Charter School: Follow that vote!
Rose City Reform digs into the portion of single transferable vote that's most puzzling to voters: Vote transfers.
Can you feel it in the air?
Portlanders are getting ready to make a big decision.
And as voters ponder the Charter Commission’s ballot measure, many are asking: Exactly how would my vote be transferred?
This edition of Charter School is for you.
Who would have thought?
Apparently Rose City Reform isn’t geeky enough.
After my primer on single transferable vote (STV) – which to my astonishment became Rose City Reform’s most popular post ever – many readers contacted me to say: “Excuse me, but you left out the most important part. I want to know exactly how my vote will be transferred!”
Well, since you insist!
Let’s have a look at STV vote transfers.
Read my primer on single transferable vote.
When and how does your vote transfer?
Let’s recap the basics of STV, which is the voting method proposed for Portland City Council elections.
STV is a multi-winner form of ranked choice voting. It’s designed to allocate multiple seats in proportion to their respective share of the overall vote.
To win a seat, candidates have to clear a certain threshold, and that threshold depends on the number of seats to be filled. In a three-seat election, which would be the case in Portland’s proposed districts, the threshold would be 25%+1 of the vote.
So how can your vote move within the STV system?
There are two ways:
Scenario 1: Your first choice has been eliminated. Then the voting machine is going to move your entire vote to the next choice on your ballot.
Scenario 2: Your first choice has won a seat with more votes than they needed to get elected. Then the voting machine is going to splice off a portion of your vote and allocate it to the next choice on your ballot.
Today, we’ll break down how that happens.
Keep in mind, we’re speculating here.
There’s a reason I was vague about Portland’s proposed system.
We don’t actually know the specifics of how vote transfers would work in Portland. Single transferable vote can vary between jurisdictions, and when new voting methods get adopted, the real nitty-gritty gets worked out after adoption.
In Portland’s case, the Multnomah County Elections Office and the City of Portland would need to work together to iron out the final details, and City Council would ultimately need to sign off on the plan.
But we have a pretty good idea of the Charter Commission’s intent. And that’s because the commission’s proposed new charter has a section that, although fairly broad, tries to capture to how the voting method would work in practice.
So let’s take some liberties and make a prediction about how vote transfers might happen in Portland, based on the information we have at hand. And who better to do that than someone who has worked with three cities that have recently adopted STV for city elections?
Rose City Reform turned to Maria Perez, co-director of Democracy Rising, a nationwide organization that helps jurisdictions implement ranked choice voting.
Look to Albany and Palm Desert for cues, Perez says.
Perez’s guess, which is more educated than most, is that Portland would use the same system as Eastpointe, MI, and Albany and Palm Desert in California.
Democracy Rising worked with Eastpointe to implement its first STV election in 2019, and is currently working with Albany and Palm Desert to prepare for their first STV elections in November.
So how do vote transfers work in those cities?
If your top candidate is eliminated, your whole vote transfers to your next candidate of choice. But if your top candidate is elected with more votes than they need, then everyone who listed that candidate as their first choice gets a little slice of their vote moved to the next choice on their ballot.
“One of the interesting things about multi-winner ranked choice voting is this idea of surplus votes,” Perez says.
The surplus is the amount of extra votes that a candidate gets on top of what’s needed to get elected. And the size of the surplus directly corresponds to the portion of your vote that will be moved to your next choice, if your first choice wins with more votes than necessary.
Pro tip: Think of your vote as a dollar.
Perez says the easiest way to understand the surplus is to think of your vote as a dollar.
“Imagine forty people are pooling money together for a pizza,” she says.
Each person puts in $1. But the pizza only costs $20, so there’s $20 left over.
“What's going to happen with that leftover money? Every single person should get some change, right? That’s what happens with STV. And you get to put that change toward your next choice,” says Perez, before she offers another example:
Imagine a race where the threshold is a hundred votes, but one candidate gets 110 votes. One hundred gets her elected. The remaining ten make up the surplus. In this scenario, the candidate has 10% more votes than she needs.
The voting machine will now take 10%, the surplus percentage, from the vote of every voter who chose that candidate as their first choice, and apply it to the next choice on each voter's ballot.
Watch Democracy Rising’s educational video about STV.
Thankfully, the voting machine does the math.
Of course, in real-life elections, numbers are more complex.
Votes may be transferred as fractions, or percentages with multiple decimal points. How many decimal points, and whether you round percentages up or down, are some of those intricate details that get worked out after adoption.
In the old days, election workers counted STV ballots by hand. These days, sophisticated machines do the work.
If you don’t understand all of the inner workings of STV, don’t sweat it, Perez says.
"Voters need to know that there's an election coming up, what the ballot is going to look like, and how you fill it out. More engaged voters might want to dive deeper into the math of it, and they certainly can, but most voters will just want to make sure their vote matters and counts."
Learn more about STV.
FairVote: A brief history of proportional representation in the United States.
New York Times: Scholars Ask Congress to Scrap Winner-Take-All Political System.
VoteGuy: What is new (and not) about the proposed Portland charter?
“Politics in Question” (podcast): How does electoral reform happen?
If you have half an hour, watch this documentary about STV in Eastpointe, MI:
“The voting machine will now take 10%, the surplus percentage, from the vote of every voter who chose that candidate as their first choice, and apply it to the next choice on each voter's ballot.”
Not quite. If the “voting machine” takes away 10% of the candidate’s first choice votes, she’ll have 99 votes remaining, which is less than the threshold of 100.
Thanks for this tutorial, but I'm raising my hand here. Please call on me! Okay, thank you.
I have a question and, if I'm allowed, a comment, no a concern, well, anyway, a criticism.
First, my question.
My question has to do with the sequence for shifting votes in the proposed election scheme for council members. The sequence matters. When, not how, is a question that has yet to be even asked.
Let me explain.
It's not that we'll see votes transferred from eliminated candidates before we see surplus votes shifted from winning candidates, as some of the scenarios in the videos show. Most likely, we'll see votes shifted from winning candidates before we'll see votes shifted from losing candidates. And, if both were to occur at the same time, it will still leave votes for the rest of the still-viable candidates frozen in place, while the second choices of those who voted for the winning candidate, and, depending on the sequence, the second choice of voters who voted for the eliminated candidate, enter the tally. This seems to me to give more voting power to those who back winners (and in some cases, those who voted for the least popular candidate) in Round 2 of the tallying. Whose votes are counted when matters, especially when the total votes are declining and the 25%+1 threshold is declining as a result. Those who vote for Round 1 winners will have their votes count earlier and more often than those who vote for lagging candidates.
This is not just a math problem, it is likely to be a real world problem. More often than not, I think that Round 1 will produce at least one 25%+1 winner. Looking at recent primary election results in council races, that seems to be a common result, even in large fields. So, in that case, the transfer of fractional surplus votes will occur before any candidate is eliminated. If that's the case, then Round 2 will involve a tally in which only the fractionalized second choice votes of the winning candidate are counted, not the second choice votes of any other candidate. If the elimination of the lowest-vote getting candidate occurs at the same time as the transfer of partial votes from the winning candidate's ballot, that would seem anomalous, because how will you know if that candidate has the lowest vote count if you haven't yet tallied those surplus votes? So the sequence that is most likely to be followed will produce in many, if not most cases, a Round 2 tally in which the winning candidate's voters are given more clout than those who votes for lagging candidates.
The questions to be answered are not just how the fractionalizing and transferring of surplus votes will be accomplished, which you say is anyone's guess or more likely the guess of the advocate you quote, but in what sequence they will occur. From the winners first in Round 1, or from both the winners and the biggest losers at the same time. Both are problematic (for the candidates who have to compete in Round 2 or later Rounds), but the latter is more problematic than the former (because of how it treats the biggest loser).
So much for the wonky stuff. If I still have the floor, may I add a comment? Thank you.
I'm sure both the Fair Vote and Democracy Rising advocates mean well. But, are they paying attention to what is happening all around them when it comes to suspicions and mistrust of our elections? Yes, those suspicions are crazy and the mistrust is unwarranted. But that's what we're dealing with now. So how will it help to improve trust in our elections to move to a more complicated and less transparent vote counting system?
You quote Maria Perez of Democracy Rising as follows:
"Voters need to know that there's an election coming up, what the ballot is going to look like, and how you fill it out. More engaged voters might want to dive deeper into the math of it, and they certainly can, but most voters will just want to make sure their vote matters and counts."
No, it's not filling out the ballot that's an issue for suspicious and mistrustful voters, it's how what they fill out is counted. Deferring to computers won't assuage them, especially those engaged voters who take an interest in how their votes are counted and whose eyes will glaze over at the explanation. And then telling them that the votes of some candidates will be moved to other candidates is guaranteed to exacerbate their suspicions.
Okay. I've used up my time. There are a lot of other hands going up now. I hope you can get to their questions and concerns as well.