102: Aaron Hamlin – Approval Voting

Play episode
Hosted by
Will Jarvis

In this episode, we are joined by Aaron Hamlin, the executive director of the nonprofit The Center for Election Science. Aaron has had great success implementing approval voting across the United States. This episode was cohosted by Lars Doucet.


William Jarvis 0:05
Hey folks, welcome to narratives. narratives is a podcast exploring the ways in which the world is better than in the past, the ways that is worse in the past towards a better, more definite vision of the future. I’m your host, William Jarvis. And I want to thank you for taking the time out of your day to listen to this episode. I hope you enjoy it. You can find show notes, transcripts and videos at narratives podcast.com.

Unknown Speaker 0:39
Well, Aaron, how are you doing this morning? Anyway, so do a great for the audience. We also have the esteemed Lars to stay with us to talk to Aaron, really appreciate having him on to kind of co host this this morning. Aaron, do you mind giving a brief bio and some of the big ideas you’re interested in?

Aaron Hamlin 0:58
Sure. So my kind of technical background is in the social sciences. I have a law degree. And the bulk of my professional career has been starting up nonprofits, I’ve done a couple of them. One was providing funding for pharmaceutical drug development to get new male contraceptives to market. And the one I’m working with now, which I also founded is the center production science. And there we study different ways to vote, like how you fill out your ballot and how that’s calculated. And we work on cooker way a kind of pool voting, which lets you pick as many cans as you want. And we work to with local communities, at the city level and going into the state level to get that implemented. And we’ve taken a very whether that has never been implemented before government elections, and so far kind of implemented in two cities working on the third and are getting ready to pivot into states.

Lars Doucet 2:05
Now when you say I’ve never been implementing governments before, do you mean American governments or any governments? But

Aaron Hamlin 2:11
technically, it was used in a variation in Greece about 100 years ago for a bit. But other than that, I’m not aware of any,

Lars Doucet 2:22
I won’t hold you to that level of.

Unknown Speaker 2:27
That’s great. Eric, can you talk about approval voting just for a bit what it is and why it’s superior to other methods?

Aaron Hamlin 2:34
Yeah, so I think So normally, when you go to vote, and I think when you go out on the street, and you asked me one, what is voting, they would tell you where you have this ballot in front of you, with some candidates, pick one of them, and a candidate with the most votes wins. And that is an example of voting. But that’s not the essence of what voting is. Voting is providing some expression of information. And then that expression of information is aggregated somehow with all the voters. And that aggregation turns into a result, where you have some measure of support for all the candidates. And the candidate with the most support is the one who is selected to hold that office. Like that’s what voting is. And you can do that in all kinds of ways. A lot of times people get fixated on the information element, like what information you’re putting on the ballot. So whether you’re picking one candidate, whether you can pick as many as you want, whether you’re ranking candidates, whether you’re scoring them on the scale, but you got to do something with that information, too. So, for instance, there’s all kinds of things you can do with ranking information. You can simulate pairwise comparisons between all the candidates, you can simulate sequential run offs, you can turn them into point values and add them up that way, all kinds of stuff. So with the approval voting is really among this, the simplest voting methods. So with approval voting, all you do is select as many candidates as you want, and the candidate with the most votes wins. And on this surface of it, it doesn’t sound like a whole lot. And really, there isn’t a whole lot to it in terms of like what you’re doing and adding up the results. But that small difference has a big impact. So, for instance, when we’re using our choose one method, if there are multiple candidates that you like, well, you’re screwed. I mean, you got to pick one of them. And as a consequence, those candidates will get a portion of the actual support that they actually deserve. And so the pool of voting, you’re not splitting your vote and so you can, these candidates that would have otherwise gotten diminished support, now get their accurate measure of support and you won’t have some extreme candidate coming in out of nowhere. Do you happen to not have their support divided, winning as a result? The other cool thing is like we think about independence and third parties, largely to the extent that they don’t get elected, and that we shouldn’t vote for them, because we don’t want it, they’re a vote away. Or approval voting just deals with that right off in a way that other voting methods don’t do well, including some alternative voting methods. So say there’s that independent, they’ve run, your wildest dreams have come true, you get someone who likes all the policy issues. Unfortunately, like, no one, like knows this person’s name. They don’t have millions of dollars in their war chest. But he loved their ideas. They come with great ideas, all the policy concerns you’d like they’ve gotten all of them. And normally, you’d be stuck with this dilemma is like, alright, well, my favorite candidate ever is now running, do I support them and like not have a say in the outcome and ultimately, like not being able to pick a real winner here. But with approval voting, you don’t have that dilemma. You just vote for them. And if you’re in this other dilemma, and thinking like, Okay, well, I also want to have a say in the outcome, if I don’t think this candidate can win, well, you can support a one of the front runners at the same time. But importantly, the candidate that you like, they get the support they deserve. And they can grow on that. And their ideas don’t get marginalized in the same way. So it has all these other important outcomes as well.

Lars Doucet 6:25
So you’re saying that even if you know, Mr. or Mrs. Perfect, but also Mr. or Mrs. Complete, obscure outlier? approval voting, even if they don’t win that election, you can still show that they had 10% 25% 30% support, right? Is that what you’re saying? And then they can take that and be like, see, I’m not a marginalized? Nobody I did way better than anyone expected. And then they can go on to be a front runner in the future. Is that what you’re saying?

Aaron Hamlin 6:51
That’s right. And so the voting method is not only used to determine the winner, but it’s also used in polling as well, because people can’t wait until election day. And they gotta have some kind of inkling in terms of who they think is going to win. And so they they match the polling to the actual voting method. And we I mean, you’d expect this to be true as well. But polling itself is used in a lot of important context. Not some of them are used to determine whether someone gets a debate or not. And if you have low polling, well, you don’t get on the debate stage, no one gets to hear your ideas, which is particularly important with presidential elections. But with approval voting, the polling would match. Because why wouldn’t it I mean, polling is used to make a prediction and you don’t use a different measure, then is actually going to be used on election day. So if you’re using approval voting on Election Day, letting people select as many cans as they want, then you’re going to get the polling the same way. And so if you’re able to get much more accurate reflection support, then you get that support. And if someone is using that criterion of polling to decide whether you get into debates, where now you get into debates in a way that you otherwise wouldn’t have people get to hear your ideas in a way they otherwise wouldn’t have. Are

Lars Doucet 8:11
you? Are you making the point that the mechanism of polling is essentially approval voting, but the way we actually vote is like first pass the post or whatever? So we’re tests we’re predicting how the elections kind of going to go in a way that doesn’t match actual mechanism for we elect? Or is there more nuance to it?

Aaron Hamlin 8:26
Well, right now, we use this choose one voting method for polling as well. Sometimes you have approval polls. But typically, they’re looking to figure out who’s going to win on election day. And to do that they match the voting method. But if the voting method was approval voting, they will use approval voting for the polling as well.

Lars Doucet 8:46
So you’re saying we would get better polling as well, if we had approval, elections and approval polling.

Aaron Hamlin 8:53
If you had approval voting, you’d also have a pool voting in terms of polling as well. And it would be used in the same context as trying to pick the winner. But you would also get the benefit of approval voting just as a far better job than our choose one method and many other methods. In fact, in terms of accurately measuring candidates support, even candidates who do not win, they all get a good accurate measure of support. This can also be shown empirically, which we’ve done and published on. But you get that advantage when you’re using Pro voting for the voting method, as you also get this accurate measure of support during the polling phase as well.

Lars Doucet 9:30
So New York, I believe, was his last New York mayoral election where like, Yang ran against Adams and stuff they used I believe, rank choice voting correct me if I’m wrong, that’s correct. Yeah. And so can you talk about the differences? That’s another often hyped example, like, well, if we just did rank choice voting, that would be better. And apparently they got that in New York. So can you compare and contrast? first past the post rank choice and approval voting, and just give us your thoughts on those?

Aaron Hamlin 9:56
Sure. So I guess first I’ll explain what rank choice voting is. So earlier I said the voting method has all these parts. One of them is that expression element like what you put on the ballot? Well, with rank choice voting, big surprise, your ranking, which, to me is a little weird in terms of naming because in actuality, there’s an entire class of voting methods that involves ranking. The traditional name for rank choice voting is instant runoff voting, and some other countries that call it by different names like the alternate vote. So it’s kind of weird that they call it rank choice voting, because like, there’s actually a whole class of voting methods that involve ranking in any case. So you go and you rank your your candidates. In many cases, because the ranking process can take up a lot of space on the ballot, it’s often truncated to like your top three or top five. So you can kind of be losing information right off the bat. And so when you’re when you’re ranking these candidates, when we first look at is the first choice votes. In terms of the calculation after everyone is done ranking. Now you look to see if any candidate has more than half of the first choice votes. If yes, you’ve got a winner. If no, you look to the candidate with the fewest first choice votes, you eliminate that candidate, you look at all those ballots next choice preference, you now treat that as a first choice vote. And now you go and you add everything up again, you say okay, now, among all the remaining ballots, does anyone have more than half the first race votes, if yes, you’ve got a winner. If no, you keep repeating that process, with the remaining balance, which also may not be all the initial balance, because through this process, not everyone ranks or all the candidates down the line, or maybe they ran out of space on the ballot. So really, you’re just talking about the remaining balance here. So you have a kind of simulated sequential runoff is kind of what’s going on here. And that’s rank choice voting. Now, one thing to keep in mind here is that virtually any voting method is going to be better than our choose one method. So you do hear a lot of folks say like, well, like it is better than our choose one method, well, anything is better than our choose one method you can pick, there are many, many voting methods out there, you can pick any of them in any of them will do better than our choose one voting method.

And in terms of how it does in other contexts, it can do some weird things. So for instance, like in our choose one method, we think of vote splitting, among our top preference, which is our only preference in our choose when voting method, we can kind of have that same kind of thing going on with rank choice voting as well. So if you have similar candidates running, and one candidate has their votes, but Well, if that candidate has the fewest first choice votes, merely through votes playing, they can get knocked out of the race early on. So that’s one issue, you could prematurely eliminate a good winner, who would otherwise be a good winner. The other component is it doesn’t do a perfectly good job of measuring the support of all the candidates, particularly ones who don’t win. It does do better than or choose one method, but again, like not by a whole lot. And also, pretty much any other voting method would also do better than or choose one method. And so what you wind up with as a result is a lot of independent third parties get behind victorious voting, I would say a bit erroneously thinking that they’re going to get more support people can choose her on his favorite. Well, in fact, there are certain conditions when folks are kind of punished by choosing their favorite is first which we seen in actual elections, such as the 2009 Burlington, Vermont election. But in addition to that, even when folks even when it doesn’t kind of go against them. The other issue is that it, it doesn’t really show their support, because if someone ranks, say, an independent third party a second, or that independent third party has already eliminated those other rankings, showing support for that third party never appear in the results. And again, this is something that we’ve seen empirically, looking at polling using experimental designs, but now we’re seeing raytrace voting being used in actual elections. And you can see this play out in real life. So for example, Maine, has been using retrace voting and it has a real independent streak. So you would think, well, if anybody is going to be treating third parties and independents nice, it’s going to be this folks in Maine. But when you look at the 2020 election, and you look at the Presidential results there which you just rank choice voting, they do no better than they do in other years in terms of when you look at the Libertarian Party, and the Green Party, and a very terse voting was gonna be able to kind of pave the way for them, you would have seen it there. But you saw no difference in terms of results.

Lars Doucet 15:16
There’s also the there’s also the question of like, the cognitive load of actually building, right, like, I remember with the New York City campaign, you saw these people saying it’s like, okay, so like, for like people voting, it’s like, okay, you can vote Yang, but rank him here, do this. And like everyone arguing about who gets the top spot and stuff like and approval voting, it’s just like, if I want to vote for Smith, and Johnson, and, you know, lastly, I just stamp all three boxes, right? And I’m done. I either would be happy with them winning the office or not. And that’s the only decision, right? Is that all there is to it? Is there any like pros and cons to it being that simple.

Aaron Hamlin 15:53
So that is a big component. So if, for example, if you have a large list of candidates, it’s way easier to just go through and say, These are the candidates I like, versus trying to rank them all. I mean, you can go through the process in your head, like just imagine a list of like 25 or 30 movies, and find all of what you’ve seen, and rank them versus go through and see like, okay, what are the ones I really like? Like those two processes are very different, and one is way quicker than the other. And for as an example here, and this kind of goes at some of the limitations of rank choice voting is that you can also, it can also kind of stumble a bit in, if you’re trying to use something like open primaries, where everyone can run at once and say, like the top two or top four, go to the next round. As an example. In Alaska for, for instance, they have a an RCV election, where there are 48 candidates on the ballot. And like, and, and because our CD has challenges of trying to nominate multiple candidates to the next round, what you’re seeing there is that they’re forced to just use choose one voting in that first round. And so you can only imagine the amount of votes voting that’s going on there.

Unknown Speaker 17:16
I’m curious, Aaron, you’re super smart guy, you’ve got a lot of options. Why did you choose approval voting? Was there like an aha moment that you know, was it a third party candidate that didn’t get any recognition? Just the intransigence of American politics, like what kind of pushed you to work on this problem.

Aaron Hamlin 17:37
So when I was in grad school, I was in the student group for healthcare reform. And I was out with a bunch of folks in the student group. And we were all this was 2008. So we were all talking about who we were going to vote for. And as we were going around the table, I was a bit surprised, because all my friends were talking about voting for people who I knew didn’t align with their, their interests. And also like, they didn’t actually support the health care policy that we were in the student group for. So it was kind of giving them some flack about that. And they were saying like, well, you know, if I support this candidate who really thinks the way that I do, they’re not going to win. And so I was kind of annoyed. And I thought, like, well, like, either I give my friends a hard time, maybe I wind up with fewer friends in grad school, because it’ll be what’s put up with me, or looking at it from the terms of like, okay, well, there’s some parameter here going on that shaping their behavior that’s causing them to act this way, which really, they were seeing explicitly saying, like, hey, like, I don’t want to throw my vote away. And so that’s what got me interested in voting methods. And I just kind of got obsessed with them from there. But also, I spoke with other folks. And really was trying to look at this parliament in a fundamental way. And when you’re when you’re looking at a problem like this, you have to ask really just kind of basic questions. One of those basic questions is what makes a voting method good in the first place? And when you and asking that question, I think there are really three big criteria that come up. One is winner selection, which is one of the voting methods, core jobs can’t do that, well, then you might want to start looking elsewhere. The other component is how accurately it measures candidates support. So it’s a voting method has more than a job of just selecting the winner. You want to make sure that I mean, election is about also being able to measure at this point, different ideas. And if a voting method doesn’t measure support accurately, then these other ideas that other dialogue happens in a in a in a poor way. And then the third one is practical. lobby. So you don’t want something that’s needlessly complex, or more complex and has to be given the value that it offers. You want it to be easy to implement. To the extent that it can be an approval voting, this really checks all those boxes really well. And that was the main rationale for going in that direction. Of course, the challenge was, it hadn’t been used for government elections, it was used a lot in academia, at some academic, academic organizations. And so we had to show proof of concept first. So we did that like so in the very, very tail end of 2017, practically 2018, in the beginning, was when we got our initial funding. And within a year of our initial funding, not only do we hire staff at the organization at this inflection science, but we also got approval voting implemented in its first US city, within a year of our initial funding. And so so we had to show that proof of concept, then replicate that and scale that to show that this wasn’t some one off thing. And so we’ve been doing that at the city level, we started at Fargo, North Dakota, then moved to St. Louis, Missouri, working with the community there. It’ll now be on the ballot this November in Seattle, Washington. And looks, its polling at 70%. There, so it’s looking good there. And then we’re at the moment, looking at pivoting to statewide campaigns where we’re not just we won’t just be affecting local offices, like we have with the cities we worked with before. But we will be affecting statewide positions, federal positions, Senate seats, US House seats, as well as presidential elections. And we do this all through ballot initiatives, because there’s a inherent conflict of interest when you ask the people who are elected by the old voting method to change the way that they get elected in the future.

Unknown Speaker 22:06
Aaron, I love this, how you went from zero to one, you know, one year implement something that really has not been done before? In a say that it says super cool. What was your, you know, approach and figuring out, okay, I found this new method, I think this will be like very effective, this will really help the world to implementation. What was that process? Like? Was it like reading literature trying to figure out how to do this? Did you have previous experience which informed like valve initiatives? Those are, that’s the way to do it, versus lobbying or something like that? Like, how did that process kind of unfold?

Aaron Hamlin 22:37
As well, in terms of like the thinking process, we had, the the initial group that we had was pretty technical with our initial board. So like, we had engineers met petitions, our advisory board was really amazing. Steven brands, political science professor at NYU, one of the independent developers of approval voting as on our advisory board. So all that was very critical in terms of being able to help our thinking, in terms of the pathway, it helped that we weren’t the so we had what we call a second mover advantage. So with raytrace voting, it had a multiple decades headstart in the space. And so we don’t have to start from scratch. We can just kind of look around and see what was done before, see what was successful before and see what wasn’t. And we just don’t do the things that weren’t very successful. And then we do the things that worked. And so we just did that. And ballot initiatives seemed like the clear path there.

Lars Doucet 23:46
And what were the things that worked and didn’t work? I think,

Aaron Hamlin 23:49
one, there’s also kind of an order effect, too. So the bail initiative component was definitely one that worked well, particularly at the city level. In terms of what didn’t work, it lobbying seemed to be one of those that at least wasn’t going to work at the stage that we’re at. So

Lars Doucet 24:11
sorry to interrupt. What do you specifically mean by lobbying? Do you mean lobbying? Personal politicians?

Aaron Hamlin 24:16
Yeah, yeah. So. So like, normally, when you get something on the on the ballot, you get some super excited folks locally. And you get and they go and they collect a bunch of signatures and it gets on the ballot, people vote yes or no on it. If you lobby, people who are in office, they can put on what’s called a referendum which is a little bit different than initiative with a referendum. Like for instance, a council would just vote on it and they would put it up on the ballot. So you skip that whole signature phase. And you already come in with buying from the from the local government. So you can you can do Do that approach, but it’s not as easy when you’re dealing with something that’s a bit newer. And because approval voting hadn’t been implemented in as many places, that route is a bit more challenging. So that’s something that isn’t something that says that we can’t do in the future. It’s just something that is going to be harder in the earlier phases.

Lars Doucet 25:23
And you talked spoken before about how, specifically when you’re trying to change the voting system? You I think you just said that another reason that personal lobbying of individual politicians is ineffective is because you’re basically asking them to extinguish the voting system by which they themselves came to power.

Aaron Hamlin 25:38
That’s right. Yeah. So can you talk about that a little bit more? Yeah, people just don’t like to make it harder on themselves to get reelected. And if they got elected by certain way, then they’re like, Okay, well, this is I change that I don’t want to hurt my chances. And so they tend to be oppositional. And so we, I mean, we, we talk with, and we get the local advocates that we work with, work with the local government, but we don’t rely on them saying yes or no to us. I mean, we just say like, Okay, well, gave you a heads up, it gave me the opportunity. If you say no, here, like, we’re not going to stop, this is going to pass. So you’re just going to have to live with it and have better elections.

Lars Doucet 26:21
I think one of the reasons we’re so interested in talking to you about your experience is that the conventional, numerous wisdom about America that we’ve just gotten gridlock in every level, and nobody can accomplish anything, change will never happen. And so it’s very interesting and heartening to see you do this like was well said, like zero to one experience where you had this idea you wanted to move to get it done, you figured out how to get it done. And you got it done in a couple of places. And I was wondering if you could speak to the whole, like, gridlock and numerous just aura descending upon us and how you like, like, Are you optimistic moving forward? Like, do you think it’s just about? Yeah, just just like, how did you pierce through that veil? Or were you just like, oh, just do what works?

Aaron Hamlin 27:07
Well, I mean, I think a little bit back to like that conversation I had with the dinner when I was in grad school with my classmates, I, it would have been really easy. And I think it’s easy for a lot of us to say like, like, these folks are making terrible decisions. I’m definitely these people were acting in their interest. They were just tactically voting. But particularly when you’re looking at behavior at the population level, it becomes, I mean, you can blame everybody, you can blame millions and millions of people. You could do that. But I think a more practical and useful approach is to think about what are the parameters that are shaping their behavior. Again, if you’re in a game, or you’re playing some kind of game, you have certain rules, and that those rules determine how what your strategies are, how you behave, how you interact with others. And it’s really hard to just go through and just convince millions of people to behave differently, particularly when the rules suggest that they shouldn’t behave that way. So if the rules are really the problem, you got to change the rules. And that’s how you’re able to shape behavior and get people to begin to operate in their own interests. So when we think about gridlock, it’s like, well, yeah, sure, like people who don’t have the there is there are very few opportunities when we can’t be ignored. And one of those rare opportunities is when we like, when when we vote, like the with the outcome that we get there, they can’t just say like, Ah, well, we’re gonna ignore that one. Like they have to that’s binding like they they have to switch seats, if the results tell them to switch seats. And it’s a real shame that in that very rare instance, where we are able to have power, that the tool that we use is so bad, like we just, we are armed with this really terrible tool. And what we’re saying is, well, you deserve to have agency like this is the one time where we can create that realignment where the people in government where their interests should align with yours. So why not give people a tool that actually allows that to happen, rather than this crappy tool that we have now let’s choose one voting method. So all we’re saying is like, Okay, well, you deserve a real tool that gives you agency. Here’s that tool. And that’s really the big game changer in terms of allowing people to avoid this gridlock that we’re stuck in right now. Because if if I mean, if even if it’s like your own party, if if they’re not, like listening to you like, what are they gonna? Do? They don’t have to listen to you like, where are you gonna go? Like, what are you going to do? And that’s really the situation that we’re in now. Whereas if we have approval voting, like, nothing gets some competition, like, if somebody comes in there with some good ideas, they don’t have to worry about the same kind of dilemma that they’ve had in the past where it’s like, okay, well, nobody’s gonna vote for me. Because I’m a new person I have. It’s kind of name recognition is like, no, no, you can get traction, you can get support. And you can hold these folks accountable, that aren’t doing their jobs. And that’s something that we just have not had before.

Lars Doucet 30:41
So how do you how do you get it done? Walk us through, you’ve done this in a couple of cities. How did he get it done? How much does it cost boots on the ground? What’s What’s the formula?

Aaron Hamlin 30:50
So I think like our initial couple of cities, were a bit. folks coming to us. So the first city in Fargo, North Dakota, is marred by crappy elections, which is not unique to Fargo. It’s all just all over the place at the local level. But all of it like every level, but also the local level. And so the people were were waiting, like with under 30% of the vote and getting elected to the commission. And the commission, I guess, was getting embarrassed by this. And so they created a task force, the task force, went out looked at different voting methods, one of the folks in the task force reached out to us learned about approval voting, and brought it back to the task force and said, like, hey, like taskforce is on board, this is really easy to implement, seemed to address the vote splitting issues that they were having recommended to the commission. And commission just sat on their hands, they didn’t do anything. And so the the member from the task force, got everyone they knew, gather signatures, got on the ballot and passed it. And that was the first implementation. And they worked with us along that process. The second in St. Louis, I had done a podcast called 80,000 hours. And one of the folks in St. Louis, had heard that and was talking about a cool voting. And they were initially looking at raytrace voting, which the machines in St. Louis couldn’t handle. And they needed something that wasn’t going to be expensive for the city. And so they reached out. And, again, similar. They’re like, there was a story where the former mayor and St. Louis, you may have seen this, like, there were a lot of racial issues there. And there were protests. And this was also that during that same time, or a couple, like a couple, outside their mansion, like pointed their guns and a bunch of protesters is that same place. And so the former mayor doxed protesters during that, during that time, and just gave out their personal information, like where they lived and stuff. And like, that’s their, that’s their mayor, like, that’s their former mayor. And that person got their office, see, through a bunch of those putting, there was a bunch of those putting in the black community and the progressive community. And as a result, there wasn’t as much of a split among her ideology. And so she wound up winning. And when So, when approval voting was getting momentum, it got on the ballot, and it passed there are passed by 68% in Fargo passed by 63%. The moment have passed that mayor who docks are their citizens names who were protesting. All of a sudden decided, as an incumbent that it makes sense for her to run anymore. And so she saw the writing on the wall, so to speak, and she decided not to run. And then the in the next election, you didn’t have that expiring and you had, like a more progressive community. You saw the leaders under approval voting, once it was being implemented. Those tended to be more progressive voices, which is common in big cities. And also, it didn’t neglect the black community as well. And so St. Louis selected its first black woman Mayor after approval voting was was implemented. And, and Seattle is a bit different. And that, again, the vote spreading all over the place like every city, and Seattle came about as a result of our chapter system. So we have a nationwide chapter system, which our director of campaigns has set up. So folks in every city in the country, every state in the country, although we right now are particularly focused on balancing of cities and states. So the leaders in that campaign came out of our chapter program. So we worked with them. We gave him some initial reason sources that they needed to be able to help them move along, give them advice about different components of the election. And as a result, like they’re ready to run a campaign and they, right now, they, the city of Seattle is in the process of verifying all their signatures. And so, so it’s a little bit different in Seattle. And we’ve pivoted more towards that chapter approach.

Unknown Speaker 35:26
So it sounds like, one thing that’s really important is to find kind of a first wedge, you know, where’s your Fargo, North Dakota, where you can be successful? Like, if someone reaches out to you like, it’s friendly territory, it’s fairly friendly. And you can kind of push it through and then kind of use that momentum to go on to bigger and bigger kind of places to implement given policy, is that a fair kind of strategy

Aaron Hamlin 35:49
approach? Yeah, I mean, you’re basically thinking about an adoption curve, you go where the lower hanging fruit is, first, where or like, where things are more cost effective. And we’re likely to get your early wins, and then you build up over there. So for example, like, lobbying is a harder approach, you tend to need a bit more momentum already, to have more wins, so to speak. And so like, that’s something for like the later part of the adoption curve, whereas right now we’re focusing on where are those wins are a bit easier, a bit more cost effective?

Lars Doucet 36:30
And so on the cost effective aspect, it’s like, Can you can you put a price on a ballot initiative to get approval voting past? Like, you know, fundraising is a big part of any of these organizations, you know, can you break down like your cost structure for us in terms of, you know, money and, and human power and just boots on the ground trying to get it done?

Aaron Hamlin 36:48
Yeah. So there’s the direct costs for campaigns, it’s also like, for CDs, like to be able to have the infrastructure that we have, which we are underfunded as an organization for concrete for what we do. We, in our first year, ces started off with a budget of like 700,000, we now have a budget of around 2 million a year. But for what we’re doing is, we could use a whole lot more. And that’s for like, being able to do all the research that we do, looking at the results of all these elections, using experimental design to compare different voting methods, being able to do the the outreach, getting people to learn about these different voting methods. And in terms of the actual campaigns, kind of the Euro stick is like, at the city level, like maybe $1 or so per person in the population, because and one of the big parts that eat up a lot of costs is the signature gathering, which is just not cheap. And then actually running the campaign, having a campaign manager here, like the, the Local Group is the one that hires that person. And then at the state level, things can get even more expensive, because signatures, like you need a whole lot of them, even in proportion to the, to the population itself. So, I mean, you can see, some campaigns cost way more than that. So, for example, like I think the Alaska campaign was like four or $5 per person in the population, for their campaign. But you can do it more efficient than that. And when you have a voting method that’s simpler, you don’t have as high of educational costs as you do with some of these other approaches. So you get a good bang for your buck when you’re doing your pro voting campaign. But nonetheless, it’s still a non trivial price tag, but it is an efficient price tag for what you get out of it.

Lars Doucet 38:59
So you’re talking about signature gathering, right? And so if you’re coming in as an outsider, you’re inevitably gonna have to go through the signature gathering. So given that, I wish I could remember which jurisdiction I suppose but there was a scandal recently in some jurisdiction, were like three or four or five of the candidates, I forget how many but it was, like several of them have all used the same, like organization to go harvest signatures for them. And they like all got invalidated at once. Because like, this organization had basically like not, like not done it right, and had like, made up a bunch of signatures, like use the same pen in the same handwriting, like five lines in a row. It was like super obvious. So like, what are best practices for signature gathering? Like doing it? Right? You know, that you’ve just, I mean, is it just hard work and how do you how do you do it efficiently and effectively?

Aaron Hamlin 39:50
I mean, the, you just have to do your research and identify your really good suit together and firm at the at the scale that you’re talking about. Like you you can’t do this with But volunteers, it’s just the the amount of signatures you get as far to events. So you just have to get a reputable firm. But even then, like, even with reputable firms, like we’ve seen firms, not on our end, like fortunately, like we’ve haven’t had any issues, but we have seen other campaigns with firms and Tim good. And it’s just super easy to make mistakes. And some of these, and as you can imagine, there are a lot of folks who don’t want to see the voting method change. And so they have their eye on you at all times ready to stop you in your tracks in terms of the campaign. So and practice, you just have to really kind of double check things and make sure that you have a firm with a good track record.

Lars Doucet 40:53
But you see, you really are hiring professionals with expertise and signature, you’re not sending out a bunch of teenagers just wait outside the Dairy Queen. And just like ask people to sign up, like, you’ve got professionals who know where the population centers are and know how to like pitch your thing, like how much do you interface with them, that you have to train them on your messaging and everything like like, how does that work when you when you when you when you when you sign that kind of a contract was with outside firm. And

Aaron Hamlin 41:20
what we’ve done with at the city level so far is the local groups have been the ones to work with their signature gathering firms, and they could work with them on messaging, they communicate with the firm themselves to make sure that they have the messaging Correct. Of course, like you’re playing a little bit of, of telephone, and like when you have tons of folks are in the center gathering, like maybe the message isn’t communicated 100% to degree that you like every time, but they do a good job overall. And so it’s a lot of just communicating with the firm, being clear on what it is. And, and, and again, like it goes back to like, when you’re dealing with something as simple as approval voting. It keeps up fidelity a lot higher than it can be in other situations, other types of balance sheets,

Lars Doucet 42:11
I was gonna say it helps to have a simple direct message, right? Yeah, if you have some complicated thing you’re trying to get across, not only to your firm, getting signatures, but to the person on the street trying to convince someone to Hey, will you sign up for the last we’ll doodle super on policy? You know, that might be a hard sell.

Aaron Hamlin 42:29
Yeah, that’s trade art. Because right choice voting has gotten a little bit of a head start our largest issue, I think, with a lot of these folks asking, like, is this rank choice voting? And it’s like, no, this is way easier. And actually, it even works better.

Lars Doucet 42:44
Okay. So on that angle, there’s another, I don’t want to go too deep into all the different kinds of voting methods. But there is one that’s kind of salient in our milieu, which is not ranked choice voting or approval voting, but it came out of Glenn Wiles book, radical markets, quadratic voting. Are you familiar with what that is? And do you have any opinions about that you’d like to share?

Aaron Hamlin 43:03
So with quadratic voting, and this is not one that I focus on as much, but the and you correct me if I’m off it, because I don’t focus on it as much. The gist is that based on your time and an area, that you can allocate your, the weight of your vote to different offices, how you see fit. And it’s like, technically, you can nest the voting method within that. And that can be kind of like a higher level. But like, overall, like, I don’t know, like, I I suspect that there may be some, like, maybe 14th amendment stuff with like, Equal Protection Clause, because you’re maybe giving people who live there, longer, disproportionate weight, and perhaps their issues with that. But the, it does kind of go and look at something like okay, well, what are we ultimately trying to address? And how complex is it like, and what are the what’s the complexity cost that we’re that we’re paying for this improvement? And are there easier ways to improve that have a lower complexity cost?

Lars Doucet 44:11
Right, I think I agree. I mean, I think we might be talking about different forms of quadratic voting, and when I’m thinking about is the one where you have, say, 100 points to allocate, but when you vote for someone and you assign a point, they get the square root of the point you get. So if you to give someone two weighted points you have to spend for your points. And I have my own thoughts about it, but I’d be interested in yours.

Aaron Hamlin 44:33
Yeah, and my recollection, and again, this is not a voting method that I pay as much attention to because it kind of falls out outside the kind of like the traditional classes of voting methods. My initial understanding was that the amount of weight that you had could change based on how long you lived in a particular jurisdiction. Oh, but if If we kind of like single that out, and it does avoid some of those other issues that could come into play, such as someone who lives there longer getting greater weight overall. Putting that aside, like the essence, I guess it’s like, okay, well, what are the elections that I really care about? And I’m only going to focus on those, for instance,

Lars Doucet 45:22
right. I mean, I guess I guess the the critique, typically that I’ve heard is that it’s like, Okay, now it’s even more of a math problem than rank choice voting. You know? So,

Aaron Hamlin 45:30
I mean, let’s come back to that complexity cost, right.

Lars Doucet 45:33
I mean, um, maybe we’ll have to learn why alone and ask him his opinion on it. So you can give a more robust defense, but so you are very much in favor of these very simple voting methods. And I think it’s really interesting how you get out there, and you’ve just kind of engineered this way to, like, get these things done. So you say you’ve had two cities, you’ve done it in so far, you’ve got an active campaign and a third, like, what’s the end goal here? Just approval voting on everything throughout America?

Aaron Hamlin 46:02
Or? Yeah, I mean, I think we could do that. And there are roughly 20 or so states that use ballot initiatives, we will just exhaust all those states working with folks across the country. And once those are exhausted, we’ll do lobbying. And we’re working on picking up the remainder.

Unknown Speaker 46:27
Very cool. Very cool. Aaron, I have a quick question just just on fundraising. What is your approach? Ben, do you just reach out to wealthy people that might be interested? Is it more grassroots? What’s been effective? What has been less effective? Yeah.

Aaron Hamlin 46:43
So in the very beginning, we had no money, which is very challenging. So we would do things like crowdfunding campaigns, all my friends got tired of hearing from me during that time. And we would do like, we got a couple good projects out of it, like one was, we got an explainer video on approval voting held a bunch of like animated fruit called plantsville, in a place called plantsville. So it’s a nice way to explain how a pool voting worked. And then we did another one, that provided funding to do experimental design, using polling through a large polling firm, that allowed us to empirically compare different voting methods and pair that to a control measure, which allowed us to see whether pool voting, empirically measured support for candidates Well, compared to other voting methods, which it was able to do. And we looked at that in the 2016 election. And then in 2017, we really just got lucky with our networking. And I was able to get in front of some folks with resources for our initial funding. And our initial funding came from platypi, which is a group that looks at highly effective impact interventions. And we were able to get funding repeatedly from from that organization, getting wins, that also helped our kind of brand awareness and what we were doing. So that feeds off of itself. So that helps a lot of smaller and mid level individual voters. And then from there as our winds increase, and as other folks give Biden, like they tell their other friends, and when you’re dealing with folks who have that more wealth, all those folks tend to talk with each other. So that also helps to get introductions to other folks.

Lars Doucet 48:43
You know, they often talk about, you know, it’s not what you know, it’s who you know, but I’ve heard an even better version, which is it’s not about who you know, it’s about who knows you. Right? And it’s very interesting that you talked about how your first case with Fargo, North Dakota came to you. So presumably, you didn’t necessarily know these people in Fargo, North Dakota, but you made yourself visible to them. Do you have any idea how you make yourself visible to them in that early stage?

Aaron Hamlin 49:08
We so early on when we had no money. We put a bunch of articles on our website, analyzing talking about different voting methods. And our Google Search score was evidently high enough to come up onto a search result. And that was virtually it. Fortunately, like, when you will Fortunately, for us at the time when you search for voting methods, not a whole lot comes up like there’s not as much competition. And so we had enough search engine optimization, I guess to make it so that we were at least visible on the internet for folks who are searching for our space.

Unknown Speaker 49:52
Very cool, very cool. So something like you need to get out there. You need to make yourself known to the world for a given policy so people can find you in Some way, Loris, did you have any other question about approval voting? I’ve got a fun one.

Lars Doucet 50:05
No, go for it go for you. Take us out here. Awesome. Well, Erin,

Unknown Speaker 50:09
I’ve really been enjoying reading your work kind of prepping for this. I did have one question. I had my bike stolen a couple of months ago, what’s the easiest and safest way to prevent my new bike from getting stolen.

Aaron Hamlin 50:21
So I have a personal website, which is Eric hamblin.com, where I put out all my essays on voting theory, and other work I’ve done in contraceptive technology, and also stuff on like, nonprofits and setting and as well as technical aspects of giving, quickly dealing with US tax law. But oddly, by far, like the most popular essay I have, is none of those things. It’s an essay I wrote on bike security. As a hobby, one of the things I do is a do lock picking, and which is just purchase locks, you have lockpick set, you understand the mechanics how the locks works, and essentially the way that a key works, open the lock, you can also use bypass stuff where you don’t deal with the the key. But in any case, like I’m in this kind of community that thinks about that kind of thing. And I also am a big fan of bikes, I’m big fan of transit without cars. And there are two big impediments for people riding their bike. One is lack of infrastructure, top one. So people can feel like they can read safely. And then the other is people want to ride quickly on a nice bike with the fear of getting it stolen. So I did a put an essay together. And it talks about layered security. So the idea of if you hit so you don’t have any one failure point. So the idea is like, if someone just happens to take that one lock, then you’re locked, then your bike is gone forever. So avoiding that kind of situation. So the approach that I put out there, using this kind of idea of layered security just looks at one like thinking about where your bike is making sure it’s and secure area. And then I look at the types of tools that are diffuses. The number one effective tool for a thief is an angle grinder, which is a spinning circular blade that tears through virtually anything. And looking at locks that put up as much resistance to that as possible in terms of how long it can resist an angle grinder because an angle with it when you’re dealing with an angle grinder, it’s only a matter of time. So how much time do you have? And there are other different types of end, you have to kind of think about how much is your bike worth, but also like the opportunity cost of like, okay, like, do I want to have to go and find another way home affects the how you got out that day, or going and finding a new bike and setting everything up. So other steps that you could take looking at, there are some devices like one is called like Boomerang has a built in accelerometer that’ll go to your phone. I use it personally. But it’s it works. Okay. It’s that perfect technology by any means. And then that’ll make it so that you can least go back to your bike, if someone’s messing with it. It also has a GPS, so you can go and find it. And then there are other things you can do such as you go to bike index.org to register your bike, because like if your bike is stolen, and the police find it, or you find it easier to prove than tours. And so if you register your bike ahead of time, that’s when you need to do that. And so like also like police, like they don’t do a great job with recovering bikes. But to the extent that they do get your bike. And you don’t know about it? Well, they’re going to look on bike index set org. And if your bike is not on there, well, you’re never going to see your bike again. So ended up at if none of that works. There’s also insurance too, which is company code below insurance, that insurance bikes in the US.

Unknown Speaker 54:12
Cool. Cool. Well, Aaron, thank you so much for coming on the show. Where can people find you? Where should people where should we send people.

Aaron Hamlin 54:20
So for making sure that you have that agency so you can actually elect people who represent your interests. You go to this reflection sites at election science.org. You can sign up for our newsletter, get all kinds of great stuff in terms of learning about what we’re doing learning about events that we put on. And you can join our chapter program. We have a discord that folks can join, to get involved at the local level, learn about other people who are also interested in this and being able to share our work to one of the obstacles that we face at the moment is that A lot of people don’t know what approval voting is. So to the extent that you go out and share with other people, our work and tell others about approval voting, that will really help things go along. And then also, if you’re interested in the bike stuff, go to Erin hamlin.com have a bunch of essays on there. Also, if you are thinking about giving to this reflection, scientists have a lot of essays on technical aspects of giving with the US tax law. And you can learn about how to give appreciated stock using bitter advice funds, and make your gift very efficient to this sort of action science and invest in the responsive and giving yourself agency so that you have a more responsive government.

Unknown Speaker 55:47
I love that. Well, thank you. I really appreciate it. Thank you. Thank you. Thank you. Special thanks to our sponsor, does market analysis for the support. Bismarck analysis creates the Bismarck brief, a newsletter about intelligence grade analysis of key industries, organizations, and live players. You can subscribe to Bismarck brief and brief dot Bismarck analysis.com. Thanks for listening. We’ll be back next week with a new episode of narratives. Special thanks to Donovan Dorrance, our audio editor. You can check out documents work in music at Donovan dorrance.com

Join the discussion

More from this show