I can potentially see some benefits. But there's definitely some drawbacks too. Single member districts are good because it tends to make that member more responsive to the needs of their district. That's a good thing IMHO.
Also, single party rule doesn't have to end badly. Look at Utah. Deep red, but with a well functioning government.
I suspect their strong religious background has something to do with it. More people that are truly concerned about being public servants rather than personal enrichment.
This was an interesting piece. That said at the same time, I'm always suspicious of electoral reforms that it appears are also likely to increase the frequency of outcomes the promoter defines as desirable.
The beauty of finding the right electoral system is that it can improve outcomes for nearly everyone. It's the limits of our current election system makes us think that if someone wins someone else has to lose. But actually, win-wins are very possible and very neglected.
But a good electoral system is supposed to be representative, responsive, and functional - and we can improve on all those things simultaneously. If LA goes proportional, it helps niches interests and it helps consensus. If it expands the number of members at the same time, the current members probably don't even have to risk their seats.
It's only those that have an interest in perpetuating the status quo (at any cost including stagnation) or those who have an interest in generating frusteration / distinterest in elections (who prefer authoritarian or machine politics). But if we can agree on a functional democracy as the goal, all else follows.
> proportional representation elections for multi-member districts encourage the formation of multiple parties.
I feel like this argument rests on this 'encourages' being understood as 'will', which I don't really find convincing. This is America, it's a deeply two party place, and has been for hundreds of years. What if we just got more anonymous Democrats, more disenfranchised Republican bomb throwers, and nothing changed?
You can already organize branded sub parties today, and I think there are important reasons why that doesn't work and people don't do it.
Interesting but Americans are used to having a person to whom they can go when a problem needs a solution. How will this system deal with the pretty widespread desire for an ombudsman?
<I> When selecting a candidate for Congress, a voter can instantaneously know a candidate’s position on abortion, gun control or climate change just by knowing which party that candidate belongs to.</I>
That’s not true. It depends on what it takes to win. If you’re a congressional Democrat in Texas (there are 13) you modify your position as needed.
The Center for Election Science has so far had some success getting cities to adopt Approval Voting for single member districts, but their actual recommendation is Proportional Approval, either for your full legislative body or for multi-member districts. Approval has the advantage of being cheaper to implement, and its tabulation system (which was invented by Thomas Jefferson, and used for the first apportionment of the House) is much easier to follow compared to Instant Runoff's spreadsheet-level complexity.
In any case, I definitely would like to see a proportional system for more legislatures. California ought to have some Silicon Valley Libertarians in the legislature, and some Central Valley representatives in the tradition of the "Democratic Farmer Labor" alliance.
Does anyone know whether San Francisco has this "privilege" system for its Board of Supervisors? I can't seem to find anything about it online one way or the other.
LA city council has this in a particularly strong way. I don’t believe counties in California tend to operate that way with their five member boards of supervisors, and San Francisco is for some reason governed as a county rather than as a city.
Nice. Thanks for sharing. I wonder how, practically speaking, political entities (cities here, but why stop there?) might initiate a political self-reconceptualization. Like: what's the mechanism that drives a political entity to reimagine its own organization. Given the existence of stakeholders highly committed to the dysfunctional status quo, given the lesser intensity (usually) of those who advocate reform, given that a reconceptualization might arouse the intensity of any number of new possible stakeholders (good or bad!), what does it take to get things moving forward in a productive direction without running off the rails? There must be examples of this around the world. Maybe a topic for a future think-piece!
Spot on, Stan. I agree about the benefits of strong parties and multi-party democracy, and I also think cities are the unit of government where it makes the most sense to start implementing these changes. It’s more tractable to begin a bottom-up improvement in the system since local parties can reasonably abstain from a lot of the culture-war paralyzing Washington since they wouldn’t have jurisdiction over it anyway.
I very much agree on the problem of one-party government, which happens all over the place, but, for LA then, how do these proposed reforms get implemented? Is the idea to do a proposition that is voted on by LA voters? Or would it be a statewide thing for all city councils in California? I'm not sure how you persuade turkeys to vote for Christmas if the current city council have to approve such measures.
This is a great article, and one I've been thinking about with the upcoming mayor's race.
One thing that I've been wondering about is the number of cities in the LA Metro Area, and how that might be making things more dysfunctional. As I go through things like city council agendas each week, it's hard to see that we're really gaining much by giving, say, City of Industry it's own city council.
I've been pleased by how serious LA has taken stuff like their housing element update though.
In Los Angeles, no matter how you slice and dice the electorate, I am afraid you are searching in vain for a large constituency that is saying "Yes, build a high rise apartment building in my neighborhood, take away a lane of traffic and I will obediently trade my car for a bicycle and a bus pass."
I've been that constituency in LA, and here's why: A high rise apartment would have made it cheaper to rent near my workplace; I didn't own a car; Living near my workplace it would have been convenient to bike (until my bike got stolen but that's a whole nother LA problem).
I’m curious, why not a list system? (Either MMP or elect by lists from larger regions/districts, e.g. 15 members each from 5 regions). Certainly it’s simpler.
Agreed. Maybe part of the council could be made up of single-member districts, and part could be 'at large' from a list. Seems like it would resolve the parochial issues while still having members who could take a larger view of the city's issues
Yeah I do think American cities could be a good place to try MMP. But we do have to be careful that it is MMP (where the party list vote determines the overall composition of the legislature, even if you sometimes have to add seats to the legislature to preserve that principle) and not a parallel voting system (where the party list vote only determines the composition of a fixed block of seats, without reference to the outcome of the elections for single-member districts). The latter isn’t really PR and also invites gerrymandering (it’s how Orbán keeps control of Hungary).
Also, random thought: If you then also made the Mayor elected by the council, you could run the election as though it were 2 separate ones (column A: Which person do you want to be your councilor, column B: which person do you want to be Mayor).
I think there are two basic problems with MMP in this situation.
First, how do you do primaries for a list? And I can't see Americans comfortably accepting party-chosen lists.
Second, if the Democratic Party enters a list, then that will just dominate the election.
The big benefit of STV is that it works with informal parties just as well as with formal ones, so the factions of the Democratic Party can run against each other, even if every candidate formally has "Democrat" on the ballot paper.
Proportional representation is a fantastic goal, but incredibly complex to implement. A simpler solution would be moving to approval voting for all of the single winner elections, which would help elect moderates in each of the districts, and would ultimately represent a moderate city Council for Los Angeles. Center for Election Science, and California approves is here to enable that to happen. We’re happy to work with you.
I don’t think moderation is the issue in Los Angeles city council - rather, it’s that candidates don’t stand for issues at all, but instead just have a bunch of personal relationships.
I sure hope approval voting helps! I find it interesting, and would particularly like to use it for party primaries. However, decades of evidence from real-world implementation of AV shows that the vast majority of users just bullet vote- they only vote for one person. Bullet voting rates of 80% have been found consistently, which is why organizations as diverse as the IEEE and Dartmouth College have tried and eventually rejected AV.
Also, AV doesn't do much when there are too many candidates- with 6-8+ candidates per election, the winner typically has a small plurality. This is specifically why Dartmouth College ceased using AV
Various flavors of proportional representation are how almost every democracy in the world does elections. The American-centric nerd war among single winner election reforms (IRV, approval, STAR, etc) is peculiar to the US's strong aversion to the idea of political parties.
I think that if we used an electoral system that could elect minority parties (which is the whole point of PR), Americans would discover that there are parties that match their political interests. It was choosing between only two parties that they hated.
Also, if you did want to implement proportional representation, proportional approval voting is a much simpler way to do it, rather than fiendishly complex single transferable vote.
>>At this number, each new member would theoretically represent about 52,000 constituents, (one-fifth of each 259,000 person district.)<<
No, they wouldn't. Each of the five district members would represent 259,000 people. They'd share the 259,000 with the four other council members and none of the five would be particularly motivated to see a group of residents as *their* constituents. Perhaps those council members would be motivated to seek the higher aims of the city as a whole or perhaps they'd be beholden to money interests or some kind of party machinery: it's hard to know.
And perhaps breaking the intimate tie between the council member and a specific geographic area and its residents would be good, as the bus route example illustrates. Maybe. Or it could be like Israel, with a single 120-member national constituency and no real time between any member of Knesset and a local group of voters, but only ties to larger and smaller (and very tiny) parties.
I don't know what the best solution is. The other evening I attended the public comment meeting on the proposed LA City Council redistricting and heard impassioned remarks by organizations and citizens mostly centered on keeping their neighborhoods and areas of common interest intact in any proposed council district. Maybe that would further undermine the city by emphasizing parochial concerns, or maybe it would enhance democracy by strengthening ties between citizens and their elected representatives. I think the tradeoffs are so great that it's hard to know exactly what the ramifications would be of swapping electoral systems out.
Each of the five district members would need at most 52,000 votes/supporters to be elected. You can modify the demographics of those 52,000 and pick up other types of supporters in your district, but in the end, you are representing the interests of those who voted for you and will vote for you. Why would the YIMBY slate representative in district X need to represent the 3/5 of the district that is NIMBY?
The thing about proportional representation is that it _lets_ you win a seat based on appealing geographically -- if you have deep enough support from one geography to put together the requisite votes, you get a seat.
But it doesn't _make_ you win a seat that way. There are other paths to victory, other ways to piece together a winning coalition. It allows people to foreground other identities, rather than ONLY thinking about geography.
I can potentially see some benefits. But there's definitely some drawbacks too. Single member districts are good because it tends to make that member more responsive to the needs of their district. That's a good thing IMHO.
Also, single party rule doesn't have to end badly. Look at Utah. Deep red, but with a well functioning government.
I suspect their strong religious background has something to do with it. More people that are truly concerned about being public servants rather than personal enrichment.
This was an interesting piece. That said at the same time, I'm always suspicious of electoral reforms that it appears are also likely to increase the frequency of outcomes the promoter defines as desirable.
The beauty of finding the right electoral system is that it can improve outcomes for nearly everyone. It's the limits of our current election system makes us think that if someone wins someone else has to lose. But actually, win-wins are very possible and very neglected.
But a good electoral system is supposed to be representative, responsive, and functional - and we can improve on all those things simultaneously. If LA goes proportional, it helps niches interests and it helps consensus. If it expands the number of members at the same time, the current members probably don't even have to risk their seats.
It's only those that have an interest in perpetuating the status quo (at any cost including stagnation) or those who have an interest in generating frusteration / distinterest in elections (who prefer authoritarian or machine politics). But if we can agree on a functional democracy as the goal, all else follows.
> proportional representation elections for multi-member districts encourage the formation of multiple parties.
I feel like this argument rests on this 'encourages' being understood as 'will', which I don't really find convincing. This is America, it's a deeply two party place, and has been for hundreds of years. What if we just got more anonymous Democrats, more disenfranchised Republican bomb throwers, and nothing changed?
You can already organize branded sub parties today, and I think there are important reasons why that doesn't work and people don't do it.
Interesting but Americans are used to having a person to whom they can go when a problem needs a solution. How will this system deal with the pretty widespread desire for an ombudsman?
You get five (or whatever number your district has) people you can go to, instead of only one.
<I> When selecting a candidate for Congress, a voter can instantaneously know a candidate’s position on abortion, gun control or climate change just by knowing which party that candidate belongs to.</I>
That’s not true. It depends on what it takes to win. If you’re a congressional Democrat in Texas (there are 13) you modify your position as needed.
The Center for Election Science has so far had some success getting cities to adopt Approval Voting for single member districts, but their actual recommendation is Proportional Approval, either for your full legislative body or for multi-member districts. Approval has the advantage of being cheaper to implement, and its tabulation system (which was invented by Thomas Jefferson, and used for the first apportionment of the House) is much easier to follow compared to Instant Runoff's spreadsheet-level complexity.
In any case, I definitely would like to see a proportional system for more legislatures. California ought to have some Silicon Valley Libertarians in the legislature, and some Central Valley representatives in the tradition of the "Democratic Farmer Labor" alliance.
Does anyone know whether San Francisco has this "privilege" system for its Board of Supervisors? I can't seem to find anything about it online one way or the other.
Yes, SF's supervisors exert significant (informal) control over changes affecting their district; Peskin is notorious for this.
LA city council has this in a particularly strong way. I don’t believe counties in California tend to operate that way with their five member boards of supervisors, and San Francisco is for some reason governed as a county rather than as a city.
I agree that this would be an improvement over the status quo, but wouldn't STAR voting be simpler and also address these concerns?
Nice. Thanks for sharing. I wonder how, practically speaking, political entities (cities here, but why stop there?) might initiate a political self-reconceptualization. Like: what's the mechanism that drives a political entity to reimagine its own organization. Given the existence of stakeholders highly committed to the dysfunctional status quo, given the lesser intensity (usually) of those who advocate reform, given that a reconceptualization might arouse the intensity of any number of new possible stakeholders (good or bad!), what does it take to get things moving forward in a productive direction without running off the rails? There must be examples of this around the world. Maybe a topic for a future think-piece!
Spot on, Stan. I agree about the benefits of strong parties and multi-party democracy, and I also think cities are the unit of government where it makes the most sense to start implementing these changes. It’s more tractable to begin a bottom-up improvement in the system since local parties can reasonably abstain from a lot of the culture-war paralyzing Washington since they wouldn’t have jurisdiction over it anyway.
I very much agree on the problem of one-party government, which happens all over the place, but, for LA then, how do these proposed reforms get implemented? Is the idea to do a proposition that is voted on by LA voters? Or would it be a statewide thing for all city councils in California? I'm not sure how you persuade turkeys to vote for Christmas if the current city council have to approve such measures.
Exactly.
This is a great article, and one I've been thinking about with the upcoming mayor's race.
One thing that I've been wondering about is the number of cities in the LA Metro Area, and how that might be making things more dysfunctional. As I go through things like city council agendas each week, it's hard to see that we're really gaining much by giving, say, City of Industry it's own city council.
I've been pleased by how serious LA has taken stuff like their housing element update though.
In Los Angeles, no matter how you slice and dice the electorate, I am afraid you are searching in vain for a large constituency that is saying "Yes, build a high rise apartment building in my neighborhood, take away a lane of traffic and I will obediently trade my car for a bicycle and a bus pass."
I've been that constituency in LA, and here's why: A high rise apartment would have made it cheaper to rent near my workplace; I didn't own a car; Living near my workplace it would have been convenient to bike (until my bike got stolen but that's a whole nother LA problem).
I’m curious, why not a list system? (Either MMP or elect by lists from larger regions/districts, e.g. 15 members each from 5 regions). Certainly it’s simpler.
Agreed. Maybe part of the council could be made up of single-member districts, and part could be 'at large' from a list. Seems like it would resolve the parochial issues while still having members who could take a larger view of the city's issues
Yeah I do think American cities could be a good place to try MMP. But we do have to be careful that it is MMP (where the party list vote determines the overall composition of the legislature, even if you sometimes have to add seats to the legislature to preserve that principle) and not a parallel voting system (where the party list vote only determines the composition of a fixed block of seats, without reference to the outcome of the elections for single-member districts). The latter isn’t really PR and also invites gerrymandering (it’s how Orbán keeps control of Hungary).
Also, random thought: If you then also made the Mayor elected by the council, you could run the election as though it were 2 separate ones (column A: Which person do you want to be your councilor, column B: which person do you want to be Mayor).
I think there are two basic problems with MMP in this situation.
First, how do you do primaries for a list? And I can't see Americans comfortably accepting party-chosen lists.
Second, if the Democratic Party enters a list, then that will just dominate the election.
The big benefit of STV is that it works with informal parties just as well as with formal ones, so the factions of the Democratic Party can run against each other, even if every candidate formally has "Democrat" on the ballot paper.
List-type systems tend to require formal parties.
Proportional representation is a fantastic goal, but incredibly complex to implement. A simpler solution would be moving to approval voting for all of the single winner elections, which would help elect moderates in each of the districts, and would ultimately represent a moderate city Council for Los Angeles. Center for Election Science, and California approves is here to enable that to happen. We’re happy to work with you.
I don’t think moderation is the issue in Los Angeles city council - rather, it’s that candidates don’t stand for issues at all, but instead just have a bunch of personal relationships.
I sure hope approval voting helps! I find it interesting, and would particularly like to use it for party primaries. However, decades of evidence from real-world implementation of AV shows that the vast majority of users just bullet vote- they only vote for one person. Bullet voting rates of 80% have been found consistently, which is why organizations as diverse as the IEEE and Dartmouth College have tried and eventually rejected AV.
Also, AV doesn't do much when there are too many candidates- with 6-8+ candidates per election, the winner typically has a small plurality. This is specifically why Dartmouth College ceased using AV
Various flavors of proportional representation are how almost every democracy in the world does elections. The American-centric nerd war among single winner election reforms (IRV, approval, STAR, etc) is peculiar to the US's strong aversion to the idea of political parties.
I think that if we used an electoral system that could elect minority parties (which is the whole point of PR), Americans would discover that there are parties that match their political interests. It was choosing between only two parties that they hated.
We’re already working to bring approval voting to Culver City!
Also, if you did want to implement proportional representation, proportional approval voting is a much simpler way to do it, rather than fiendishly complex single transferable vote.
>>At this number, each new member would theoretically represent about 52,000 constituents, (one-fifth of each 259,000 person district.)<<
No, they wouldn't. Each of the five district members would represent 259,000 people. They'd share the 259,000 with the four other council members and none of the five would be particularly motivated to see a group of residents as *their* constituents. Perhaps those council members would be motivated to seek the higher aims of the city as a whole or perhaps they'd be beholden to money interests or some kind of party machinery: it's hard to know.
And perhaps breaking the intimate tie between the council member and a specific geographic area and its residents would be good, as the bus route example illustrates. Maybe. Or it could be like Israel, with a single 120-member national constituency and no real time between any member of Knesset and a local group of voters, but only ties to larger and smaller (and very tiny) parties.
I don't know what the best solution is. The other evening I attended the public comment meeting on the proposed LA City Council redistricting and heard impassioned remarks by organizations and citizens mostly centered on keeping their neighborhoods and areas of common interest intact in any proposed council district. Maybe that would further undermine the city by emphasizing parochial concerns, or maybe it would enhance democracy by strengthening ties between citizens and their elected representatives. I think the tradeoffs are so great that it's hard to know exactly what the ramifications would be of swapping electoral systems out.
Each of the five district members would need at most 52,000 votes/supporters to be elected. You can modify the demographics of those 52,000 and pick up other types of supporters in your district, but in the end, you are representing the interests of those who voted for you and will vote for you. Why would the YIMBY slate representative in district X need to represent the 3/5 of the district that is NIMBY?
The thing about proportional representation is that it _lets_ you win a seat based on appealing geographically -- if you have deep enough support from one geography to put together the requisite votes, you get a seat.
But it doesn't _make_ you win a seat that way. There are other paths to victory, other ways to piece together a winning coalition. It allows people to foreground other identities, rather than ONLY thinking about geography.