“Well, you are the only candidate that came to my door, so you got my vote.” This is what a very nice older lady told me on a particularly hot summer day while I was out canvassing, before I even made my campaign pitch. On one hand, it was an invigorating example of the power of retail politics. But it also highlighted the biggest problem with the new city council election system.

With so many candidates running for so many seats, it was effectively impossible for candidates to break through or be taken down by media scrutiny. Almost everyone involved in local politics this year saw how the parking ticket scandal killed Carmen Rubio’s campaign and gave Keith Wilson his opening. We will probably never know how many similar scandals might exist among the hundred-odd candidates running for city council because there is nowhere near enough local journalists to do that level of digging. Instead, everything was lost in a sea of overwhelming choices.

The biggest predictor of who won the city council races seemed not to be policy positions or ideology but name identification and existing political connections. Looking at how candidates' votes were reassigned to second or third choices, there was surprisingly little ideological or issue consistency.

The current city council has very poor ratings, which explains why 3/5th of the entire council all lost the mayor's race, but Dan Ryan was still able to win the third place seat with just 23% of the vote in District 2. Similarly, in 2016, Steve Novick lost his city council re-election bid with just 44.65% of the vote, with his worst performance in the city coming from what is now District 3. However, this year, he managed to win a seat in District 3, finally surpassing 25% support in the 20th round of vote reassigning. Lorretta Smith lost her last two runs for Portland City Council but won this year with 21% of the vote, thanks to so many exhausted ballots making that good enough. Candace Avalos won less than 9% in her 2020 run for City Council (17,966 votes) but was part of the charter reform commission. There she pushed for a voting system that would let someone win without majority support. She then ran under this system, securing her seat this time with just 10,478 votes. Among the other eight winners who were not previous candidates for city council, five were political employees for important local or state officials. 

https://projects.oregonlive.com/maps/eudaly-win/

Fortunately, some of this will be partially fixed next time, since only half the seats will be up for re-election every two years. Even so, merely 50 candidates running can still overwhelm journalists and voters. A slightly higher requirement for becoming a candidate could help some, but six seats open at the same time will always generate a large number of candidates.

Recommendation: Replace the $75 filing requirement with a 100-signature filing requirement. This is enough to reduce the number of candidates modestly but still easily doable for someone willing to put in the work.

Portland’s Small Donor Program is an incoherent lie that incentivizes cheating

Repeatedly, both city officials and candidates said the small donor program would match small donor dollars nine to one, but this was simply not factually true. An incredibly tiny percent of donations were actually matched nine to one, while a huge share of small donations were never matched at all. This happened in part because the program did not have enough money for all of the candidates, so they changed how it worked. But when a government agency changes how a government program works, the agency NEEDS to be honest with voters about it. The agency can’t just keep the messaging consistent if the messaging is a deception.

In practice, the small donor program was not a matching program at all but a terribly designed donor bounty program. There were three donor total tiers a candidate had to hit (250, 750, 1250) and they would effectively get $40,000 for each threshold they hit. So if by the end a campaign had only 749 donors, they would get $40,000 from the program, but if you could convince just one more voter to give you $5, the city would reward the campaign with a total of $80,000. As you can imagine, having tiers where that last few small donations are worth $40,000 created a horrible incentive to boost donation numbers by any means necessary. This terrible incentive structure is why so many scandals about donation swapping, donation buying, and donation harvesting have come to light. This is how you end up with candidates who focus entirely on getting donors instead of voters, which is why candidates ended up spending huge amounts getting almost no votes.

The number of donation abuses was greatly enhanced by the city auditor's office refusal to regulate the program. After the donation swapping news came to light, I called the auditor's office to ask what they were doing about it; they told me they didn't consider it their job to audit city money they gave out. This meant there was every incentive was to exploit the system and ask for forgiveness after. Expect far more abuses in the future if the possible penalty remains less than the reward.

The program design also undermined a major point of the new charter: local representation. It treated donations from outside of a district the same as donations from inside the district, as long as the donor lived somewhere in Portland. This goes against the entire reason for having districts and a small donor program. The purpose of districts is that only people in those neighborhoods get to decide what they want, while the purpose of small donor programs is that regular people can boost the people running to represent THEM. 750 small donors in District 4 shouldn’t be able to use city money to completely overwhelm candidates they don’t like in District 1, but that is what can happen. 

Finally, the program’s design stacked the deck for incumbents and the politically connected in multiple ways. First, the three tiers gave a massive financial boost to campaigns who sometimes only had a nominal amount of more donors. Second, to qualify, a campaign needed 250 donors by August 27th -- way before most people are paying attention to local elections. If a campaign hit that threshold in early September, they got nothing from the city, but for campaigns that did qualify by August 27th the city would keep rewarding donations made until October 25th. There is no real justification for two separate dates, except to benefit incumbents and people who can run really long campaigns. If the city is able to process and verify donations in October, it should be able to do that for every campaign. 

I’m a supporter of matching fund programs in practice, but Portland has one of the worst possible implementations imaginable. It is so poorly designed I honestly wonder if it is the product of overt favoritism.

Recommendation: The small donor program needs a full redesign and should never ever ever use tiers that strongly incentivize rule breaking. Never should the city give one campaign twice as much money as another simply because they had a few more than an arbitrary threshold. It should be an ACTUAL donation matching program, like voters were promised. There should be a modest threshold to qualify and after that each donation is actually matched after it is made, until the campaign reaches a maximum amount from the city. Only donations from inside the district should count for matching. A flat four-to-one match would fit the current budget better but in a much fairer, honest, and less abuse-prone way. The deadline for campaigns or donations to qualify should be the same date for everyone in late October. In addition, the city needs the department in charge of the program to actually engage in oversight to stop abuses of it.

Portland managed to create the worst of all possible worlds voting system

Three countries use Single Transferable Voting (Australia, Malta, Ireland). The most important thing their ballot designs have in common is that you get to rank all the candidates by simply writing the number next to their name. Being able to rank all the candidates might not seem like a big deal, but it is hugely consequential. It is what gives STV its only real benefits. Despite the complexity, voters can vote their principles without engaging in strategic voting or worrying about their vote being wasted. In these places, political organizations often provide a full ranking of all the candidates based primarily on how much they would advance the group's priorities. 

Portland adopted a fake version of STV by only letting you rank six candidates. This means it has all the complexity of a proper STV voting system but with the added dysfunction of needing to engage in strategic voting as you see in first-past-the-post elections. With our fake STV system, you not only risk the possibility of your vote being thrown out if none of your six choices make it to the final four, but actually making the outcome worse since your vote changed how winners are calculated.

While most voters probably didn’t realize how much limiting ranking distorted the system, it was clear as a candidate. I spoke with numerous groups in endorsement meetings, and the issue that kept coming up was how to game out strategically voting to avoid having a vote wasted. The result was many groups still choosing to endorse who they thought had the best shot of winning instead of candidates with the best policy platforms. Most groups endorsed just a few candidates as part of complex game theory instead of providing simple best-to-worst lists like you would expect with true STV.

Our first election results already make it clear our truncated voting options have distorted the outcome. Normally in real STV, all three winners can claim victory by eventually getting 25% plus one. In this election, five of the 12 council members won with less than 25% due to so many exhausted ballots. In these five seats, the number of exhausted ballots significantly exceeded the margin of victory over the fourth place finisher.

For many, being able to vote AGAINST an incumbent you dislike is even more important for democratic accountability than being able to vote for the candidate you like the best. This is the revealed preference of voters, since in places with non-partisan primaries to whittle down the field more people usually vote in the general election round than the primary.  This right existed under Portland’s old run-off election system for city council, since you were able to vote for the final two candidates in the run off. It also exists in a true STV, which follows international best practices of letting you rank every candidate. If you fully rank your ballot, you can be assured you always vote against an incumbent you oppose, no matter how many rounds the voting count goes to. 

This right does not exist under Portland's unique bastardization of STV, creating a massive legitimacy problem. Deborah Scroggin, who ran the city’s election system, misled the public about this during a Reddit AMA. She said, “If you ranked 6 candidates and all of them are preliminarily defeated, then your ballot was counted as intended and the ballot has done its work (it's exhausted!). This is just like any other election method - if the way you vote doesn't work out to support winner(s) - you still voted and your support for those candidates was counted.” But this is not true. Under the old system, if your preferred candidate lost in the primary, you still had your right to vote against the candidate you most disliked in the run off. Since we merged the primary and general, if your ballot is exhausted you effectively lose your right to vote in the final round, something that is not supposed to happen in true STV. 

The big problem will come in two years when the voting system creates a huge advantage for the three incumbents in each district. Many voters will like two of their council members but want to hold the third one accountable for something unforgivable they did. With proper STV, a voter could vote for this outcome by simply ranking every candidate above the one council member they opposed. With Portland’s fake STV the strategy becomes way more complicated. This voter should rank the two council members they like among their 6 but that leaves just 4 slots to try to guess who stands the best chance of unseating the council member they dislike. This will of course be complicated by the fact other voters may feel this way about the other two council members and choose their own candidates accordingly. Voters will often guess wrong about who is the most likely to come in third in such a complicated dynamic, letting an incumbent sneak by with well less than 25% because the bad system deprived voters of their right to vote in the final round. For example, Eric Zimmerman managed to win the third seat in District 4 with just 18% after all the rounds of reassignment. 

It is likely unpopular incumbents might even recruit/prop up a bunch of spoilers to divide the vote. This would be useless under actual STV, but it's another major potential problem of Portland's Fake STV. I would go so far as to argue the city is in violation of the charter, since the system we have is so distorted from the core principles of STV, it may not legally be considered Single Transferable Voting. 

Recommendation: We should adopt actual STV like we were promised when we voted for charter reform by letting voters rank all the candidates. Other countries have done it for decades. It is by far the best, simplest, and most logical solution. If we don’t do that then we need some primary or second round to bring the number of candidates down to the number of ranking slots minus one. As it is currently designed Portland’s system will undermine basic democratic accountability and is legally not what the charter requires. We have a system arguably worse than what we had before. We should not do STV unless we are going to do true STV.

Overall, I found the campaign encouraging in many ways. I was happy to see so many people involved in local politics and so many organizations trying to make the best of trying to cut through the confusion and data overload. I’m also extremely grateful we got rid of the commission system, which had some of the worst political incentives for any government design imaginable. 

My greatest fear coming out of the campaign is that we replaced one huge accountability problem (how do you judge a commissioner when they are part legislator and part a random bureau director?), with an even bigger accountability problem. 

We will see if the new council works to fix any of these glaring issues, but rarely do politicians vote against a system which lets them win, no matter how broken.