One of the hard things about changing voting systems is that there are many alternate voting systems and there’s no obvious best choice. Provably.
So, although almost any other voting system is better than the First Past the Post one we use now, picking between them is hard, and often quite technical.
The percentage of spoiled ballots in a typical Australian election is about 5 or 6 % (eg here - 60,000 out of about 1.3 million registered voters in WA at the time).
BUT - you have to factor in that an unknown number of these are deliberate protests against the compulsory voting system. You have to turn up to vote or get fined - but nothing can make a person actually vote once they’re in the booth. So people who really really don’t want to vote can turn up, get crossed off, get their bit of paper - and then drop it in the box without writing a thing. They will be counted as a “spoiled ballot” along with any people who were “honestly confused”
I must say, I am not altogether distressed by the prospect that someone who truly can’t write the numbers from 1 to 6 in a group of boxes, does not have their opinion on the best people to run the country taken into account.
Wow. Guess I don’t understand English then.
So in your first post, when you said:
You weren’t talking about a specific situation? Do you really expect anyone to buy that? Honestly, admitting an error is not the end of the world.
There you go with that irrational confidence again. If you had ranked choice voting, more people would run. Notably for this year’s election, I think there would be/have been a greater chance of seeing Trump, Bloomberg, or Webb, etc. run as an independent. [RIGHT][/RIGHT]
Ranked choice voting encourages more parties to participate, encourages enfranchisement in regions with voting records opposite of a given individual and incentivizes candidates to broaden their appeal rather than ignite their fringe base. The only legitimate downsides are some initial confusion in the geriatric community and possibly a day or two more for vote counting. I say legitimate downsides, I don’t consider the secure status of the only two parties we currently have a priority.
I am absolutely in favor of it, for local school boards on up to presidential elections.
I’ve done a lot of scrutineering at elections at all levels in Australia, and I’d estimate that it’s only about 1% of voters who try to cast a valid vote, but make a mistake which makes their vote informal. Most informal ballot papers are blank, and most of the rest are deliberately informal (e.g., a cross in every square, or some rude remarks about the honesty of politicians).
We have a useful control in Ireland, where voters also have to number candidates in order of preference but, unlike Australia, voting is optional - you only have to turn up and vote if you choose to. In Ireland the spoiled vote rate runs at 1% or thereabouts - and, again, a certain proportion of these are deliberate protests, a certain proportion are spoiled for reasons other than the voter’s inability to write numbers in sequence, and a certain proportion are spoiled because they don’t number their preferences sequentially. We’ve no hard data on how many fall into the third category, but it’s clearly much less than 1%.
In the US (where, again, voting is optional) I believe the rate runs at between 1.5% and 2.5% under the current system, so unless you think that US voters are, on average, much stupider than Irish voters there doesn’t seem to be much reason to think that the use of Irish-type ballots would lead to an increase in invalid ballots.
Of course, by the same mechanism, Buchanan’s votes would have gone to Bush. But I’m sure Gore would’ve won handily anyway,
Yes. Nader had 97,488 votes in Florida, as opposed to Buchanan’s 17,484. Bush’s margin over Gore was only 537 votes. If Nader’s votes had transferred predominantly to Gore, and Buchanan’s to Bush, Gore would certainly have won.
Except . . .
Nader’s vote, and Buchanan’s, were low partly because of the sense that, in a first-past-the-post system, a vote for a minority candidate is “wasted”; it won’t directly influence the outcome of the election. The effect of voting for Buchanan is likely to deny Bush a vote, which advantages Gore, which is probably not the outcome a Buchanan voter would hope for. Which means that an awful lot of potential Buchanan voters decide to vote for Bush, knowing that the winner will be either Gore or Bush and, of the two, they’d rather Bush.
With preferential voting, this effect disappears. So if Florida had preferential voting in 2000, you’d expect that both Nader and Buchanan would have had significantly more votes, and Gore and Bush fewer votes.
And, over time, the impression that the victor must be either Gore or Bush tends to weaken, because now you have more than two parties who regularly poll strongly. One thing we note about countries with preferential voting is that they tend not to have, or to retain, overwhelming two-party duopolies. Quite how this would play out in the US, where the party system is radically different from other democracies, it’s hard to say. But I think a weakening of party loyalties (on the part of voters) and of the entrenched advanatage enjoyed by the Republican and Democratic parties, is likely. It might be that under preferential voting not only would Nader’s vote be larger, but it might flow less reliably to the Democrats on second preferences. And the same goes for the Buchanan vote.
Are we keeping the Electoral College and having states select their own electors, the way we do now? Because let’s say we get the weakening of the two-party system (as described by UDS), and we get four or five parties that can win states. That means that it would be far more likely that no candidate gets a majority of the EC and the election gets thrown to the House, right? That’s probably not going to be the desired result.
So, if we did use IRV for President, we’d probably want to get rid of the EC and have the Feds conduct the election themselves. Or maybe we could try an interstate compact similar to the National Popular Vote compact. But I’d have to think for a minute how to design that.
Good question. Couple of thoughts:
Assume that people want to stick with the electoral college mechanism. Abolishing that is an interesting question, but a whole different one.
Assume also that people want to stick with the “winner-takes-all” rule that prevails in nearly all states - that is, whichever candidates wins Florida gets all 25 of Florida’s electoral votes. Again, there are arguments to be had over abolishing that rule, but they are different arguments to the ones we are having here.
Right. Assume that three or more candidates have electors in the electoral College, and none has 50% or more of the electors. What can be done?
One possibility is to stick with the present rule; the election is decided by the House. As you say, this happening more frequently than at present (which is practically never) would not be seen as a good outcome.
A second possibility is that the electors vote in the same way as the citizens have already voted; that is, by numbering the candidates in order of preference. If no candidate secures a majority of the first preference votes, the candidate with the lowest total is eliminated, and his or her votes distributed according to the next preference, and so on until someone is elected.
This makes sense, and has the added attraction that there is an even lower chance than under the present system of the election being thrown back to the house. But it does raise an interesting question . . .
7 . . . which is, are electors to be pledged only as regards their first preference, but free to distribute second and subsequent preferences as they wish? Or are they pledged as to all their preferences?
8 The latter seems to give individual voters more control and to keep the electoral college members as the ciphers that they are currently expected to be. But it does raise an interesting problem. Suppose my choice for president would be (1) Bush (2) Gore (3) Buchanan (4) Nader, but no campaign is offering electors pledged to that particular sequence of preferences. If I vote for the Bush electors, it may well be that they are pledged to transfer their second preferences to (say) Buchanan.
So, giving me a preferential vote allows me to express my preferences with great precision, but much of the benefit of that is taken away again if an electoral college, with its own voting, is interposed between me and the actual election of the president, since the preferences expressed by the electors will determine the outcome of the election, not the preferences expressed by the voters.
The democratic but cumbersome way around this problem might be to decree that, if no candidate gets a majority in the electoral college, the lowest candidate is eliminated and then the general election votes in all states are reallocated on the basis that votes for the eliminated candidate are redistributed according to the next effective preference. The result of this, in one or more states, may be that the state now goes to another candidate, and the electors from that state are replaced with electors pledged to that other candidate. And this process continues until one candidate secures a majority of votes in the electoral college.
Off the top of my head, I think this could be done with an interstate compact, which wouldn’t require a Constitutional Amendment.
But… if we do it this way, would the final result approximate the results if we did a national IRV? My gut is telling me maybe, but not necessarily. But I’d have to work the math on that, and I’m not sure I’m up for that tonight. But it might be that from a practical perspective, it doesn’t really matter if the results from doing it this way align with the results from doing a national IRV.
What the electoral college does is, of course, to attach disproportionate weight to the votes cast in less populous states. Each state - or nearly each state - also seeks to maximise its voice by adopting a “winner-takes-all” rule in the allocation of electors. The combined effect of these two features is that it’s possible for a candidate to be elected as President even though another candidate has actually secure more popular votes and, as we know, this does occasionally happen. If you simply have a single national election in which all votes are accorded equal weight that won’t happen.
Therefore, an electoral college system won’t necessarily produce the same outcome as a uniform national vote. Switching to preferential voting doesn’t change this.
nm…
Bumped.
A ranked-choice voting lawsuit is now on a fast track in Maine, and it may effect the presidential race. Excerpted from today’s Portland Press Herald:
The state’s highest court is moving quickly on a case that will determine whether Maine voters make history in November by using ranked-choice voting in the presidential election.
The Maine Supreme Judicial Court will hear oral arguments Thursday on a case that pits the Maine Republican Party against Secretary of State Matt Dunlap, a Democrat, who is appealing a lower court ruling. The high court will have until Sept. 24 to make a ruling.
At stake is whether a people’s veto initiative asking Maine voters if they want to use ranked choice in presidential primary and general elections will appear on the November ballot.
The clock is also ticking for Dunlap, who needs to prepare statewide ballots in time for printing and distribution to absentee voters, who are expected to participate in record numbers to reduce their risk of contracting COVID-19.
Attorneys for Dunlap and the Maine Republican Party filed their respective responses with the high court Monday.
If the court allows the question to appear on the ballot, ranked choice would not be used for the race between President Trump, Democratic nominee Joe Biden and three other candidates.
If the high court rules in favor of Dunlap, the veto question would not appear on the November ballot and voters would be allowed to rank their choices for president.
Maine, one of only two states to split its Electoral College votes by congressional district, awards one vote to the winner of each of its two districts and two votes for the statewide winner. In 2016, Maine split its electoral votes for the first time since 1988 when the 2nd Congressional District voted for Trump by a 10-point margin, but Democrat Hillary Clinton won the 1st District and the statewide race.
This year, five candidates appear on the ballot, including Trump, Biden, the Green Party’s Howard Hawkins, the Libertarian Party’s Jo Jorgensen and the Alliance Party’s Rocky De La Fuente. That could spell trouble for Trump if he isn’t preferred over Biden by voters whose top pick is a third-party candidate.
Maine’s ranked-choice system has now twice been endorsed by voters statewide but has been rejected by 2nd District voters. Maine uses ranked choice in legislative and gubernatorial primaries as well as congressional primary and general elections.
The system will also be used for the first time in a U.S. Senate race in November. Incumbent Republican Sen. Susan Collins is running against Maine House Speaker Sara Gideon, a Democrat, Green Party candidate Lisa Savage and independent Max Linn.
The system allows, but doesn’t require, voters to rank candidates in descending preference. If no candidate wins more than 50 percent of the vote in the first round of tabulation the candidate with the fewest votes is eliminated, and their voters’ second choices are reallocated to the remaining field. This continues until one candidate has more than 50 percent of the votes.
The system has withstood repeated challenges in federal court. Most recently, a federal judge in Bangor again rejected a legal challenge by the Republican Party to its use in the U.S. Senate race in Maine and the party chose not to appeal.
Last week, the Maine Republican Party got its first legal victory when a Cumberland County Superior Court justice overturned Dunlap’s determination that the people’s veto effort did not collect enough valid voter signatures to place it on the ballot . Just under 1,000 signatures had been rejected because they were collected by circulators who were not registered to vote when they were collecting signatures, which is required under the Maine Constitution.
Leaning heavily on a prior U.S. Supreme Court decision, Cumberland County Superior Court Justice Thomas McKeon ruled that Maine’s constitution was in conflict with the U.S. Constitution and that the requirement that circulators be registered voters violated the circulators’ 1st Amendment rights.
But Dunlap’s office pointed to a state Supreme Judicial Court precedent that had rejected a similar appeal based on the same federal case McKeon had cited.
Dunlap’s appeal, filed by Assistant Attorney General Phyllis Gardiner, noted that a federal court had found in 1999 that Maine’s requirement that circulators be registered voters was not in conflict with the U.S. Constitution. Gardiner also notes that the Maine Republican Party did not argue or show it had any difficulty in finding registered voters to serve as petition circulators.
“They successfully engaged a total of 543 circulators, only 10 of whom were found to be unregistered at the time they collected signatures,” she wrote.
Gardiner also noted McKeon made a factual error in not rejecting signatures that were duplicates or signatures of people who were not registered to vote. Even if the high court agrees that the petitions from unregistered circulators were allowed to stand, the people’s veto backers would still be 22 signatures short of the number needed to put the question on the ballot.
But in his response, Patrick Strawbridge, the attorney for the Maine Republicans, argued that court rules allowing for an automatic stay of a superior court justice’s decision while a case is being appealed do not apply in this case.
Strawbridge, who has also served as Trump’s attorney in cases before the U.S. Supreme Court, also contended that both the Committee for Ranked Choice Voting and Dunlap are now attempting to introduce new evidence that was not in question during McKeon’s deliberations.
He also argues that allowing the veto question to move forward does not cause any harm, and that keeping the question off the ballot this close to the election would likely add to voter confusion.
“Their claim for irreparable harm is weak, given that the Superior Court’s order at most simply maintains the longstanding status quo for Presidential elections in Maine,” Strawbridge wrote. “And the competing harms are great, given that a stay would deny the Appellees the ability to exercise their fundamental right to place the veto before the people of this state within the timeline mandated by the Maine Constitution.”
Dunlap’s appeal asks the Law Court to stay the lower court decision and allow his decision to stand, leaving the people’s veto question off the ballot in November. The court is likely to act earlier than Sept. 24 given that the ballot needs to printed in time to be sent to absentee voters by Oct. 3…
Traditionally the ‘Cock and Balls’ garners quite a few votes at all elections.
For more on this subject, see www.fairvote.org.
Just want to say I would love this.
My aunt is a politically-savvy Mainer, and she loves RCV.