Plurality voting is the most common system in the U.S. A voter picks one candidate in each race and the candidate that receives the most votes wins. Then Maine enacted a new system called ranked choice voting for the November 2016 election. Now Alaska has joined Maine, and will use ranked choice voting for the first time this year as well as a new open primary system in which the top four candidates advance to the general election. Rep. Jonathan Kreiss-Tomkins (D) of Alaska is the first guest on the podcast and he discusses how Alaska came to adopt the system, and the challenges and costs it posed to the state’s election administrators. This discussion took place the day before Alaska’s open primary on June 12. The second guest is Ben Williams, a principal in NCSL’s elections and redistricting program and author, along with an advisory panel, of a new report on ranked choice voting that will be published in July. He discusses the national landscape for the new approach to voting and some of the information they discovered in surveying election administrators around the country.
Ed: Hello and welcome to “Our American States," a podcast from the National Conference of State Legislatures. This podcast is all about legislatures, the people in them, the policies, process, and politics that shape them. I am your host, Ed Smith.
JKT: Nobody was predicting or anticipating the sort of debut of ranked choice voting being in a sort of high political pressure environment that we are having with a special election and the New York Times, The Washington Post has written about it. What are we at? Forty-eight candidates. Santa Claus among them and Sarah Palin.
Ed: That was Representative Jonathan Kreiss-Tomkins of Alaska, who is joining me on this podcast to discuss ranked choice voting. Until 2016, nearly every state used a similar system to vote, a system called plurality voting. A voter picks one candidate in each race and the candidate that receives the most votes wins. Then Maine enacted a new system called ranked choice voting for the November 2016 election. Now Alaska has joined Maine and will use ranked choice voting for the first time this year as well as a new open primary system in which the top four candidates advance to the general election.
Kreiss-Tomkins talked about how Alaska came to adopt the system and the challenges and costs it posed to the states election administrators. Note that we talked the day before Alaska’s open primary on June 12.
My second guest is Ben Williams, a principal in NCSL’s elections and redistricting program and an author, along with an advisory panel, of a new report on ranked choice voting that will be published in July. He discussed the national landscape for the new approach to voting and some of the information he discovered in surveying election administrators around the country.
Here is our discussion, starting with Representative Kreiss-Tomkins.
Representative, welcome to the podcast.
JKT: Thanks so much for having me Ed.
Ed: Hey so we are going to talk about ranked choice voting and I wonder if you could start out describing how Alaska adopted ranked choice voting and what role you played in that?
JKT: Yeah. Alaska adopted ranked choice voting in 2020 by ballot initiative by sort of hair on one’s chinny, chin, chin margin of 51/49. The ballot initiative also included other components relating to transparency in disclosure of dark money in elections and a few other things, but probably the center piece of the initiative, the most substant of policy change was adoption of an open primary ranked choice voting system. We’re on the precipice of our first ever ranked choice voting administered election, which is actually a special election for our open congressional seat that was created with the passing of Congressman Young. That is going to happen this summer. And I think the only other thing I would add is Alaska, as with the state of Maine, has had a history of three-way races whether it’s because there has been a strong third party or independent candidate or sort of a disaffected Republican as well as a mainstream Republican. Just all sorts of unconventional mixes that have created spoiler effects and unwieldy the electoral dynamics. And so, I think to some extent in Alaska, that electoral history sort of created an orientation and an appetite for potentially creating this new election system, which arguably manifested with the passage of the initiative.
Ed: Well, I think those of us who are interested in elections always do watch Maine and Alaska because they are some of the more – often some of the more interesting races in the country because of that three-way dynamic. Let’s talk a little bit about money. That’s always something I think your fellow legislators are probably interested in knowing about. And I wonder in Alaska, what was your experience with the cost of switching to ranked choice voting?
JKT: Through the initiative, there was a little less than a million dollars that was effectively set aside for onetime adoption and implementation costs. There were tabulation machines that needed to be procured. Programming on the backend with the Division of Elections. A lot of public education and information related investments. I think there is also sort of a harder to calculate going on incalculable sort of costs just in terms of political resources and energy both within the Executive Branch in the Division of Elections, but also legislators and other sort of opinion and public leaders in the state to socialize the new concept and help people wrap their minds around it that has been sort of across the board.
Ed: How about the technology? You mentioned the tabulation machines. From what I understand, in some cases, machines can be reprogrammed with new software. In other cases, new equipment is needed. What was your experience there in Alaska?
JKT: My understanding is that DOE had to purchase one hundred plus new tabulators for RCV. However, I don’t know, you know, in other jurisdictions how transferable or adaptable existing technology is in making the conversion to RCV. I realize that that can sort of be a disincentivizing upfront costs to conversion for jurisdictions.
Ed: I’m just wondering from your perspective was it worth it? Is it a better system than our – than the traditional system?
JKT: I should disclose my colors I guess in that I was a proponent of the ballot initiative. A proponent of ranked choice voting. And I, you know, lent my name and sort of fed the rah, rah, rah pass the initiative drum a little bit although there were others who did vastly more than myself when the initiative was going to voters in the 2020 cycle. I guess the other hat that I’ve worn is that for the last six years I’ve chaired or co-chaired the committee in the Alaska House that has jurisdiction over elections and so we have held a couple of informational hearings on the subject of ranked choice voting. I think largely to sort of provide a forum for the division to get information out to the public, out to decisionmakers and legislators. I’ve sort of had different, I guess, relationships to the initiative, but as far as how it is working, I’m cautiously optimistic. Nobody was predicting or anticipating this sort of debut of rank choice voting being in a sort of high political pressure environment that we are having with a special election. And The New York Times and The Washington Post have written about it. What are we at – 48 candidates. Santa Claur among them and Sarah Palin.
Ed: Some very high-profile candidates for sure.
JKT: I think that’s notable in two ways. I mean certainly that’s an even statewide races in Alaska do not have you know 40 plus candidates on the ballot so that’s a really sort of edge case I guess for ranked choice voting to debut. Although I don’t really anticipate it to sort of challenge the system in any fundamental way. The second thing, though, is Division of Elections, you know, had a huge task in front of itself to make this transition in time for the general. Well, the primary and general election as normally scheduled and that was going to be you know a big lift in and of itself. And then like boom special election just kind of falls out of the sky and so Division of Elections is, you know, having to work triple time basically to pull this off.
So, we will find out shortly really how it performs on an administrative and technical level. Although in my sense is there haven’t been any warning bells that there is stuff falling through the cracks or there is sort of administrative mishaps. Basically, this sort of like lack of news I think is good news so far. Primary results for the special election are just around the corner. I think it’s sort of premature to have any assessment until we get through this special election cycle.
Ed: I would think you might have to see this system in operation for a few cycles before you’d want to draw any conclusions about its effect on, for example, having a moderating effect on the candidates, which sometimes is suggested as possible outcome of that in the same way that the so-called jungle primary in California is supposed to bring that about. Let me ask you about ballot design, particularly when you just told me you’ve got 48 candidates for that House seat. What kind of a challenge has that been for your election’s division?
JKT: Ballot design. I mean even in a non-rank choice voting, contact is incredibly important as we’ve seen in unfortunately some negative examples from across the Country. I mean there’s been a collaborative process with DOE on ballot design. And I think everybody seems happy. So, the organization behind the initiative, I think to their credit has not just said ah, we got the initiative passed, you know, peace out. We’ll see you down the road. But has maintained the presence and continued to invest in public education and also maintained a collaborative relationship with the Division of Elections including on questions like ballot design. That has proceeded copacetic. Everybody seems happy and the Division of Elections has a process whereby they share draft ballot design with parties for consultation. So far, so good. It hasn’t proven to be a bugaboo, at least up here.
Ed: Let me move on to something we talked about before we started, which is voter education. As I was saying, I think it is sometimes a little bit of a difficult thing to wrap your head around how it works. And I wonder what kind of approaches you folks are taking up there to just try to get people more information about what they will see on their ballot.
JKT: That’s been an all hands-on deck effort. A lot of public officials, myself included have tried to sort of chip in by using different platforms or bully pulpits to get the word out. The organization I referenced previously, which is now under the banner of Alaskans for Better Elections, has I think probably invested over seven figures in public education and outreach traveling to regional conferences to sports and cultural events to the state fair to you name it just to sort of be as anomy present as possible and familiarize Alaskans with what is coming. And I think the Division of Elections has done a good job given the circumstances of getting the word out and sort of having people be familiar with the approach. The Division of Elections ran a mock election under the RCV sort of paradigm with I think like for a moose or a caribou or a halibut or a salmon or etc. You know sort of did that dry run and again with all of the investment and outreach that is happening, I haven’t heard any warning bells. I haven’t had like stray conversations with constituents or people in my district you know who are just completely befuddled and feel like you know somebody is speaking Greek at them. But it’s all anecdata and you know I think the rubber will meet the road when we have the special election, general election and people actually have to fill out the ballot.
Ed: Well, it sounds like having an assertive nonprofit partner has been helpful if they are putting that kind of money and that kind of effort into educating people so that might be a good thing for people in other states to keep in mind that that is as you say an all hands-on deck effort. So, we’ve certainly heard an awful lot about auditing the elections in the last two years. How does auditing an election work here?
JKT: You know that’s come up with the Division of Elections in hearings in my committee actually and the Division doesn’t foresee any problems. I could be confusing issues a little bit because we were looking at actually kind of an overhaul of different audit processes and laws perhaps as part of the broader conversation that you alluded to in Alaska although ultimately nothing came of that. We almost got something across the line but didn’t quite. The division hasn’t attested to any problems on executability of it. So, you know full steam ahead.
Ed: I think there is a perception, particularly after the New York City mayor’s race, that using ranked choice voting slows down the results, and I wonder if you think that’s going to be the case? If you think it can be improved or maybe that was just a case with the New York City elections office.
JKT: It is so unfortunate to have New York City as a prior. In Alaska, we should know the results when all eligible absentee in question ballots are counted and that’s 15 days after the election, so that is a delay. In Alaska, we also have the law that I think is a prohibition on counting any absentee ballots prior to Election Day, i.e., sort of getting a head start. So, you basically preempt any lag period. I’m pretty certain that is currently the law in the books. We didn’t end up changing that although we were contemplating doing so. So, yes on the lag, but hopefully without the confusion and mayhem that existed in New York City. I think at the very least like having known/unknowns or sort of certainty about what the process and the timeline looks like so there is not such guesswork involved.
Ed: So, seeing what you have seen so far in Alaska when you talked about your election division, what kind of challenge should election administrators in other states expect if their state decided to go this way?
JKT: The biggest hurdle and these might be you know famous last words cause the special election or the general election could go sideways in ways that I’m not anticipating in this moment. The biggest hurdle I think is political, which also has sort of been the forum that I’ve been most involved in and that sort of wrapping one’s head around the concept and understanding or attempting to understand how it shifts the electoral incentive structure in ways that one might like or one might not like and different incumbent interests are going to react differently to that change in electoral incentives.
You know being halfway familiar with some other efforts around the country to adopt RCV, that is proving I think to be the highest hurdle in terms of initial adoption. The technical questions, the administration – I think that’s all secondary to sort of political will. I am wanting to think once RCV sort of has its day in the sun especially with an open primary function, which I think is really worth meaning that what we have in Alaska is pretty fundamentally different than what exists in Maine. Maine is open primary RCV structure. Has its day in the sun that it will be really hard to sort of argue with how it sort of shifts the discourse, the Overton Window if you will, on electability and who wins elections and if and how the median voter is well represented by their representatives. That sort of political puzzle and finding the political will and capital to get this across the line, I think is the biggest question mark.
Ed: I’m sure we will all be interested to see what happens in the open primary, the runoff for the congressional seat and then the general election. As we get ready to wrap up, I wonder what advice you would share with colleagues around the country about rank choice voting.
JKT: Two pieces of advice. One, do it, cause I think the county desperately needs--I mean innovation is kind of a wishy/washy word--but innovation in its election systems. And maybe the less wishy/washy, more substantive exhortation I would offer is that we really need election systems that represent in a representative sense the electorate and don’t create what I might describe as sort of distorted outcomes in terms of the median voter not being represented. Or that, a majority of the minority can constitute the slate of elected officials that might exist from a certain jurisdiction. That seems really important, and I would encourage colleagues around the country to sort of lean into that effort.
And then I think, No. 2, you know, if successful in getting this passed, that’s as the cliché goes only half the battle. To follow up, the outreach, the attention, detail and implementation is equally important. Arguably more important because if you fall flat on your face that has such a chilling effect. You know, unfortunately as we’ve kind of seen in New York City and the sort of negative reverberations that have occurred from that election cycle has caused a chilling effect on other people who might be interested in trying to adopt it. So really nailing the execution and sticking the landing super important and just keeping that in mind if trying to get this passed.
Ed: Representative thanks so much for taking the time to talk about this. This subject I find is fascinating and I will be watching the results in Alaska closely. You take care.
JKT: Thanks so much Ed.
Ed: I’ll be right back after this with Ben Williams from NCSL.
(TM): 18:22 Music/Advertisement
Ed: Ben thanks for coming on the show.
BW: Thanks for having me Ed.
Ed: So, Ben, earlier in the program I spoke with Representative Kreiss-Tomkins of Alaska and he explained how ranked choice voting has worked in that state. First question in my mind which is where else is it being used, and I’ve heard a story here and a story there in the news about it, but could you break that down for us and tell us is it becoming more popular?
BW: Ranked choice voting isn’t used widely in the United States, but it has been gaining steam lately. So, if you look at the website of the organization called Fair Vote, which is a ranked choice voting advocacy group, there are 53 cities and counties that use RCV for all voters in their election. There are two states that do it – Maine and Alaska obviously. Beyond that, there is a handful of states in the deep South that use ranked choice voting for specific kinds of voters. So, you know in the South there are states that have runoff elections. You are familiar with them--Louisiana, Alabama. You’ve heard of these elections before.
BW: In order to comply with federal law for their runoffs that they have, military voters who are overseas and overseas voters who still vote in state are sent rank choice voting ballots in case the race goes to a runoff. So, there are specific applications in some states where it is used. So, you may look at that 53 plus two plus a handful of specific applications and think that’s not a huge number right. But less than 10 years ago, there weren’t any states using ranked choice voting for all of their voters. So, it’s increasingly common and popular. Both Maine and Alaska voted for it via a citizen led ballot initiatives despite, one state being red at the presidential level and one state being blue. So, it has traction certainly.
Ed: As you mentioned this was not a big deal for a long time and in the time that I worked at NCSL, I do remember trying to build an animated graphic to show how ranked choice voting works. I think it might still be hanging around on the website somewhere, but it hadn’t been a big focus of your program. Now you are and I’m wondering what this new project is that you are working on.
BW: Sure Ed. So, we’re looking into several under-researched aspects of ranked choice voting. You know like all NCSL projects we aren’t taking a position on the merits of the policy or the wisdom of enacting it, so our work is really targeted at those minute details. Think what factors affect the cost of switching to ranked choice elections from a traditional plurality winner election or a runoff election like I mentioned earlier. Or when a state, city or county chooses to adopt ranked choice voting for some or all of its elections, how many pieces of equipment or types of software need to be replaced on average and stuff like that. So, we are writing a report and our report will include details on the policy side. The things that if you look at advocates, they talk about like does ranked choice voting lead to less political polarization. That’s a common claim that you will see from advocates. So, we summarize the existing research on that. Another one is does ranked choice voting leading to any unique election security issues. We address that as well. But the bulk of our research it is focused on detailed oriented administrative questions, not policy questions.
Ed: Let’s get into it a little bit then. What would – in that area, what would some of the key findings be? The key takeaways that legislators might want to think about.
BW: So obviously it’s focused on details, but I’ll go into a couple of questions and give you some of the answers that we found. You can imagine that it’s tough to get information on rank choice voting when only 50 or so cities and counties are using it. That’s a pretty small sample size for a policymaker to analyze. So, what we did is we sent a survey to every election office in the Untied States that administers ranked choice voting elections. While not all of them responded of course; a majority did. And one of the questions was what is the average cost of switching to rank choice voting. So, I think initial transition costs. Not the recurring long-term costs over a plurality election. We found that the average cost for a city or county to switch is around $154,759. Like that was the average on the nose. If you remove the outliers, it drops significantly to just under $40,000. The reason that is such a dramatic change is that the most expensive jurisdictions spent dramatically more money than others and that’s largely a factor of size. A lot of these municipalities that are switching to ranked choice voting are relatively small but there are a couple of really big cities that have adopted it as well. That’s one question.
Another one directly related to the question of costs was the average number of types of equipment that need to be changed to run an election using rank choice voting. So, you can think of things like when you go into a polling place, the machine that marks the ballots. A lot of places now they use optical scan ballots for elections, but you have the option instead of bubbling it out in some cities and states to fill out on a touchscreen and then it prints out an optical scan paper that you can review before you feed it into the feeder. That tabulation machine, which is like a big iPad, does it have the capability to present a ranked choice ballot. And then the machine that tabulates the ballots itself, can it tabulate using ranked choice voting. Even if it has the physical capability in terms of the size of the scanner, does the software in it? Can it be swapped out for something that can tabulate ranked choice voting? It’s not always the case that there’s interoperability here. We found that on average the tabulator was the most common piece of equipment that had to be switched out. That’s the weak point for jurisdictions thinking about switching to ranked choice voting. A majority of our respondents said they had to change that to be able to run an election using ranked choice voting. There are a few cities and counties that had to change their ballot design software or their ballot printer, but that was a minority. That was not as big of an impact.
Third question, keep in mind there were over 20, so I’m only giving the listeners a taste of three. Things like staff time. How much extra time and effort did an election office of staff put into things like ballot design. So, if you think about it, a ranked choice voting ballot is inherently more complicated than a traditional election ballot. So, in a typical, an American election, the candidate’s names are listed on a column or columns with a single bubble next to their name and sometimes it’s a touchscreen and you hit a square. But in general, this is how it works. You make a selection next to a name. In a ranked choice election, it’s not. That’s not the end of the process so you have to have bubbles for every candidate to have for their possibility of that candidate having a first, second, third, fourth, fifth choice vote and so on. Unless the jurisdiction limits the number of options that a voter is allowed to select. And so, as you add more candidates on the “X” axis thinking about like a graph on the other axis you are going to have more and more and more bubbles and it’s increasing exponentially. So, if you had a rank choice voting election with three candidates, there are nine bubbles on the ballot. If you had it with four, there would be sixteen. Five candidates, twenty-five. Eight candidates, sixty-four. You can imagine how quickly this becomes a ballot design issue.
Keeping all of that in mind, the thing that most voters typically don’t think about is how many different types of election ballots there are. So, in most states, we elect many different types of offices on a single ballot. You are all familiar with this. You’ve seen your member of Congress, your state Senator, your state Representative. If you live in a city, you have a city councilmember. You may have a school board or a water board. Some states elect judges. All of those are districted elections and those district boundaries don’t necessarily line up with one another. Oftentimes, each type of district is drawn with no regard to the other district boundaries that exist. And every single intersection of every single district line is a different type of ballot that a local election office has to send out to make sure that you aren’t necessarily voting for the wrong state representative. You are voting for the district you live in, but also your correct schoolboard member and your correct county judge.
Ed: I think most of us are able to get that ballot and fill it out and think well that was simple. But I’m beginning to appreciate the greater complexity of it from you Ben. Please go ahead.
BW: Yeah. No, it’s extremely complex. But not only is it complex, but you want it to be readable. I think every election official has this nightmare in their mind of being the election official in Palm Beach County, Florida, in 2000 where you had the poor ballot design that led to some voters being confused. And then there was a problem where the voters couldn’t punch the paper with the old punch card. That was the way I had voted for the very first time was with those punch cards. Butterfly ballots. There was a problem with hanging chads. Every election official wants it to be as simple and painless for the voter as possible. They really focus on a customer experience aspect. That means though if you are designing 4,000 different ballot types, every one of those 4,000 has to make sense and be clear. And so that’s a lot of legwork and rank choice voting and according to our survey results does complicate that process. So, it adds staff time.
Ed: Boy this is going to be report that legislators I think who really want to dig into the issue are going to be keen to read because there are a lot of complications as you say in any election, but the user interface and the computational challenges that ranked choice voting presents sound formidable. I know you are holding a preconference the day before NCSL’s upcoming August summit which is going to be held here in Denver. Can you tell folks who may be interested in attending what they will learn?
BW: For those who have never attended an NCSL Summit, it’s our biggest annual meeting. This year it will be August 1st thru 3rd as you just alluded to. And it’s in our hometown, Denver, Colorado. So, we are really excited to show folks our home city and where we live and why we love living here at the foot of the Rocky Mountains. But for most NCSL meetings of this size, there are a series of pre- or post-conferences that are held that are adjacent to the meeting and those could be invitation only or they could be open to the public. This is an open to the public meeting. We would like everyone who is interested to attend. You do not have to attend the Summit to attend the preconference although we certainly hope that you do. We think there is a lot of value in attending the NCSL Summit, but this is open to anyone and you don’t have to register for the Summit to attend our preconference on ranked choice voting. So, it’s going to be on Sunday, July 31st from 1 to 6 p.m. It will be at the Convention Center, which is the same facility where the Summit itself is going to be hosted. And we are going to be talking about what ranked choice voting is, whether it lives up to the promises advocates offer for why it should be adopted and whether it is possible for states, cities or counties to switch to it not withstanding the low cost that you just referenced. And we are going to ensure that folks have plenty of snacks and caffeine during the meeting itself. And to try to make it fun and more interesting, we are going to have a reception for that last hour from 5 to 6 and we are going to have hors d'oeuvres and we are going to have a bunch of different cocktails and mocktails for people to try and they can rank which one is their favorite. Easy rank choice voting Ed.
Ed: Very nice transition and I can testify that the snacks at NCSL meetings are definitely first rate. You mentioned that the inspiration for this came from conversations with legislators at NCSL meetings. What about legislators who are interested in other projects? How do they get their ideas in front of your team and possibility of you guys doing a deep dive?
BW: The best way is to just talk to us. You know we like to think that we keep the pulse on what our members are interested in and we certainly have put a lot of effort into tracking legislative trends and talking to as many people as we can. But the best way for someone to get an idea in front of us is to just reach out. You can either call on our numbers. All of our phone numbers are on the NCSL website. I’m sure that I can give you something to put in the podcast description Ed. You can also just email me. My email address – every email address at NCSL is just our first name.last name@NCSL.org. It’s a pretty easy nomenclature to remember. Anyone at NCSL would love to talk to a legislator or legislative staff about a policy area that they are interested in. And if we don’t have a lot of resources on it, we want to make sure that we get that for you because I guarantee if you reach out and ask, you are not the only person who is thinking about it. So, this is about being member oriented and member driven like everything else we do.
Ed: Well Ben thanks very much for taking a few minutes to talk with me about this emerging area of voting. It is really very interesting, and I have a feeling that we are going to be hearing a lot more about it. Take care.
BW: Thanks so much Ed.
Ed: And that concludes this episode of our podcast. We encourage you to review and rate NCSL podcasts on Apple podcasts, Google Play, Pocket Casts, Stitcher or Spotify. We also encourage you to check out our other podcasts: “Legislatures: The Inside Storey”and the special series “Building Democracy.”Thanks for listening.