutional law, court rulings, executive orders, and regulations; probably what you usually think of when you think of law. I want to discuss whether it’s important to make the laws themselves - rather than general descriptions of them - public.

Making law public maximizes accountability

This is pretty simple. Sure, you could just make a general description of the law public. But to maximize accountability, you can’t beat just making the law public.

Following the law requires knowing what it is

Another basic reason for transparency in law is that people should know what laws they have to follow. If the government makes smoking illegal, people should know that so that they know not to smoke.

I will say that this rationale is weaker than it might seem. Often people don’t know most of the laws they have to follow - they don’t know what the fine is for speeding, or what exactly the laws regarding alcohol are. People make reasonable inferences about what the law likely is, and how likely it is to be enforced, and live their life based on that. It wouldn’t change that much if nobody knew what the fine is for speeding until they actually had to pay it.

The people who really benefit from transparency are lawyers, and then anyone who relies on them. If people want to know whether something is legal or not before doing it, they can consult a lawyer, who knows what the law is and can advise them. I think this case is especially strong for businesses, who try really hard to optimize for not paying fines.

I suppose it could be possible to create a system where only lawyers are allowed to know what the law is, like getting a law license gives you access to the rulebook. But things might get thorny if the government tries to restrict access, and anyway I don’t know what the point of that would be.

A government of law, not of men

Often, though, law constrains what the government is allowed to do, not what you’re allowed to do. In that case, the main purpose of making law public is ensuring that the government actually follows the law.

Again there’s the question of how much transparency actually matters. How exactly does making law transparent ensure that it is actually followed?

I think there are a few key actors involved here. One is the courts - if the government breaks the law, it is the courts’ job to say so and get them back on track. There is a strong norm that the government must abide by court rulings, and if not the court has some mechanisms to force their hand.

But how do you ensure the courts are following the law? Again you rely heavily on norms. When they are broken, I think this is where some other actors come in. As before, lawyers are the ones who know what the law is and inform others. Then it is the role of the media to highlight when the courts or the government are breaking the law and disseminate that information. Then, hopefully public pressure is enough to get the courts or the government to back down. The idealized mechanized here is that the public will vote out politicians that break the law, although I think the actual mechanisms in place are more like “some people in government, who aren’t that dissimilar from the public at large, are outraged and use their leverage to force legal compliance”.

Legislative votes

Lastly, I want to discuss one of the most interesting parts of transparency, legislative votes.

When the legislature votes on a bill, the public gets a quick snapshot of every legislator’s political position on an issue. Legislators can always downplay the importance of their vote by claiming that the bill’s wording is biased, or that they were just voting against the bill because they wanted certain specific changes, but the vote itself is still undeniable. I’m not sure how much people actually learn about these votes, but it be might more than you think (the votes could come up in ads, journalists could use them to characterize candidates, etc.)

Symbolism

Sometimes these votes are symbolic, when they are on a measure that’s destined to fail because the majority bloc is voting against it and is just using it as a tool to expose the minority as holding an unpopular position.

These votes provide less information for the public than substantive votes because legislators can just vote dishonestly with no substantive consequences, but still they can be useful. A legislator might have moral qualms with voting against a symbolic resolution even when it’s popular, or they might be torn between different political interests that cause different parts of the electorate to get mad at them no matter what they vote.

This introduces an asymmetry: because the majority bloc in legislature (usually) decides what comes to a vote, they can expose their opposition in this way, while the minority bloc cannot expose the majority’s unpopular positions. Even if you like the incumbent party having a leg up in public opinion, I don’t really think this is the type of advantage it makes sense to give them. So to address it, you can allow the minority bloc to force votes by requiring that any resolution that gets the signature of more than 1/3 of the legislature gets voted on. I would limit these votes to symbolic resolutions, to match the symbolic nature of the votes that the majority party forces. I also do think there are legitimate accountability reasons to allow the leader of the majority bloc to control what substantive legislation gets voted on.

(The only thing with this mechanism is that you just have to limit the number of such symbolic votes to prevent this process from being used to jam up the legislature’s time.)

Status Quo Bias

Another asymmetry in votes is that the legislature only has to vote on what to change and not on what to maintain. This allows politicians to never have to bring up an unpopular position for a vote as long as it is the status quo. This is especially impactful when there is bipartisan consensus, so an issue never comes to a vote no matter who is in power.

This is harmful if you don’t think politicians should be able to freeze out the public view like this. The solution is to force a piece of legislation to be voted on if a public petition with it gets a certain numbers of signatures. See my post on direct democracy for more discussion of this issue and this proposal.

Secret Ballot

What about the other direction: what if you want to insulate politicians from the public view by making each legislator’s vote secret. Are there any instances where this would make sense?

Most laws are going to be supported by the majority bloc in the legislature, and the party will attribute them to the majority bloc, so it doesn’t really matter if these votes are secret.

Secret ballot becomes important when some members of the majority bloc break off and vote with the minority. This can happen either with bills that pass or with bills that fail. I don’t think secret ballot is a good idea with normal majority-vote bills - it is democracy after all, people should know the views of their legislators.

The one version of secret ballot I think is plausible is as follows. Legislators can sign on to secret petitions that the public cannot see. If any such secret petition gets the signature of a supermajority of the legislature, it goes to either to a vote by the entire legislature or to a public referendum. If it is then approved, it goes into law, over any veto of the leader. (And perhaps you could have this process for constitutional amendments if a supermajority of the public approves of it).

The main thing this mechanism targets is a case where a lot of members of the majority-bloc in the legislature are afraid of voting for a bill because they don’t want punishment by their own party - for example, if they are afraid to be primaries because they are led by a leader with a cult-like following like Trump. So they can instead sign on to these secret petitions to pass legislation. This is a bit logistically difficult to implement, because you have to hide who signs on to the referendum from party leadership themselves so they cannot punish members who sign on, but it might be possible.

This mechanism hurts accountability - part of democracy is being able to control what laws get passed when you’re in power so you can be judged for them - but that might be worth it.

I think if you’re looking for a simple rule, public actions, public facts, and private formulation is a good one to aim for. But if you’re willing to think more carefully, I think you can do better than that.

But before any of that, some social choice theory background is in order. What are we even looking for in an electoral system?

Social Choice Theory Background

Maximizing Societal Utility

If you want a prescription for who objectively “should” win an election given certain voter preferences, there’s basically two you could choose. The first is to define each voter as giving some utility to each candidate, and then choose the candidate that maximizes total societal utility.

There is one ethical philosophy, preference utilitarianism, which views this as close to the optimal choice1, assuming you view these utility functions as people’s true preferences. But if you don’t hold this stance, either because you’re not a philosophy nerd and just want something vaguely democratic, or because you have different ethical beliefs or ways of viewing democracy (like me2), I don’t think this method comes out looking all too amazing.

This method says that someone who hates one candidate and loves another should have more say in an election than someone whose preferences are milder. This might be desirable because not doing so can lead you to tyranny of the majority, where 51% of the population hogs all the resources and is happy while the other 49% is miserable. But on the other hand, I think people can get really passionate about politics in ways that don’t merit more say, like being terrified of immigrants that don’t actually hurt them. And for electoral systems specifically, there are other methods in society for people with stronger preferences to have more of an effect (party candidate selection being the main one, but also activism, interest groups, higher turnout in elections, etc.), so it’s maybe not the right mindset here.

So overall I’m kind of ambivalent on this prescription.

The Condorcet Criterion

The other prescription for who should win an election is to choose the Condorcet winner (called the “Condorcet Criterion” for a voting system). The Condorcet winner is a candidate that would defeat every other candidate in a head-to-head election. If a population is 40% left-wing/20% centrist/40% right-wing, the Condorcet winner is the centrist, assuming left and right-wing voters prefer the centrist to the opposite party. In this way, the Condorcet criterion is a way of encoding a preference for moderate candidates.

I will also point out that the Condorcet criterion doesn’t go maximally far in the direction of moderation. You might instead imagine electing the candidate that the fewest people really hate; that would be too far in my opinion. In the real world, the ideology model doesn’t hold that literally, so you can easily see a left- or right-wing candidate win in a 1-on-1 election with a centrist.

There isn’t always a Condorcet winner. Instead, you can have a Condorcet cycle, where A would beat B in a 1-on-1, B would beat C, and C would beat A—for example in an election where 1/3 of people prefer A>B>C, 1/3 prefer C>A>B, and 1/3 prefer B>C>A. I’m honestly not sure how often we should expect Condorcet cycles in practice. I don’t think it’s some crazy hypothetical; it’s quite imaginable for left-wing voters to prefer left-wing candidate (L) > traditional conservative “center-right” candidate (CR) > right-populist “far-right” candidate (FR), center-right voters to prefer CR>FR>L, and working-class voters to prefer FR>L>CR. Then again I would be surprised if it happened that often. Maybe 1 in 10 real-world elections feature Condorcet cycles?

I quite like the Condorcet Criterion. But you don’t necessarily have to share my preference; I’ll try to go through the most notable electoral systems and discuss the pros and cons.

Single-Member Electoral Systems

Methods for Polarized Elections

This is the bucket of electoral systems you should choose if you like having a two-party system with politicians toward the extremes. Note that since politicians strategically moderate to win elections, none of these systems do as poorly at electing moderates as you might naively think, but still the politicians elected are usually well to the left or right of center.

FPTP

First-past-the-post (FPTP) is the simplest method for electing a single person: everyone votes for one person, and the person with the most votes gets elected. This system leads to the spoiler effect, where people voting for third parties are essentially throwing their vote away, and thus everyone is incentivized to vote for the two best-polling or most established parties.

Top-Two Runoff

You can expand upon FPTP by having a top-two runoff, where the two best-placing candidates in a first round of voting face each other in a one-on-one second round. This system allows people to vote more honestly during the first round instead of strategically coalescing around the two best polling candidates. In practice, you’re usually just going to get the same left vs. right election as you would in a regular FPTP election. There are three main differences from FPTP:

• People are less likely to throw their vote away (abstaining, voting third party) in a runoff than under FPTP. This is because:

  • (a) they don’t have to rely on polling to know who the two front-runners are, and

  • (b) they feel like they have had the opportunity to vote honestly in the first round and so are happier to vote strategically in the second round

• Sometimes, a runoff will allow a candidate not from one of the two main parties to gain enough traction to make the top-two when they wouldn’t have under FPTP.

• You can use this system as a replacement for primaries. I don’t recommend this though if you can avoid it, because of the case when two candidates from the same party make the runoff.

  • When it’s two candidates from the less popular side, eg, this is just a failure

  • When it’s two candidates from the more popular side, the more moderate candidate should win. I also see this as a failure - if you want a moderate candidate to win you should pick a different system, the fact that this system randomly picks a non-plurality moderate candidate 1/10 times is unintended

RCV

In Ranked-Choice Voting (RCV), also known as Instant-Runoff voting or Alternative Vote, people rank their choices. At first all votes are allocated to people’s first choice; then, in successive rounds, the lowest-scoring candidate is eliminated, and each voter who had chosen this candidate has their vote reallocated to their next highest-ranked candidate. This is done until there is only one candidate remaining.

Much of the stuff I said about top-2 applied to RCV as well:

• People may be less likely not to vote for (ie rank) either of the top 2 candidates than under FPTP because they don’t have to rely on polling and are satisfied with their ability to vote for their top choice first

• Sometimes, RCV will allow a candidate not from one of the two main parties to gain enough traction to be one of the final 2 candidates when they wouldn’t have under FPTP.

• You can use this system as a replacement for primaries

The advantage of RCV over top-2 is that it can be more robust to having many candidates, eg it’s less likely to see two candidates from the less popular side make the runoff.

The main downside is “ballot exhaustion”: you never truly get 100% of voters to fill out their entire ballot. In practice, the main failure mode is that some people who vote eg Republican and generally prefer Republicans to Democrats don’t bother filling out any other Republicans besides their top choice. I’d also extend ballot exhaustion to include any time a voter’s full ranking isn’t as accurate as it would have been if they really thought long and hard about it, even if they did fill out most of their ballot. One reason for ballot exhaustion might be limited knowledge about all the candidates just because of how many there are. Another is just that some people are lazy and just mark their top choice.

To remedy this, you can have two rounds of voting, where the first round is RCV and the second round is a manual runoff for the top two contenders. Alternatively, you can do something like Alaska is doing, where in the first round everyone votes for just one candidate, and then the top four placers make it to the second round which uses RCV.

Two-round RCV (of some form) is just about strictly better than FPTP. The main downside is really that it’s more complex. I think I’d consider two-round RCV to be the best system if you want polarized elections, but I think you can make the case for any of these systems (FPTP, FPTP+runoff, RCV, two-round RCV) and it wouldn’t be crazy.

One last thing. I do think RCV can be more likely to select the Condorcet winner just based off vibes. As in: under RCV people are more likely to strategically upvote the Condorcet winner, and Condorcet winners are more likely to run, just because it seems like a thing the system is supposed to encourage (even when it doesn’t actually).

Condorcet Ranked (Ordinal) Methods

The second bucket of electoral systems is ranked methods that explicitly pick out the Condorcet winner with an algorithm, which is easy to do once you have everyone’s full ranking.

As with with RCV, ballot exhaustion can be an issue. The right way to deal with this is to do a top-4 system and mandate that people rank all 4 candidates in the second round. I do worry though that a Condorcet winner would not even be able to make the second round, like in a very polarized landscape you might need a decent amount of strategic voting to get them there. But there’s not really any way around this.

These ranked Condorcet methods differ in how they handle a situation where there is no Condorcet winner, their cycle-breaking method. Once you narrow it down to 4 candidates though, I find it unlikely that these systems would differ in practical cases of no Condorcet winner. The simplest and most intuitive approach is probably Minimax.

Condorcet ranked methods also tend to have minimal strategic voting. There’s a formal statement here:

Strategic voting in ranked Condorcet systems can only occur by manipulating the configuration so that there’s a cycle, and then exploiting the cycle-breaking method. In most sensible systems though, including Minimax, this is really tricky and I’d expect it not to happen at all in real elections. The one possibility I can see is that voters think they can manipulate the system and end up doing stupid things that lead to a worse result. Probably I just wouldn’t publicize the cycle-breaking procedure too widely and emphasize that it’s rare.

On complexity

The most cited drawback of these ranked Condorcet methods is that they are complicated. Honestly though, I think that’s overstated. Condorcet methods can be easily understood with a single chart: the head-to-head matchups. The winning candidate is just the candidate whose row is all green.

I think most of the public is intelligent enough to understand this chart. Some of the dumbest people in the world obsess over sports numbers, I feel like this is easier to understand than a lot of those. Understanding how the chart is produced - counting how many people rank candidate A above candidate B, vs. vice versa - is more complicated, but still not insane and not critical for understanding the method. And crucially, none of this is necessary for understanding how to use the method as a voter: the optimal strategy is the genuine one, to rank candidates in order of your preference.

I’ll have more to say about complexity at the end, but I do think these methods are too often dismissed.

A Brief Note on Strategic Voting

These ranked Condorcet methods are the only ones that guarantee a Condorcet winner wins. An interesting result, however, is that almost all electoral systems should select the Condorcet winner (assuming one exists) if everyone is super strategic and public polling is good enough. This is easiest to see with an example from FPTP: if it’s a left vs. center vs. right and polling shows that the left would lose to the right in a 1-on-1, left-wing voters should all vote for the centrist so that the right-winger doesn’t take power. Here is a proof for score voting, a system I’ll talk about shortly.

In practice, people aren’t strategic enough and public polling isn’t good enough for this to hold. But it’s an interesting result, and can hold to some extent in some voting systems.

Rated (Cardinal) Methods

The last bucket of voting systems is rated methods, where people rate every candidate on the ballot, eg from 1 to 10.

Score and Approval Voting

Score voting is the simplest: have everyone rate the candidates, sum the ratings for each candidate, and then seat the candidate with the highest total rating. You can have a rating scale like 0 to 10, or you can have a binary approve/disapprove, which is called Approval Voting.

This system needs fewer strategic voters to elect the Condorcet winner by the above logic, but again this relies on public polling being good enough. I feel like in real life this system might elect the Condorcet winner it’s a standard left v. right election and polls show a landslide (because the losing side would strategically vote for the centrist). If polling shows a tight race, or if things are more complicated than left v. right, I feel like it would be pretty unpredictable. The hope is that people would be more honest in that case, since it’s not clear what the right strategy is, and thus you would get something close to the candidate satisfying total utility3. And idk, maybe. Or maybe they’d behave irrationally in some other way, like giving their candidate the max rating and every other candidate the min rating because they’re lazy.

There’s one other failure mode of score voting that upon reflection seems quite bad. Let’s say you do score voting with a top-4 system to help with consolidation, and the public favors the left wing candidates 60:40 over the right wing candidates. Then imagine the runoff is two left-wing candidates polling relatively similarly, one right-wing candidate, and one irrelevant candidate that everyone knows is not going to win; this all seems like a very normal situation. But now, the voters for left-wing candidate A and left-wing candidate B are caught in a sort of prisoner’s dilemma with one another. If all candidate A voters “defect” by rating A 10/10 and B 0/10, while candidate B voters “cooperate” by rating B 10/10 and A 8/10 (and the other two candidates 0/10), then candidate A will win. But if they both “defect”, rating the other candidate 0/10, then all of a sudden the right-wing candidate will likely win. Now in the real world the numbers probably wouldn’t be quite as drastic, but still this seems like a very real dynamic.

One upshot of this dynamic is that score voting probably can’t replace party candidate selection (through primaries or some other means), so you shouldn’t have a top-4 system. But even with party candidate selection, I do worry that this would happen anyway, and it’s made me more reluctant about score voting.

STAR Voting

STAR is very similar, except instead of seating the candidate with the highest total score, it uses the total scores to narrow the field down to 2 candidates, and then seats the candidate which would win in a runoff (ie, the candidate which more voters rated higher than the other one). STAR has the (IMO) advantage of making it more likely the Condorcet winner wins, because they would win that final round. STAR stands for “Score Then Automatic Runoff”, but you should probably actually make it a manual runoff to help with public knowledge.

The biggest downside of STAR I can see is that it’s not cloneproof: if the winning candidate has a “clone” which has a very similar voting base, these two candidates will make the runoff and it will basically be score voting. Clones seem like a pretty gaping flaw if parties can nominate multiple people. You could try to mitigate it with some sort of “party only nominates one” rule, or with a threshold system where candidates need to have (eg) 10% of voters rate them highest in order to make the runoff, but even with them STAR seems liable to collapse into Score voting.

Highest Median

This is what the name suggests: choose the candidate with the highest median rating, with some secondary system for breaking ties (read about that here). This voting system lends itself to pretty pictures, which makes it fairly easy to explain (except for the tiebreaking procedure) even to someone who doesn’t know what a median is. Highest median is supposed to be more strategy resistant than score voting, but honestly it seems even more uncertain how it would work in real life. Like, in a standard left vs. center vs. right election, highest median should turn on if left/right-wing rank the centrist candidate higher than centrist voters rank the left or right-wing candidate, which seems highly uncertain.

Recommendations

The two best systems in my view are score voting and ranked condorcet methods. If you have free choice over which electoral system to adopt, I think there are solid arguments for both of them; for reasons I’ve discussed above I’d probably lean toward ranked Condorcet ranked methods just to get rid of all strategic voting when there’s a Condorcet winner.

I don’t usually go too far into the practical aspect of what you should actually advocate for in the real world, but let me do a bit of that here. In practice, adopting a new electoral system is such a tall order that people are desperate for methods simple enough to explain to people. Once an electoral system is adopted, however, governments are quite good at following them to the letter. There have been some cases of mistabulations that have gotten resolved later on (like this case out of California), but in general I’d expect these cases to get much less frequent as an electoral system is widely adopted and used repeatedly.

My recommendation for what to advocate for when I initially wrote this piece was solidly score voting, because of its apparent simplicty. The more I thought about it though, I’m not sure simplicity actually works in its favor. Score voting is simple to describe, sure, but understanding why we’d expect it to work well - how it would lead to somewhere between the Condorcet and utilitarian winner - is actually very complicated.

And how do you sell politicians on why to adopt score voting? The first instinct I think is to say “people should be allowed to express their degree of preferences rather than just their top choice/ranking”. But the natural followup question, and the one that is actually really complicated to answer, is why people shouldn’t just rank their top choice 10/10 and all the rest of them 0/10. And the line about degree of preferences is also just not very compelling: who looks at politics and says “the underlying issue is that voters aren’t able to express degree of preference well”?

(I do think score voting is the obvious choice for primaries, as I discuss in this piece, and I think the degree of preference thing is more intuitive there.)

I think this complexity of justification, even when the system itself is not very complex, explains why score voting hasn’t been adopted basically anywhere. Approval voting, the binary version of score voting, has been adopted, and in those cases it’s been sold as a tool to promote positivity in politics. But that explanation is straight-up BS; limiting people’s choices to binary yes-no does nothing to encourage positivity if people are rational. I also just intuitively dislike approval voting. It does not sit right with me to have a system whose optimal strategy is a randomized one.

Ranked Condorcet systems also haven’t been adopted anywhere, probably on account of their apparent complexity. I do think though that they would probably be easier to adopt in real elections, and top-4+Minimax is the system I would advocate for IRL. I wrote another post for a wider audience advocating this system and the best terminology to use for it: “head-to-head” for Condorcet and “highest floor” for Minimax.

If you enjoyed this part, be sure to read the second part of this post, about electoral systems for electing a multi-member body like a legislature!