The latest lunacy to seize Minneapolis – a city that seizes lunacy like Lindsay Lohan seizes appletinis – is “Instant Runoff Voting”. An idea husbanded by the Green Party, it basically brings to elections the sober reflection of the Lotto with the insight of a high school popularity contest.
Here’s how IRV works (and I’ll note in advance that my disdain for the idea will make this description seem about as dismissive as I intend it to):
- For each race, the voter fills in their their choices for candidates, in descending order. For example, in my most recent Mayoral race, I’d have put a “1” by Randy Kelly, a “2” by the written-in name of my dog, a “3” by Mary Jane Reagan (if she were running – and she’s always running for something), a “4” by the written-in name of “Idi Amin”, a “5” next to DFL-endorsed Chris Coleman, and a “6” next to the endorsed Green candidate (whose name I’m fuzzy on – I think it was Moonbeam Birkenstock, but don’t quote me on that).
- The computers would count everyone’s first choices.
- If nobody gets a majority, then the computer takes the second choices.
- And so on.
- And so forth.
- And I’m actually very fuzzy on how it works at this point.
- It’s confusing. Really.
- And we come out with a winner.
What I think is most interesting, by the way, is that the same people who kvetch and barber about electronic balloting and how Diebold is run by Republicans are the same people, in many cases, who want to turn our democracy (whatever that means in Minneapolis) over, entirely, to a ranked sorting algorithm.
Seem…opaque to you?
We have a rare moment of bipartisan consensus. On a St. Paul politics discussion board, a commenter with long-standing ties to the DFL writes:
…remember who is most empowered by the IRV system, it is those voters who chose candidates who get dropped from the ballot first. So, if you would agree that in most municipal elections the candidates who finish last are not some brilliant issue oriented candidate who just didn’t have the resources to be heard, but rather a candidate with lots and lots of tin foil on their head. And, remember those voters who voted for that candidate then get their second choice. You really start to see how the IRV system is
just about as far away from Jefferson’s goal of having an educated and informed electorate as possible.In fact as someone who has been running campaigns for over thirty years in Saint Paul, it doesn’t take a lot to figure out the purpose of the IRV system, it is to achieve by random chance and confusion what one can’t do in an open an informed election — elect Minor Party candidates.
While I agree with the writer about as often as I french-kiss Marisa Tomei, I think that’s a great point; IRV makes the fringe, nutcase bloc inordinately powerful. Combine that with a slick, coordinated message that “democracy is broken” and a faith-based plea to make trite protest-voting a way of life, and you will have…
…a city even more strangled by fringe politics than Minneapolis already is.
The essence of the new campaigns once IRV happens is vote Green, it doesn’t matter. Don’t worry, our candidate won’t get elected, your second choice will win, so it doesn’t matter that you voted for us first. Send a protest vote! So, if you can be convinced that it really doesn’t matter who you vote for and it won’t hurt you to throw away a vote, and if they can get enough people comfortable with that, then they get to win a few seats.
Democracy and governance is no longer of some value, its just a parlor game.
They want to bring the system to Saint Paul, by the way – and the proposal is getting some traction in high places.
[IRV proponents] want to tell you that it will be cheaper to not hold primaries. What they don’t mention is that we will need all new computers for every polling place to read IRV ballots and they don’t mention that the school board won’t be on an IRV ballot so, you might need two separate ballots and machines, or at least you will have one contest where you IRV and one where you don’t
IRV, that will make the average voter happy…
Another thing about IRV; I measure Usability for a living; it involves answerign the questions “Who is using a product (software, hardware, toy, shopping process, whatever), “What are they trying to accomplish, how important is it to them” and so on. I (and a class I was teaching at the time) did a usability evaluation on the infamous Butterfly Ballot, for example – and founds scads of things that could make it incrementally more difficult to use more correctly. Fact is, when things are designed by people whose first interest (or competence) is not designing things to be usable by people whose primary goal in life isn’t using that thing, you’re going to have problems.
And if American election authorities can’t design a punch card ballot – a book with a pinhole – to be clear and usable, what makes anyone think that they can design a rank-based ballot that will be any clearer? And before you answer that question, remember – there a small but solid number of people out there who earn a very living wage, myself included, precisely because industry realizes exactly how dismally bad most people are at making things usable.
And, having done campaigns, if you ask voters why they don’t vote in primaries, the normal answer is because they don’t know who all of the candidates are and that they don’t pay attention to the election until the field is narrowed, and of course that will never happen in IRV. So, you are as likely to turn off voters as stimulate them.
But, remember it isn’t the general public that we are concerned about, or what the average voter will be most comfortable with, the real issue is How In the Heck Can We Come Up With a Scheme To Elect A Green Candidate????
Exactly. If you can’t win ’em over with your platform and candidates, baffle ’em.
Mitch: While I fully agree, you didn’t quite represent what IRV does. If you vote for a candidate in the #1 position, that vote is recorded as long as the candidate is still ‘in the race’. It doesn’t automatically skip to the next choice until the candidate is dropped because of a low vote total. So in your example, IRV would have counted Randy Kelly to the end of the counting. Now, if you had voted #1 for your dog, #2 for Amin, #3 for Kelly…and after the first count Comrad Coleman didn’t have enough to win, then on the second count, your #1 drops and your second choice is included.
Here’s my rub: EVERY year the Greenie-boob-losers run an IRV poll at their State Fair booth, and EVERY year they count the ballots to show how great IRV is. Except…EVERY year they end the counting with LESS voters than they started with! Gasp…we’ve disenfranchised people!
You know why? Because voters WON’T rank people after the first or second choices, and ultimately are not counted when the final count is made. They start with 500 voters…end with 450. The Independence/Reform/ Jesse-loon party did the SAME thing in chosing their mascot. And the SAME thing happened, some voters were lost in the process.
Well, I fully disagree with you, and I think you didn’t get what IRV does. Among other things, it allows you to vote for your first choice — whether it’s the DFL, Republicans, Independents, Greens, Constitution or your choice of independent nominee — and rate your subsequent choices according to your preference, with the effect of your expressed preferences always mattering.
That said, I’ll bow to your professional expertise that the design of a form to express that — that ranking your own choices really can matter — is, at the very least, difficult.
Cui bono?
The American winner-takes-all system leads to a government dominated by two parties, each trying to get 50% +1 votes on election day, not a bad recipe for mush and definitely not bad for the richest and militarily strongest country in the world.
I think it would be an interesting experiment to pay people not to vote. Give them a choice of voting or a $10 bill when they show up at the polling place. Then we’ll see what people really think their vote is worth.
So….you’ve french-kised Marisa Tomei at least once? You lucky devil.
Man, I don’t know if I have the energy/time to debate this one right now but I’m definitely a supporter of IRV. I went out canvasing, went to city council meetings, got signatures to get the referendum, placed signs, etc.
I’ve got a couple posts about it from when it was an open issue:
http://rphaedrus.livejournal.com/5445.html (mainly a plea for people to support it)
http://rphaedrus.livejournal.com/16963.html (explaining how it works)
For the record, the “lost votes” thing is a non issue – if you don’t rank all the way down (which there is no reason to do) and all your candidates got eliminated, then none of your candidates got elected – exactly the same as it currently works but you got to make your second choices. If you only want to vote for one person, that’s just fine. You’ll add to their total and increase their chance of winning – pretty much the same as now.
Also for the record, in places where forms of this is used (Cambridge MA, Ireland, etc.), it does not get the fringe candidates elected. I know that in Cambridge, most of the time (I think it was between 70 and 80%), the first count does it and someone wins with a clear majority. I think its probable that the other locations see similar results. In the other cases, it is pretty much always someone who got a fairly strong showing on the first round of counting. It is pretty much impossible for a fringe candidate to win – not enough people would give them rankings for them to avoid elimination let alone get a majority. No candidate that did not get at least 50%+1 votes can win. There are various opinions on how to deal with NO candidate getting at least 50%+1 votes. They tend towards reverting to taking the highest vote getter or having a re-election. It doesn’t have a high chance of happening and if it does, isn’t really any worse than our current system.
To my mind, the fundamental purpose of all of this is to remove gambling from elections. If a lesser known candidate has my “perfect platform” I have to either vote for them and “waste my vote” or I have to gamble that a semi-OK candidate with more support may actually be able to win so I vote for them even if I don’t like them. This is a situation that our current voting system makes necessary and while many people may be OK with it, I am not. It isn’t a necessary aspect of representative democracy and I really think that the “gambling” aspect of voting – while it may be entertaining – should be eliminated if possible.
The reason its called instant RUNOFF voting is that it functions exactly like a long series of primaries. Would you vote differently if instead of wondering if Hatch was going to beat Pawlenty (or vice versa), everyone except for the one candidate getting the least number of votes was going to move on to the next round? Wouldn’t you wait to vote for the “lesser of two evils” until there was no one else left? (And obviously, one hopes that someone who isn’t the lesser of two evils makes it to the final choice) Have you never “held your nose and voted” for someone you didn’t really like even though there was a lesser known candidate you would have much preferred to vote for – but the odds say you had to gamble rather than vote? The current system creates that situation. Its a good system, it has worked for a couple hundred years, but I believe we can do better. I believe that living in the era of mass media and massive communication, the degree of various forms of propaganda (advertising, polling, web logs, “news”, etc.) has significantly increased the gambling involved.
I AM glad we’re doing this on a small scale in various places first to shake out any issues. There are going to be education and adoption issues that will need to be field tested and figured out. I don’t mind being a beta tester on this but I really do believe it is a significant step forward for representative democracy.
Incidentally, due to the fact that this is also being applied to some of the multi-seat elections, I think you’re going to start seeing some conservatives actually make it onto the city boards – right now, since over half the city is DFL, they get pretty much every seat. I’m pretty sure that there are enough conservatives in the city that they should get some representation.
Regarding the electronic ballot point – I DO have some concern about it. For the record though, “we’re” not against electronic voting machines or electronic vote counting. What we’re against is paperless non-verifiable voting. The ATM model works just fine.
Here’s my ideal voting system:
You select your choices on your nifty electric voting machine. There are two additional options at the end of the ballot. This may not be the best wording, but its the idea:
[ ] Print a receipt
Printing a receipt prints a copy of how you voted and has a randomly generated unique ID.
It may also be a good idea to print a copy of the ballot for you to visually verify and enter into the ballot box in case a hand count becomes necessary. I’m somewhat ambivalent to that but could understand why it might be wanted.
In any case, back to the votes – the records are obviously put into some sort of database.
In my system, the unique ID is stored along with the vote records. A “verified copy” of that database is created and made publically accessible. The folks responsible for making sure that ballots are counted would need to do some random data checking to help confirm that an accurate copy was made.
Using your unqiue ID, you can call an automated government phone system OR a government website to verify that your vote is what you think it is. You can also download the entire voting record (on larger scale voting, this may have bandwidth concerns but in the era of YouTube, its hard to imagine it would be too bad) so you can run your own counting software against it.
Additionally, both the counting algorithm and possibly even the actual software used should be available to anyone who wants it.
There’s a pretty good description of how IRV works right here. It’s been used in Australia for some time, and has seemed to work out well enough.
In the short run, in Minneapolis, the only party to get an advantage will be the Greens — which doesn’t much matter, as the Minneapolis DFL really is indistinguishably loony, and will get the majority on the Council anyway. In the medium run, though, if a sane party — or a challenger — can get enough interest and support, they/he/she could actually end up getting elected, without the voters who honestly prefer them/him/her having to worry about “wasting” their vote on an apparent underdog.
It never ceases to amaze me that so many DFLers go along with this wicked idea. We already have a big problem with people not understanding or marking a ballot correctly, it would be insane to make them hugely more complicated. IRV is basically a way to disenfrancise the old and the poor.
RickDFL-
My pay-people-to-not-vote scheme is the better way to disenfranchise the old and the poor!
No, when the vote is lost…its lost. Stop ignoring the fact that this will happen.
But even beyond that, voters will not be treated equally. Let’s say two Greenie loons vote for their choice for the Soil and Conservation Board. One Greenie votes for just the Greenie and nobody else. The other Greenie votes for the Greenie, and ranks the rest. Once the first choices are gone, then one voter is given a SECOND vote…on their second choice. In other words, some voters get to vote two or more times.
Now, in Chicago, where the dead can vote…that might not be a problem.
There’s also this notion that voters will actually take the time to rank choices. Anyone who believes this has never paid attention to what actually HAPPENS in an election. Once you start at the top of the ballot, by the time you get to the bottom you lose about 10% of the voters. They just don’t care about voting for city races, or soil and water conservation directors or unchallenged judges.
Making voting MORE complicated will SUPPRESS the vote…not open it up.
Dave, no, in IRV, all voters whose first choice doesn’t immediately win get to have their vote counted two or more times, even if they haven’t done more than selected their top pick.
In IRV, the only voters whose votes aren’t counted more than once are the folks who specify only one choice, and whose one choice is the least popular first choice.
In your example, the voter who only voted for the “Greenie” will certainly have his or her vote counted at least a second time — in Minneapolis, the Green Party candidate will certainly not have the least first-place votes of all the candidates; that’ll be one of the fringe, self-nominated candidates who always get on the ballot, like Mark Koscielski.
There’s nothing wrong with thinking that IRV is a bad idea; that’s fine. There’s a lot wrong with getting how it works wrong.
No, when the vote is lost…its lost. Stop ignoring the fact that this will happen.
No, not really, it isn’t. The closest it could be said to being “lost” is that you didn’t choose to use it.
Again, looking at the fact that IRV functions the same as a series of elections – each election acting to eliminate one candidate, imagine the following ballot.
[ ] Ghandi
[ ] Jesus
[ ] Count Dracula
[ ] Pol Pot
On this ballot, I’d probably give my 1st two ranks to Jesus and Ghandi. After that, I might consider which of the other two would be least intolerable but I may well not rank either. If Jesus and Ghandi lost the first two rounds of counting, I’ve got no more votes to count and my ballot is “exhausted” – the case you are describing as lost.
It wasn’t though. I could have voted in the final round, I just chose not to. It is exactly the same as if there were a week between each round and each vote was cast again. The first week, I go in and vote for Jesus. Ghandi gets eliminated. The second week, I go in and vote for Jesus again. Jesus gets eliminated. The third week, I’ve got a choice to vote for Count Dracula and Pol Pot. Screw it, I stay home and watch a movie.
Alternatively, The first week, I go inand vote for Jesus. Jesus gets elimianted. The second week, I go in and vote for Ghandi. Ghandi gets eliminated. Again, looking at the remaining options being Count Dracula and Pol Pot, I stay home.
On the other hand, maybe I am fortunate enough that one of my two candidates makes it through the first two rounds. Then I go to the polls in the third week so I can vote for them again.
In no case did I have more or less voting power or more or less opportunity to vote. The only issue is that in some of the cases, I had no interest in voting and chose not to exercise my vote. Perhaps that’s a lost vote from your point of view. Not choosing to vote is not the same thing in my mind as losing a vote. If they’re the same to you then yes, you’re losing a vote.
I don’t think you are. Not chosing to exercise a vote is not the same as losing your vote. I skip votes all the time on things I don’t really care about or don’t feel informed enough about. It doesn’t mean that I don’t have a vote.
Regarding Ballot Complexity, sadly, you may well be right. The fact that we can make an ATM that pretty much anyone can use but we can’t manage to figure out a way to vote boggles my mind but it seems that we can be very clever at being stupid. Fortunately, you’ve got places like Minneapolis, Cambridge, San Fransisco, and Burlington Vermont who are running the trial cases of this. Hopefully we can get it sorted out so that the rest of y’all don’t have any trouble with it. The Australians and the Irish don’t seem to have much trouble with it, but they’re clearly smarter than us.
Well, the ballot may or may not be complex — I think a good designer can (and has; see Australia) make it somewhat easier.
But the idea isn’t complex, not really. The notion of still maybe influencing the result of an election even if you don’t get your first choice is, at bottom, pretty simple.
Btw, any chance that the DFLer that was against IRV is named Steve Cross?