This transcript has been lightly edited for clarity.
David Beard:
Hello and welcome. I’m David Beard, Contributing Editor for Daily Kos Elections.
David Nir:
And I’m David Nir, Political Director of Daily Kos. The Downballot is a weekly podcast dedicated to the many elections that take place below the presidency, from Senate to city council. Please subscribe to The Downballot on Apple Podcasts and leave us a five-star rating and review.
David Beard:
Now, we’ve got quite the interesting show for our listeners today, don’t we?
David Nir:
We do indeed. First, we have to talk about an effort by Montana Republicans to rig next year’s Senate race that just went down in flames. Then we are going to discuss why New York could get a new congressional map for 2024, now that the state’s top court is finally back at full strength. And then coming up, we have a fascinating guest, Professor Matthew Shugart, who is a professor of political science and an expert on electoral systems around the world. He is going to tell us all about how other countries elect their governments and the changes that we could make in America to improve our democracy. Fantastic episode. Let’s get rolling. So for our Weekly Hits, we have to follow up on some news we talked about last week, and this time we have good news to report.
David Beard:
I know. That’s always so much better than bad news when we come and report.
David Nir:
So much better.
David Beard:
As we talked about last week, Republicans in the Montana state Senate passed a bill to change the rules just for the 2024 U.S. Senate Election, basically to try to make it harder for Jon Tester to win reelection by making sure the battle was just Tester and a Republican, as opposed to potentially a Libertarian who’s been on past ballots. Now, fortunately, the Montana House Republicans did not feel the same way as the Montana Senate Republicans. The Montana House Committee has voted down that bill on a 17 to 1 vote, so nearly unanimously, which almost certainly ends its chances of becoming law. Now, an interesting quote I saw from one of the reports was from a Republican House member, Greg Frazier, and he said, “I’ve had a lot of my folks from back home reach out to me and ask me to vote no on this a lot, more than what I thought. It’s actually been pretty interesting.”
And I think there’s two reasons for that. One, it was a really unfair bill. It was not changing the electoral system of Montana. It was targeting one elected official to try to make it easier for Republicans to beat him. And I think everyday people have a sense of fairness about elections. They obviously have a candidate they want to win, but most people understand that electoral systems should be fair and we should follow the rules of the system that’s laid out for everybody, not change them for just one candidate or one race. And then, secondarily, of course, people get used to certain systems and they don’t like change, and that can also sometimes inhibit good change. But in this case, I think it probably helped a little bit. People didn’t want to change the system. They’re used to the system. So for those reasons, I think the Republicans got a lot more blow back here than they would’ve expected and as a result they’ve decided to end this little quest to make Tester’s reelection a little harder. And so he’ll go forward with the same system that he’s won election three times before.
David Nir:
I think you’re exactly right. One of the topics that most lights up the Daily Kos community is redistricting, particularly gerrymandering. And you think, “Oh, this is an obscure issue. Who really gets into the nitty-gritty of this? Who follows this stuff?” Well, you don’t really have to get into the nitty-gritty of it to understand the unfairness of rigging the maps. And this was essentially electoral gerrymandering. Like you said, they wanted to carve out this one race from all the others, and not just this one race, but also only this particular election cycle that they were going to go straight back to the old system for future Senate races. And when you try to do that kind of thing, it tends to rub voters the wrong way.
I think that Massachusetts Democrats actually faced similar blowback ahead of the Scott Brown special election when they tweaked the rules to ensure that they could get a placeholder after Ted Kennedy died, but before a special election could take place. And Scott Brown wound up winning that race. So Montana Republicans, I don’t know, I think this would’ve made Tester pretty sympathetic had they gone forward with it. They probably caught a lucky break that they got all this blowback when they did.
David Beard:
And I think it’s a lot easier to swallow. Obviously, it’s extremely disappointing when you lose a race, but when the other candidate gets more votes, that’s how democracies are supposed to work. It’s an opportunity for a reflection; why did our side lose? Why weren’t we able to convince enough voters that we were the right choice? But when things like gerrymandering, things like this, happen, and gives folks the sense that the candidate with the most votes might not win or the candidate of the majority of the electorate’s not going to win because of rules changes or trickiness behind the scenes, it really poisons democracy and makes people cynical about the whole exercise.
David Nir:
So the one other story that I wanted to make sure we hit this week is back in my home state of New York where a long, long saga finally came to an end. The state’s highest court, which has the confusing name of the Court of Appeals, is finally back at full strength after being down a member for the better part of the year. The Democratic-dominated state Senate confirmed Rowan Wilson as chief judge. He is currently an associate judge on the court, so to take his place, they also confirmed Caitlin Halligan as a new associate judge. She’s a former state solicitor general who’s currently in private practice. I should add, full disclosure, I worked for Halligan at the New York Attorney General’s Office 20 years ago—but the most important question going forward now is, will all of this lead to New York getting yet another new congressional map?
It’s very possible that it could. You’ll recall that last year the Court of Appeals, then dominated by a conservative majority, said that state lawmakers could not step in to draw their own congressional map after the state’s evenly divided bipartisan redistricting commission failed to do so and instead handed the job to a state court. A group of voters has since filed a lawsuit saying that, “Okay, well, you had that court drawn map for 2022, but for the rest of the decade, the Redistricting Commission should be the one to draw a new map,” and they want the courts to order the commission to reconvene so that it can create a new map. That argument was rejected by a trial court, but the case is on appeal now before the state’s intermediate appellate court. That’s called the Appellate Division. And one way or another, this case is likely to wind up back in front of the Court of Appeals.
If the Court of Appeals agrees with the plaintiffs, then, yes, it would send the map making process back to the Redistricting Commission. Now, there is no way to know how the court will actually vote. Wilson, the new chief judge, was in the minority who dissented when the Court of Appeals ruled last year that lawmakers couldn’t draw a new map and ordered a court to draw a map instead. There’s no way to know how Halligan will vote though. But let’s say that the Court of Appeals does agree with plaintiffs and sends this back to the Redistricting Commission, then the real question becomes, is there any reason to think that this commission will actually produce a map in accordance with the state constitution? Like I said, it’s evenly divided between Republicans and Democrats, but in order for it to pass a map, any map has to win bipartisan support.
That failed to happen in 2021, which is why Democratic lawmakers tried to step in and they were told that they couldn’t. So is the endgame here that plaintiffs are hoping to send the process back to the Redistricting Commission, have the commission fail again, and then see if Democratic lawmakers try to draw a map again, but hope that the Court of Appeals this time says it’s kosher for them to step in? I know it’s very roundabout, a lot of moving parts here, and we’re definitely a long ways off from a final decision. But if that all comes to pass, then we could wind up with a Democratic gerrymander of New York’s congressional map, which is exactly what Democrats had hoped to put in place last year and were denied by the courts.
David Beard:
And, of course, this is really important because there’s a number of seats in New York that Republicans won in 2022 that are already going to be very competitive. And obviously with a new map that is more pro-Democratic, those seats would become even easier for Democrats to pick up under a potential new map. Of course, it’s important to remember that we expect Ohio and North Carolina to implement more Republican maps because they have control of the Supreme Courts in those states and they’re expected to be allowed to basically draw whatever maps they want due to those Conservative Supreme Courts. So that’s something that’s already being factored into 2024 and getting a positive result from New York with a more Democratic map to help balance that out could be an important factor as we look to the House in 2024.
David Nir:
And this is a good time to say once again as we have many times in the past, we are adamantly opposed to gerrymandering and we think that a national ban on gerrymandering is something Congress absolutely should pass. And the only reason why it hasn’t is because Republicans oppose it. Democrats in Congress have universally backed a ban on congressional gerrymandering. It would be good for the country, it would be good for democracy, but that isn’t going to pass unless and until Republicans get on board.
And since they haven’t and since they aren’t going to, then we cannot unilaterally disarm and since, Beard, like you just said a moment ago, we know that North Carolina and Ohio Republicans are going to ram through maps that are much more favorable to the GOP, Democrats have to fight fire with fire, and if they can get the chance to do so in New York, they have to take it.
That does it for our Weekly Hits. Coming up, we are going to be discussing electoral systems in other parts of the world and how they differ from how we do things here in the United States. Our guest joining us is Matthew Shugart, who is a political science professor emeritus at the University of California, Davis. It is a really fascinating in-depth and weedy conversation to chew on.
Matthew Shugart is Distinguished Professor Emeritus in the department of Political Science at the University of California, Davis. Professor, thank you so much for joining us on The Downballot today.
Matthew Shugart:
Thank you for having me on the podcast.
David Nir:
So, professor, this show is focused entirely on U.S. politics and I think most Americans really probably are only familiar with our electoral system here where we have Congress and the president, and then after that, most folks don’t really have a reason to be familiar with how things are done overseas. But I would be really grateful for you to kick off this conversation by giving us an overview of some of the most popular electoral systems that you’ve seen used in other parts of the world.
Matthew Shugart:
I’m going to give you an answer that is both the common categories that people use when thinking about electoral systems and then another way that’s perhaps a bit more useful for thinking about them. Most people when they think about electoral systems, including people in my profession of political science, say there are two broad categories: majoritarian, also sometimes in the U.S. more often known as winner take all, and on the other hand, proportional, proportional representation.
So, that’s useful for some reasons. But in thinking about comparing across systems, proportional is an extremely wide category. There’s so much variation within proportional that it’s almost not a meaningful category, it’s just that all of those systems that we call proportional somehow allow … The roles are different in different countries, but the principle is to allow parties to get seats in the legislature that is somehow roughly proportional or sometimes perfectly proportional, depending on details of the system with their vote shares. So, a party that gets 40% of the votes should have about 40% of the seats, but there are lots of variations within that.
And where they stand, of course in stark contrast to what we use in the U.S., is that we generally have every member of the legislature, and certainly this is the case in the House of Representatives. Every member of the legislature is elected in his or her own local district, and that whoever wins the most votes gets therefore all the representation of that district. And so, when we contrast them, we tend to say majoritarian versus proportional, but there’s a world of difference among proportional systems.
And maybe we can get into that a little bit. That’s just a starting point. The most important thing I would say about that variety of proportional systems is how many seats are there. Does your district elect 150, as in the Netherlands, or is the country divided into several, say five or six seat districts like Ireland?
David Nir:
So, that’s a very interesting breakdown between majoritarian and proportional. Can you maybe give us a few well-known examples of countries, especially any that Americans might be a little more familiar with, like other English-speaking countries that fall into the proportional category and how they operate?
Matthew Shugart:
Sure. Among English-speaking countries, actually, there are very few that use any kind of proportional system. The majoritarian winner take all type system is really a legacy of Britain, because it’s what Britain itself uses. But the most prominent example would be New Zealand. So, New Zealand formerly used a majoritarian system, technically what they would call and what many people in Britain and Canada call first pass the post. And you hear that term used in the US, but it’s not quite as common here as a term for the system as it is elsewhere.
But in 1996, New Zealand started using a form of proportional representation. And so, it’s our best example. They use what’s called mixed member proportional, which we might talk about in further detail a little bit later. But the broad description of that is they still elect about 65 of their 120 members parliament in local single seat districts. And then they elect the rest from party lists, and in a manner that ensures that overall nationwide the parties are represented quite close to perfect proportionality.
David Nir:
So, party lists, I think that is a term that is so different from the American experience. So, maybe you can talk about that, because my understanding is that’s a fairly common approach in a lot of countries.
Matthew Shugart:
Correct. Most countries that use proportional representation use some kind of list system. And by list system, we mean that parties present lists of candidates. And see, the reason for that is that you have to have districts that elect more than one member, multi seat districts in order to do proportional representation. Because you can’t proportionally divide one person into a third of that person represents the Republican Party, and a third of that person represents the Democrats, and a third represents some other party. You can’t do that. So, you would need to have districts electing more than one member a piece. And the most common way, but not the only way that this is done, is to have each party present a list of candidates. So, they’ve nominated candidates before the election.
Let’s suppose it’s a five seat district and they’ve won enough votes to get three of those five seats, then it’s the top three candidates on their list who are elected. And then those other two maybe went to other parties. Each of them will send the top candidate on their list. And so, party lists can be ranked either by the party itself, which is what they do in New Zealand, or they can be ranked by the voters, where the voters select candidates within a list, which is called an open list system. And there are many other combinations, but those are good starting points for understanding it.
David Nir:
So, in that example you gave of that five-member district with proportional representation, the party winning three seats, they would say more or less probably have to win 60% of the vote. In other words, three-fifths of the vote to get three out of those five seats.
Matthew Shugart:
Well, not necessarily, because if they won half the votes they would still be getting three of the seats. That’s a really good point. I’m glad you asked that question, because you see, if you have only five seat districts, the proportionality will be somewhat less than complete, but still better than if you had only one seat. Imagine you have the largest party getting only half the votes, which is a pretty common situation in the U.S., they would still get all the representation in the district. The one seat. If there are five, they’re going to get three of them probably, because what’s half of five? It’s two and a half, but you can’t give them two and a half seats. So, it would round up and they get three. But if you have 10 seats available, then they’ll probably get five of the 10.
So, the larger the district magnitudes are… I use the technical political science, so let me stop and define it. District magnitude is the number of seats in a district. So, in the U.S. it’s one, in the House of Representatives, for example, most state legislatures, but not all of them. And the example we were using, the district magnitude is five. In another system, maybe it’s 10. In the Netherlands, it’s 150 and so on. It can be in principle any number. And the higher that number is, as long as you’re using some kind of proportional algorithm for turning both to the seats, the higher that magnitude is, the more proportional the result will be.
David Nir:
Now we’ve talked about the majoritarian and the proportional, and obviously within both of those, there are variations that we’ve started getting into. Do you think there’s one system that’s maybe seen as best in class or a couple of systems that are generally seen as what you would recommend to a country reforming its electoral system or a new country, something like that, that’s really hitting the most goals in terms of what people want out of a democracy?
Matthew Shugart:
I actually co-edited the book many years ago called “Mixed-Member Electoral Systems: The Best of Both Worlds?”, with a question mark. And in research I’ve done more recently in my reading of research that others have done, I think we can remove that question mark. I think the best in class electoral system is the one that I mentioned earlier that New Zealand now uses, has used for almost 30 years now, which is the mixed-member proportional or sometimes called MMP. And so, that’s the system that has both single seat districts where the candidate with the plurality wins in that district and has party lists, and the voter has voted both. The voter, every voter votes, has the option to pick one of several candidates running in their district and also to choose a party list, that need not be the same party as the candidate they selected. And without going into the mathematical details, that the system, it’s set up to ensure that parties are compensated.
So, if they don’t win very many of the single seat districts, maybe they’re a small party or a party with its votes distributed all around the country, but not strong anywhere. If they don’t have many single seat districts, don’t win more off their list. Or if they won a lot of the single seat districts, they won’t get as many on the list. So, if the overall result is proportional. Why is this beneficial? And the research that I and others have done show that it allows for good local representation, members will still pay attention to the districts in which the election is contested, but also strong proportional representation so that small parties are represented.
And really, I think an important point that gets overlooked is that the big parties have an incentive to cater to voters regardless of where they live. They don’t just focus on their safe seats and a few competitive or swing districts, but wherever their voters are, it will help them win a certain proportional share of the seats. So, the National Party in New Zealand, which is their rough equivalent to the Republicans serve, to the British Conservatives, the center-right party in other words, they will pay attention to urban voters. And the Labour Party, which is their party at the center-left, will pay attention to voters in the rural areas. So, it’s a very good system.
Whether it would work in the United States is an interesting question. There may be some reasons why we should go in a different direction. But best in class? Yes, it’s MMP. And then maybe I’ll mention two others that are often rated highly by specialists. The single transferrable vote, which I’m sure we need to talk about, because there’s lots of reformers in the United States like it. It’s a form of ranked choice voting. Or there’s the open list proportional system, which I alluded to earlier where the parties present lists, but they don’t rank them, the voters do by selecting a candidate within the list. Every voter votes for a candidate within the list. So, those are the three, mixed member proportional, single transferrable vote and open list, probably like the top three in class. They all have problems and they all have good things about them. That’s the nature, but there’s no single perfect system. But if I had to say there’s one system that’s best worldwide, I’m going to go with MMP.
David Nir:
Professor, you threw a couple of new terms at us in your last answer, so I would love to hear a little bit more, especially the single transferrable vote, which you didn’t get into detail just yet, but I definitely would like you to.
Matthew Shugart:
Okay, yes. I call this the occupational hazard of people who specialize in electoral systems, as there’s always another terminology item to throw out into the conversation. So, single transferable vote, it’s not a great name, STV, but that is the name that political scientists generally know it by. Yet to the extent that people in the general public in the United States have heard about this, they probably tend to know it as ranked choice voting or ranked proportional.
So, we need to back up a little bit that rank choice voting means that the voter is allowed to say, here’s the candidates, that is my first choice. Here’s the candidates, that’s my second choice. Here’s my third, and so on. Usually they don’t have to rank all of them, although in Australia, often, depending on which jurisdiction we’re talking about, they often do have to rank every candidate on the ballot. In other jurisdictions, if you want to just vote for one it’s okay.
But the principle behind them is that if your first choice candidate can’t be elected, then your vote will instead be counted towards your second one. If your second one can’t, you’ll be counted towards your third and so on. So a lot of people find this attractive, because it seems to eliminate the spoiler problem that, if I really like one of the minor party candidates but I’m worried that that person can’t win, my default then is I like this other one almost as much, and so my vote can count for the second one if my first one can’t win the seat.
Now, in most jurisdictions in the U.S. that use something called ranked choice voting, they’re electing only one candidate. So it’s important to realize that RCV, or ranked choice voting, could still be a majoritarian system. If you’re electing only one, it’s still 100% of the representation in that office is coming from, is held by whichever party emerged from the count as the winner. It’s a winner take all system still. So RCV, ranked choice voting, is not by itself a proportional system. And that’s an important thing, because I sometimes hear people conflate them, say RCV is a form of proportional representation, what we in the trade call PR. It’s not, but it can. So RCV is just a ballot format, and an algorithm for turning the votes cast into representation.
If you carry the RCV process that I just described in a multi-seat district, you can then have a form of proportional representation. Because, let’s go back to our five seat district example. Let’s suppose there’s a five seat district as we were talking about in our earlier example, and this jurisdiction uses proportional ranked choice voting, or what political scientists often call single transferable vote, STV. It typically uses something called a Droop Quota, named after Henry Richmond Droop, who wrote about this system and advocated it back in the 1860s. The notion of the Droop Quota, or the mathematical definition of it, we could call it, instead of a quota, we could call it a winning threshold, because it’s what you need to win a seat. We take the district magnitude and we calculate the quota as one divided by M, the district magnitude. M plus one. And then we add one vote.
Now, why did we do that? That might not seem intuitive at first. Notice that when we do that with M=5, we’re getting one vote more than a sixth, because we’re doing one over six, and then we’re adding one vote. If you think about a single seat district, a winner take all system as we were talking about earlier, it works out just perfectly. Because then your district magnitude, your M, is one. And so you’re doing one over M plus one, that’s a half right? But note that it’s possible for two candidates to have exactly half the vote, so you need a way to ensure that somebody has reached the winning threshold at only M candidates, only whatever the magnitude is, the number of candidates have done so. What’s a majority? 50% plus one. So one over M plus one, and then add one vote. That’s a number that only M candidates can reach, so when M=5, your Droop Quota would be one-sixth. And we’ll ignore the plus one, because it gets in the way. But it’s necessary to avoid the problem I said a moment ago of getting more winners than you can actually have.
So let’s suppose you run the election with STV in a five-seat district, proportional ranked choice voting. The first step is to ask, did any candidate get the quota, the winning threshold? Did any candidate get one-sixth of the votes cast? And if so that candidate’s elected. If not, we’ll eliminate the one in last place and look at their voters’ second choices, and see if that puts, if when we redistribute those votes, if that gets somebody over the winning threshold. If so, that person’s declared elected.
Similarly, let’s suppose the candidate with the most votes, most first choice votes, has more than a sixth. Then the surplus is transferred to those voters’ second choices to see if it can help elect someone else. And it’s a way in which candidates kind of share their votes, right? It seeks to give every one of those five winners roughly the same share of support in the district, whether that was first preferences or built on later preferences. And the reason it might be based on later preferences is either someone else was more popular, had more than they needed, or others were trailing, but their voters liked some other candidate and gave them second choices or third choices.
David Nir:
I want to pause right here and ask a question. So I live in New York City. We recently adopted ranked choice voting for primaries, but of course like you were saying before, you’re only picking one candidate in a party primary. In this situation, so you need roughly 17% of the vote in this hypothetical five-member district. If you have let’s say 30% of the vote and you’re using this single transferable vote system, you’re ranking multiple candidates, whose second choice votes actually get redistributed? Because, is it the people who are … it’s arbitrary, which votes get counted first to get you to that 17% threshold.
Matthew Shugart:
Correct. If you just stop the counting and said, “Oh look, somebody’s reached the winning threshold, so all other votes for that candidate are now going to go to the second choice,” that is very arbitrary. So the way it’s normally done is you take in a fractional share of all the votes that had that candidate ranked first, and then transfer fractional shares to all of those voters’ second preferences, and you do this with all you’re transferring. So everybody’s getting a weighted, all the candidates, excuse me, all the ballots that are more than are needed for that candidate are having an equal chance of effecting who else is elected with that surplus, rather than just which ones happened to be counted later. I think that’s the most straightforward way to do it. There are different specific mechanisms used by different jurisdictions, but that’s the basic principle.
David Beard:
Now I think the single transferable vote in multi-member districts definitely has some positives for it, but I think as we’ve seen just in this conversation is that it’s also one of the most complicated. But I do know that it’s in effect in at least one country which I believe Ireland uses this.
Matthew Shugart:
Correct.
David Beard:
So can you talk about how it sort of worked in Ireland; have they seen it as a success there?
Matthew Shugart:
Yeah. That’s very important to note is it’s used very few places throughout the world, and it’s used really nowhere in really large public elections. Ireland is one of the bigger places that use it, and of course Ireland is not a very large country compared to the United States for example. So it’s worked very well in Ireland, so well that there have been two times over the last, I don’t know, 50, 60 years? I don’t remember the dates precisely that it happened, but there have been two times when an Irish government has wanted to abolish the single transferable vote and replace it with something else, probably first past the post. The Irish constitution says you can’t change the electoral system without a public referendum. And both times the voters said, “Oh no, no. We want to keep STV.” And so obviously the voters like it.
It’s important to note though that it was introduced into Ireland when they already had a functioning party system. They already had more than one party, so it wasn’t just introduced into a context where it was not partisan, as is being, in many of the cases where it’s been adopted so far in the United States, it’s local elections that are either nonpartisan or have a mix of party and independent candidates. And that’s important, because note that when I described how proportional ranked choice works, I kept talking about the candidates. Because indeed, it is a candidate based system. It doesn’t even require parties. Some people might think that’s a good thing, but as a political scientist, I’m a bit skeptical of systems that kind of undermine parties by encouraging voters to jump across party lines in getting the rankings.
Why be skeptical of that? Because parties are critical to democracy. Yeah, local elections might work fine in a nonpartisan context, I have my views about that. Some people like it, some people don’t. But for national politics or even state politics in any of our larger states, it’s really important to realize, what is policy, what is policymaking, what is politics? It’s all about collective choice, and the collective ability of one group or one set of groups to impose themselves on everybody else right? Because once a law is passed, we all have to obey it. So a politics that undermines parties or is nonpartisan, it breaks the fundamental nature of policymaking as a collective enterprise.
I’m very skeptical of STV for this reason, unless it’s introduced in a context in which the parties have strong organizational presence in society, and have a track record. Well obviously Democrats and Republicans have a track record, but a lot of that track record is of divisions and disputes over who controls the meaning of the party and so on, and so I do worry that if STV were used in the United States, it would actually potentially make some of our political problems more difficult to manage rather than easier. And I know this is kind of maybe an unusual perspective that I’m articulating, because lots of political scientists and lots of political reformers really like STV.
And I guess I kind of do too, but I worry about it not being a good choice for the U.S. context, where parties are not, they’re not like Irish parties. They’re not like European parties in general. We don’t have a party system … multiparty system, I should’ve said. We have factions that operate inside the parties, and STV would make the factionalism even more prominent rather than less. And I think that’s something that people should at least think about before they jump into the idea that oh, STV is so nice because it gives voters so much choice, or STV works in Ireland so therefore it’ll work here.
We should always be skeptical of claims like that, because any system has to function in the context in which it’s introduced.
David Nir:
So that’s the perfect intro to the next question I wanted to ask, which is, what systems or changes do you think would work in the United States? You mentioned the New Zealand system, the mixed member proportional system. You said you weren’t necessarily sure that that would be suitable or workable in the United States, and obviously of course there are huge practical hurdles to making major changes like this. But I think it helps to start the conversation with the theory. What in theory do you think actually would help improve our democracy?
Matthew Shugart:
Okay good, thank you for that question. It’s a really important one. So maybe I should start, with your permission, with why I’m not sure MMP, which I just said a while ago, is probably the best-in-class electoral system … well, then why not just come out and endorse MMP for the United States.
David Nir:
Yeah, no, I’d love to hear why.
Matthew Shugart:
MMP, well, it’s really, I do think, is the best in class. It is the best of both worlds, where the one world is the local representation and local knowledge of your legislator that we get in winner-take-all systems, and the other world is the proportionality that encourages parties to seek votes wherever they are instead of just in geographic constituencies and allows parties that are maybe with 10% of the vote to nonetheless have representation and influence. Those are the two worlds, right? And MMP has a really nice balance of these. But it works best, as I said earlier, when the voter is allowed two votes, one for that local representative and one for the party that they want to be proportionally represented.
It hasn’t been a problem in New Zealand, it has not been a problem in Germany, but it’s been a problem in a few other places that have tried MMP. The problem is that parties can get clever with it and they can game the system. They can put up what are called decoy lists. And here’s how a decoy list works. One party nominates candidates in the districts, just like we would expect in any first-past-the-post system, or for that matter any MMP system. They have district-level candidates. And whichever ones get the most votes in their districts will get represented. But then, instead of having a list with the same name affiliated with all those candidates, they put up a list with a totally different name, and they ask their voters, “Vote for our candidates in the district but vote for this decoy party list.” Then that breaks the connection between the …
Remember, I said earlier that MMP compensates so that if a party wins a lot of the single-seat districts, it will win fewer seats from the party list in order to balance it out so that the parties that have won very few district seats will get their proportional share off the list. If the party’s putting up a decoy, then it wins all those districts where it had the most votes, and it also gets a fully-proportional share, not just the compensating share but a fully proportional share of its list, its decoy list. And this may sound just like, “Oh, how could this ever happen?” Well, it happened in Lesotho in 2007, I think was the year, and it happened in Albania once. And it’s been a well-known problem in Germany, even though it hasn’t actually happened.
Why hasn’t it actually happened in Germany or in New Zealand? It’s a really important question, right? This problem that everyone knows could happen has not happened. The reason it hasn’t happened, I think, is very strict regulation of parties, that the electoral commission would not let a party get away with this. I very much worry that if we were to adopt MMP for the U.S. House, all it would take would be, say … Let’s suppose a couple of Republican states just said, “Aha, let’s let the Republican Party run a decoy list. No one will ever will never tell anyone, we’ll just let it happen.”
They could then make sure that the Republican plus its decoy got all the seats, but most of the seats from that state, even when MMP working properly would have given them maybe 40% of the seats and left others for other parties. So, with our state-level decentralized electoral-administration process, I think MMP is kind of an accident waiting to happen. So I’m in this very uncomfortable position having to say, “Don’t adopt the system that I know, as a scientist of these things, is actually the best system.”
So, what does that leave us with? It leaves us with there’s STV or there’s list proportional representation, which could be an open list, as I said earlier, or it could be a closed list. Let me clarify: by closed list, we just mean that the voter votes for a list, does not select the candidate within the list. They accept the ranking that’s been made by the party. Closed lists are used for the MMP systems in New Zealand and Germany. They’re also used for the list proportional system of Israel, of Spain, of Portugal, and several other countries. Or you can have an open list where the voter is selecting a candidate within a party and the winners are … First of all, they’re both the same in terms of how the parties get seats. They get them proportional to how many votes the list got. Where they differ is in how is the list ranked. Again, our five-seat district, if the party got three of those seats, which three? They’ve maybe nominated five candidates, which three? In a closed-list system, the three that the party said were our top three. In an open-list system, the three with the highest vote totals as individuals within their list. So, under an open list, the voters are voting for typically a single candidate.
So, I think before we go into that, I should also say, I already indicated some of the problems that I see with proportional ranked choice voting, but another one that I didn’t mention is the administrative headache. It took me some contortions, maybe, to describe how it works and I was glossing over a lot of details. It’s a complicated system. Imagine if you’re the election administrator. Of course, it can be computerized. It’s not impossible to run it. But we have quite overstretched administrators as it is because we underfund our electoral processes. We’d have to greatly increase how much we set aside to run elections, which we should do anyway, of course. What’s more important to democracy than voting in elections, we should invest in it. But that’s another conversation.
STV requires a lot of voter education and a lot of administrator education to make it work properly. What I like about open-list PR is that the ballots and the administrative process and the education process are all pretty close to what we already have. You just tell the voter, “Vote for one candidate,” and the rest of the system goes from there. Of course, the ballot should indicate which candidates are in a party so you know who they’re allied with. And typically, an open-list ballot will have columns for party and then it will list the candidates in some order, but the order doesn’t govern the election process because what governs the final order of the candidates is who got the most votes.
So, the open-list system has a really nice, easy move from where we are to where we would want to be if we were to adopt proportional representation. And it works very well in districts of, like, five, six, seven members, magnitudes of five, six, seven, even three. So I would say that it’s one of the three best systems, but it’s probably the best system for implementation in the U.S. context for the reasons that I’ve mentioned.
David Nir:
I’m really intrigued. Could a state like New York, where I’m from, we have 26 seats in the House of Representatives, could the New York Legislature pass a law or maybe amend the state constitution to create, say, five districts with five members each, and odd man out, so one district would have six, let’s say? Is that legal? Is that constitutional? Is that something that could at least conceivably happen?
Matthew Shugart:
I would say yes. I don’t know each state’s constitution, but there’s certainly nothing in the federal constitution that would stop a state from doing this for its own state legislature. And there’s also nothing in the U.S. Constitution that would stop states from doing it for Congress, but there’s something in federal law that would stop it. Federal law currently mandates single-seat districts, but that governs the House, not state legislatures.
So, yeah, as far as I know, again, there could be state constitutional provisions in specific states that would have to be amended. I’m pretty sure that is the case because state constitutions are often pretty detailed. For example, the California Constitution does define our electoral system for basically all offices as being a single-seat district system. But in general, yeah, there’s certainly no reason why states could not go down this road. And in some states, maybe it could be done by just legislation. That’s a state-by-state variation where you’d have to consult an expert in state politics.
But yeah, you said it exactly right. If you’ve got 26 seats, you could make five-seat districts, plus one, one of them would be six seats, and you could implement this immediately.
David Nir:
Professor, I absolutely am fascinated by this whole conversation. I feel like we could go deeper into the weeds for many, many more hours, but I-
Matthew Shugart:
Oh, I promise you I could.
David Nir:
… think that we have to wrap up.
But before we let you go, would you let our listeners know where they can follow your work, and also any resources, including stuff that you’ve written, for folks who want to learn more about alternate electoral systems?
Matthew Shugart:
Well, sure. Since 2005, I’ve been writing a blog called Fruits and Votes. You can find it at fruitsandvotes.wordpress.com. And on there, I have a link to my academic work so that if somebody wants to actually look up something that I’ve published in book or in article form in a academic press, they can find it there. And then on the blog itself, much of the content is about electoral systems in other countries and also the US. So, of course, I’d be delighted if people wanted to go spend some time there.
You can also follow me on Twitter. If you just look me up on my name, it will come up. I hope people will be energized over electoral systems. This will just whet their appetite so they’ll want to go find more.
David Nir:
We have been talking with Matthew Shugart, distinguished professor emeritus in the Department of Political Science at the University of California, Davis.
Professor, thank you so much for taking the time to talk with us today.
Matthew Shugart:
Thank you very much for having me.
David Beard:
That’s all from us this week. Thanks to Matthew Shugart for joining us. The Downballot comes out every Thursday everywhere you listen to podcasts. You can reach out to us by emailing the [email protected]. If you haven’t already, please subscribe to The Downballot on Apple Podcasts and leave us a five-star rating and review. Thanks to our producer, Walter Einenkel, and editor, Trevor Jones. We’ll be back next week with a new episode.
