Deconstructing the 'Vote No to STV' arguments

Note: I wrote this post for my 'Vote Yes to STV' blog which recommends that B.C. voters vote in favour of changing the provincial electoral system to STV on May 17, but it ended up being so long that I figured I'd post it here too, in the hopes of increasing the probability that at least one person makes it through to the end (you've been warned).

----

A website has been setup for a Vote No campaign against STV. The url and the title on the site read 'Know STV'. I guess 'know' is the obvious pun to go with for a vote no campaign, but I find it a bit odd since, in my opinion, the biggest hope of 'no' supporters is that people won't 'know' much about STV.

Consider what happens when people 'know' a lot: The members of the Citizens Assembly studied electoral systems for over a year and at the end, they voted 146-7 that STV is better than the current system. The Law Commission of Canada did a study and concluded that just about anything was better than First Past the Post (the current system) including STV (they recommended MMP as the best choice).

The truth is that pretty much any time any group has seriously studied electoral systems they have concluded that both STV and MMP are better than First Past the Post.

So maybe it's not surprising that, on a site whose avowed mission is to help people 'know' about STV, there is no section which actually explains what STV is, nor is there a link to any of the many animations which are probably the best online resource for understanding how STV works (I like the Australian one the best) - but they do have almost 1000 words in their section on the biographies of the Know STV backers. Seems like they want you to know 'know STV' more than they want you to know STV, if you know what I mean.

Now, it's true, the site does have a long Q&A section but this is more of a forum for them to give answers to softball questions than it is a place to get clear answers to help you 'know' what STV is.

For example, the first question is, "What is the Single-Transferable Vote (STV) proposed by the Citizens Assembly as BC's new electoral system?" - which sounds promising right?

But the answer given is, "The Single-Transferable Vote (STV) is an alternative to the First Past The Post electoral system"

Wow, that really clears it up, thanks! The rest of the answer is devoted to a comparison of the usage rates of STV and First Past the Post around the world which is interesting, but not what was asked (I always find it annoying when someone is asked something and they hijack the question - but when it's their own friggin' question you really have to wonder).

They do go into more detail about how STV works in their next question, "How does the Single-Transferable Vote (STV) differ from our current electoral system?" but rather than explain it clearly they make a few misleading statements and send the reader off to the Citizen's Assembly website for a full explanation.

Since I brought it up, here are some of the misleading statements:

"BC currently has 79 different constituencies but under STV there could be as few as18 constituencies or less. Each larger constituency would have from two or three members in rural areas to as many as seven MLAs in larger urban areas."

Aside from the confusing, bad writing, ('as few as18 or less'?) this paragraph could give the impression that there will be fewer MLA's elected under STV when in fact the number of MLA's elected (79) will remain the same.

The next line is, "Voters would rank all candidates in that larger constituency by their personal preference" which is simply not true. Voters could rank just one candidate or they could rank as many as they want. There is little point for most voters in ranking more than a few candidates and it is highly likely that the number of voters who decide to rank *all* the candidates would be extremely small. I know I have no plans to.

At any rate, I invite anyone to spend some time here on my blog and try to learn more about STV, (hint: click on the 'What the heck is STV' link) and then go to the Know STV site and see which one helps you learn more about how STV works.

----

All right, enough about how the Know STV folks don't really want you to know all that much about STV or electoral systems in general, let's take a more detailed look at their Q&A and correct some of the areas where it lacks a little in the 'fair and balanced' department.

First question: "What is the Single-Transferable Vote (STV) proposed by the Citizens Assembly as BC's new electoral system?"

I already talked about how they don't answer the question, but let's look at the question they did answer which is, "Which jurisdictions use STV and which ones use First Past the Post?"

Their answer, which is that first past the post is used in a lot more places than STV, is accurate but it misses an important aspect which is the trend. And the trend is that countries are abandoning our First Past the Post system in favour of more proportional systems. In New Zealand, they adopted the Mixed Member Proportional System. In Australia, we are seeing the gradual spread of STV (most recently adopted in the State of Victoria.

The European parliament has members elected from countries all across Europe. If you take a look at the linked map, you can see that while only 2 European countries use STV to elect their members, that is greater than the number (0) which use First Past the Post. Even England, the country which spread First Past the Post to so much of the world, no longer uses it for the European Parliament elections. Similarly, none of the recently set-up parliaments in Scotland, Wales and Northern Ireland use First Past the Post (Scotland and Wales use MMP and Northern Ireland uses STV).

The fact that so many more countries use First Past the Post than STV is more of a historical artifact than a ringing endorsement of First Past the Post.


Q: "So how does our current First Past the Post electoral system work by comparison?"

A: "Under First Past the Post, each voter chooses one candidate to represent their constituency and the candidate who wins more votes than any other is elected.

Each FPTP constituency has one MLA who is personally accountable to those voters and the constituencies are much smaller both geographically and in terms of the number of voters in each one."

So after reading this answer would you get the impression that there is exactly the same number of voters per MLA under STV as there is under the current system? If not, then you've been misled. Another good question to ask voters might be how personally accountable they feel their MLA is currently. Especially when in many cases over half the voters voted against that MLA (which wouldn't be the case under STV). And when most of the people who did vote for the MLA were probably voting for the Party the MLA represented and if they wanted to support their party they only had one choice of who to vote for (unlike under STV).

OK, next question.

Q: "Isn't STV a lot more complicated than FPTP?"

AKA, 'How often does STV beat its wife?'

A: "Yes. One of FPTP's biggest advantages is the simplicity and ease of understanding it brings to all voters. In recent New Zealand local elections using STV for the first time 11% of all votes were disqualified, more than 14 times the number rejected in the previous election."

Of course, if they had done their homework, they may have run across this quote from the Citizen's Assembly website:

"Does changing the voting system lead to more spoiled ballots?
A change in voting system or ballot form does not have a large impact on spoiled ballots. For example, when elections for the parliament of Tasmania switched from First Past The Post (FPTP - our current electoral system) to STV (the recommended system) the rate of spoiled ballots was 1.2% in 1906 (under FPTP) and 2.9% in 1909 (under STV). Another example of this marginal impact is when the House of Representatives in Australia went from marking crosses on the ballot to listing numbers, the rate of spoiled ballots changed 1.2%.

You may have heard that New Zealand had a problem with counting STV votes in its Fall 2004 municipal elections. That was a problem with its computers, not with the STV system. The Assembly has designed BC-STV so the count can be done either by hand or by computer."

oops.

Anyway, the answer continues:

"Voters may also be faced with a very large ballot and dozens of candidates in larger ridings, making it hard to rank the candidates knowledgeably."

OK, imagine someone said this to you,

"Shoppers may be faced with a lot of different cereals making it hard to choose a cereal knowledgeably"

Let's face it, any trip to the store involves choosing between a multitude of various models of just about everything, and I think if people can manage this they can manage to pick out their favourite candidate (or someone from their favourite party) out of a long list. And if the success of big-box stores is any guide people prefer more choices not fewer. Shocking, I know.

The answer gets better,

"Voters will also be confused by a mathematical quota called the Weighted Inclusive Gregory System which determines how and where exactly their vote will be 'transferred' to, by having to rank a large number of candidates in each constituency and by the need to trust computers to get the results right."

You hear that voters, you *will* be confused (whether you like it or not!). Of course, if you let the 'know STV' folks get to you with their scary sounding mathematical formula I guess it's possible. But someone who really wanted you to 'know' STV might take this opportunity to explain just how your vote will be transferred (it's not that scary):

If the person you put #1 gets more votes than what they need to get elected, part of your vote gets transferred to your #2. How big a part? Well if your #1 got so many votes he/she only needed half of them to get elected, then they only get half your vote and the other half is transferred to #2. If they got so many votes they only needed 1/3 of their votes than they only get 1/3 of your vote and 2/3 is transferred to your #2 choice.

The other way your vote can get transferred is if nobody has enough votes to get elected, in which case they eliminate the poor sap who has the fewest #1 votes. If that poor sap was your #1 choice, then your vote (rather than being wasted like under the current system) is transferred to your #2 choice.

The same rules apply to your #2 choice as well so it's possible part or all of your vote could end up getting transferred to your #3 choice as well. (e.g. You vote for an independent with your #1, but they get eliminated and your vote is transferred to your #2 who is from the Green Party, but they get eliminated as well, so your vote is transferred to your #3 choice, and helps an NDP candidate edge out a Liberal candidate.)

OK, where was I, right, next question:

Q: "Will STV increase or reduce local representation and accountability?"

Before I get to the answer, I should mention the extensive section of (7!) links that the 'know STV' site has. One of these 7 lucky links is the "The Single Transferable Vote in practice", a report undertaken on behalf of the Scottish government (before they adopted STV for use in local elections).

Here's a quote from criticisms of how STV works in practice in Ireland:

"the heavy emphasis on constituency casework, faction-fighting between candidates from the same party, a focus on constituency, localist matters in election campaigns and parliamentary work, 'friends and neighbours' voting, are all seen as resulting - at least in large part - from the candidate-centred, preference voting of STV"

Yes, a frequent criticism of STV is that because candidates have to compete with members of their own party for seats, they focus too much on local issues and local representation at the expense of 'big-picture' thinking.

The Scottish report eventually concludes that it is hard to tell whether the intense focus on local issues is a function of the electoral system or just some quirk of the Irish in general (or possibly due to the fact that they only have national government with no provincial or local governments).

Anyway, what do the 'Know STV' folks say about STV and local representation?

"There will be less local representation and accountability because STV will mean much larger constituencies and MLAs will be representing far more people over a wider geographic area."

Once again, may I point out that the number of voters per MLA will be the same under STV as it is now. So MLA's will be representing the same number of people as before. It's just that instead of having (for example) 4 ridings each with it's own MLA, you would now have them combined into one riding with 4 MLA's. So maybe the MLA's will have to figure out how to divide up the riding between them - an insurmountable challenge? I doubt it. Personally I'd rather have four MLA's to potentially go to with a problem - including one that I voted for, vs. having only one choice and that being someone I voted against because I couldn't stand them, but that's just me.

They continue, "In large rural constituencies that contain a major town, it’s possible that all MLAs elected will come from that town because that’s where the most voters are, reducing accountability for other parts of the constituency."

Let me see if I can explain this. If an area was large enough to be a riding under the current system, then that means it has enough population to elect someone to represent them that is from their area - if that's what they want (and if they make the effort to get off the couch and go vote). This might be a good time to point out how under the current system parties can bring in candidates who don't even live in the riding at all (the recent election of John Tory in a by-election in Ontario is a good example of this) and people have no choice if they want to support their party. Under STV people could choose the representative from their party from their local area. Here's another quote from the Scottish report:

"Moreover studies of Irish voting patterns also indicate that candidates from a local area will receive more votes from that part of the constituency in which they live (Farrell 2001). Accordingly Irish political parties have developed sophisticated vote management and election strategies. These tend to involve ensuring that:
candidates are picked from different corners of the constituency and voters in each locale are actively encouraged to vary the ordering of their preferences so as to maximize the efficiency of the party vote. The basic idea is that the more equal the spread of first preferences across the different party candidates, the greater chance that more will be elected (Farrell 2001, p.146)."

But wait, Ireland and B.C. aren't the same, next question:

Q: "STV supporters say local representation is very good in Ireland under STV. What's the difference with BC?"

A: "BC and Ireland are quite different geographically, with BC many times larger. However Ireland's population is very close to BC's 4 million people and they have 166 representatives in their parliament, called the Dail, while in BC we have just 79 MLAs in the B.C. Legislature."

This is of course true, but it's just as true under our current system as it would be under STV. If we need more MLA's to represent us, we should have more MLA's. If we're short on MLA's that doesn't have much bearing on a comparison between two systems, both of which have the same number of MLA's.

They continue,

"Inevitably with huge ridings and few MLAs parts of BC would likely lose local representation. In some areas it is possible that no local candidate would be elected as an MLA, removing local representation completely."

'huge ridings and few MLAs?' Did I mention that there will be the same number of MLA's under STV, yeah I think I did.

Let's face it, this whole long list of concerns about local representation comes down to a worry that MLA's may be distributed slightly more unevenly around the province than they are now. Even though this will be left pretty much entirely in the hands of voters to decide for themselves. And there is no mention of how the fact that MLA's will have to worry about being defeated by members of their own party will force them to be more attentive to local needs. Or how the absence of safe Liberal or NDP seats like under the current system will mean that people will have to get elected based on their personal reputation for getting things done vs. which party they are running for.

OK, next question.

Q: "Does STV give proportional results? That is, if a party gets 10% of the popular vote in B.C. would it win 10% of the seats?

A: "No. STV supporters say it is more proportional than FPTP but there is no guarantee that seats won will correspond with popular vote. Proportional representation electoral systems such as List PR are designed to ensure such proportionality, not STV.

If a party got 10% of the vote under STV it would be unlikely to win a seat in any constituency in BC. Look again at the example of a constituency of 100,000 voters electing three members: the number of votes needed to win is 25,001, which means that a party would need at least 25% support to win one seat of the three."

Never mind for a moment that the Green party got 12% of the vote in the last election under First Past the Post, got 0 seats, and in all likelihood *would* have elected members under STV, instead how about we ask a different question:

Q: Is STV far more proportional than our current system and has this been clearly proven time after time in elections around the world and is the proportionality of STV much closer to being perfectly proportional than it is to matching our current system?

A: Yes. OK, let's move on.

Q: "Are MLAs elected with equal levels of support under STV?"

A: "No. Proponents say because STV it is more proportional "overall" if is a fairer system. But a candidate in a two-member riding in northern BC can get elected with 33% public support while a Vancouver or other large urban centre candidate can get elected with just 13% of the votes cast.

This means some MLAs have had to win far more support than others to be elected to the BC Legislature."

OK, Time for a math lesson. The candidate in the rural riding is getting a larger percentage of a riding with fewer voters. The candidate in the urban riding is getting a smaller percentage of votes in a riding with more voters. Can we see how these factors balance each other out so the two candidates need roughly the same number of total votes to get elected.

Now it's true that there is a slight difference. Say we assume 10,000 voters per riding then in a 2 member rural riding, a candidate would need 1/3 of 20,000 (10,000 each for the 2 members) votes = 6,700 votes to get elected. In a 7 member urban riding, a candidate would need, 1/8 of 70,000 = 8,750 votes. Of course a quick look at the last election results will show that the riding of Victoria-Hillside was won with 7,878 votes while the riding of Fort Langley Aldergrove was won with 16,527 votes. And this kind of disparity, far larger than the biggest possible difference under STV, is par for the course for First Past the Post (sorry about the mixed sporting metaphors).

I'm not even really sure what this question is getting at, but if MLA's getting elected with an equal number of voters actually is important, than we should point out that STV is far superior to our current system once again in this area.

Are you starting to notice a pattern yet? When the Know-STV folks feel that First-Past-the-Post has an advantage, they go for a direct comparison of the two systems. But when they think ('know') that STV is superior, they compare STV to some ideal standard like perfect proportionality or MLA's being elected with exactly the same number of votes each. Also note how rarely they are willing to take on a direct comparison - what does that tell you?

OK, next question.

Q: "Does STV allow independent candidates to win seats?"

A: "Not necessarily. Malta has used STV since 1921 but since 1950 not a single independent candidate has been elected. Any candidate requires significant funding to win election and with STV the constituencies will be much bigger, forcing candidates to raise even more money. In a seven-member constituency as proposed for Vancouver, major parties will likely spend $1 million or more in that constituency campaign alone — an amount no independent candidate could possibly raise."

First off, this answer is simply factually inaccurate. Yes, STV *necessarily* *allows* independent candidates to win seats. Even our old First Past the Post system *allows* independents to win seats. More substantively, the Know-STV folks yet again seem to be confused about STV. A 7 member constituency would indeed be much bigger, but there are 7 seats, so an independent only needs to get 1/7 of the votes (technically 1/8th since only 7 people can get more than 1/8 of a total). The parties aren't going to spend any more money *per MLA* than they do right now and an independent won't have to either.

Let me explain why STV is better for independents, something which pretty much everybody (even most STV opponents acknowledge) but which the 'Know-STV' people apparently don't want you to, you know, know.

Under the old system, the only way for an independent candidate to get elected is to win a plurality (get more votes than anyone else) in a specific riding. Under STV, an independent candidate who gets that many votes in one riding will still (with a very high probability) get elected. But in addition, an independent candidate can get a smaller level of support, spread over a wider area. For example, an independent candidate who gets an average of 15% support across one of the big Vancouver ridings would be able to get elected under STV, but would have been shut out under the old system (much like the Green Party was in the last election).

Furthermore, under the old system, many people don't vote for independents because they know the independent has no chance of being elected so they are wasting their vote (or worse by not strategically voting for their favourite party, they may help that party's opponent to get elected). But under STV, voters can vote for an independent knowing that if that candidate gets eliminated, then their vote will be transferred to their second choice, so they don't have to worry about wasted votes and strategic votes.

Next question.

Q: "Would smaller third parties be elected under an STV system?"

A: "Not necessarily. In Malta, which has had STV since 1921, there are only two parties with elected officials. In recent elections the largest third party has won less than 2% of the vote and no seats. In Ireland small parties have won seats but so have smaller parties in BC under First Past The Post, as recently as in 1996."

OK, this is getting tiresome. How about:

Q: Are smaller parties far more likely to be elected under STV than under our current system? A: Yes.

Seriously, if these folks really want you to make an informed choice why are they trying to deny even facts which are blindingly obvious to anyone who takes some time to think about how the two systems work (not to mention clearly documented by almost 100 years of electoral history). First Past the Post is so hostile to small parties that political scientists have formed a 'law', (Duvergers Law) which postulates that First Past the Post naturally leads to a two-party system.

It's simple, STV is more proportional so you need a smaller percentage of the vote to get elected plus it eliminates strategic voting - thus more small parties can be elected and this has been shown pretty clearly in practice. It's true that in Malta small parties don't get elected - because people don't vote for them - and this is a problem because?? OK, moving on.

Q: "Are women elected in larger numbers under STV?"

A: "No. In Malta women make up just 9.2% of the country's legislators, with only 6 women elected out of 65 representatives. In Ireland just 13.3% of elected officials are women."

I've already discuss this at length here, so I won't belabour it any further, but I'll just ask one more time, why, if the 'Know-STV' people want you to truly know STV, would they fail to mention that in the STV-using jurisdiction most similar to Canada (Australia), more women are elected than in B.C. Wouldn't that information be helpful to someone trying to make an informed decision?

Q: "Does STV mean an end to so-called 'wasted votes'"

All right, I'm just warning you, the 'Know-STV' folks really jumped the shark on this one, so this may get lengthy.

A: "If no vote were to be 'wasted' that would mean every voter's candidate of choice would have to win election - it's not possible or sensible."

Actually, it is entirely sensible and possible - in fact that's what proportional representation means. According to the Citizen's Assembly site:

"STV virtually eliminates 'wasted' votes. For example, in a three seat riding, even if a voter's #1 preferred candidate is not elected, there is a good chance her ballot will help her #2 and #3 preferred candidates win a seat. About 16% of ballots don't contribute to electing a candidate."

So STV doesn't eliminate wasted votes entirely like a pure proportional system would, but it reduces the % of wasted votes (from around 40-50% in a typical First Past the Post election) down to 16% - which is pretty good since it means one in four voters who would have someone in the legislature that they voted for under STV where they would have had noone under First-Past-the-Post.

They continue: "Elections are to select which candidate in each constituency has the most support"
Actually that's a definition of the First-Past-the-Post system not 'elections', but it's instructive to see that the Know-STV people think that the First-Past-the-Post method is equivalent to 'elections'.

"...and then which parties across the province have enough support from elected members to form a government.

STV supporters say that by ranking your choice of candidates, the odds are one of your choices will win a seat. But that's a little like saying if you bet on every horse in a horserace, one of your picks will be a winner."

Actually, it's not like that at all. Voters are expressing a preference between who they want to help win the race vs. who they want to lose, not trying to profit from every possible outcome - that's why the ranking is crucial. If you really want an analogy, it's more like saying that if your friends vote down your choice of a pizza with anchovies, egg and corn, you can still have a say in the decision between the vegetarian option and the pepperoni pizza rather than being shut out of the crucial decision because you wanted to be able to express your true preference first.

"And because of the complicated transfer system, you will never know where your vote actually went in electing the MLAs for your constituency."

Wrong again. If your first choice didn't get elected, it's a good bet that your vote got transferred to your second choice, unless the person you voted for was the last one to get eliminated. If your person was the last to get eliminated then your vote didn't get transferred at all. That's tough but console yourself by remembering that if we still used FPTP your vote would *never* get transferred.

Now, if your first choice did get elected, either they were the last one elected in your riding (in which case they got all of your vote and none was transferred) or they weren't the last one elected and they got part of your vote (enough to get elected) and the rest was transferred to your next choice. You can check the result to see how many votes your candidate got compared to how many they needed. The same calculation as before applies. If your candidate got twice as many votes as they needed then they got half your vote and half was transferred to your #2.

So the truth is, if you care to know, it will be easy enough to track where your vote went. It takes a bit more effort than under the current system where - unless you voted for the person who won - you might as well have put your ballot in the trash can as in the voting box, for all the effect it had on the election.

"Under First Past The Post, your vote goes to one candidate and is counted clearly. Regardless of your choice, that's not a wasted vote."

Seriously, why are they even bothering with this nonsense? If my vote had 0 impact on who get elected, it's a wasted vote, end of story. True, you could argue that no vote is wasted because it's important on principle to vote, but the 'know-stv' folks know as well as you and I do that that isn't what we're talking about when we talk about wasted votes. Sheesh.

Q: "What would happen to nominating meetings under an STV system?"

A: "Because STV combines the smaller single member constituencies of our current system into large multi-member constituencies, the likelihood is that special interest groups would dominate the nomination process of political parties even more than today."

Umm, why is that exactly - I'm not following. Wouldn't bigger ridings make it harder for special interests to dominate a nomination meeting. Isn't that logical - the smaller something is, the easier it is for a special interest to dominate it? I'm just asking.

OK, almost finished, I promise:

Q: "What is the short version of what's wrong with STV? Why should I vote no?"

A: "STV is complicated, confusing, prone to errors and delay, it reduces local accountability, increases the size of ridings, allows MLAs to avoid direct accountability for their decisions, increases party control and allows special interests to dominate party nominations."

So here's the big finish. STV is too complicated and confusing (translation BC voters aren't as smart as voters in Malta and Ireland and Tasmania who have been using STV since before the invention of commercial air travel, the automobile or television.)

It's prone to errors and delay (translation: Elections BC is incompetent and can't manage elections as well as the Irish or the Maltans)

It reduces local accountability (never mind that one of the most common criticisms of STV is that it makes MLA's focus too much on local concerns)

It increases the size of ridings (but not the number of voters per MLA - if this is such a big concern we could always just add more MLA's anyway)

It allows MLA's to avoid direct accountability for their decisions (how exactly? - and don't you think that MLA's are more accountable when voters know they can dump the MLA without voting for a different party?)

It increases party control (Actually, no. How much control parties want to have over their MLA's is up to the parties. - If anything, MLA's who go against the party have a better chance of surviving as independents under STV than they would under the old system).

and it allows special interests to dominate nomination meetings (not that there is any evidence of this ever happening anywhere, but the 'know STV' crew have decided that it's true so there you go. Personally I think logic suggests that larger meetings will be less corrupt but that's just me.)

So that's it. It's quite a long list of questions, but somehow they missed the obvious ones. Which system is more proportional? Which system doesn't require strategic voting? What % of Citizen's Assembly members thought STV was superior to First Past the Post? What happened in the two referendums where Irish voters were asked if they wanted to switch away from STV? Which system gives voters more choice? Yeah, I wonder why they didn't ask any of those questions.

Maybe because they didn't want you to 'know' the answers.