top of page

PR-STV, MMP and Women

First, let me reinforce what the research says:

On average, 8% more women are elected in countries that use proportional representation systems, compared to countries which use winner-take-all systems. Obviously a balance of men and women and men sharing decision making power, setting priorities, working together on a range of issues is important to creating a truly representative Parliament.

I'm no expert. But I've learned enough to understand that when you dig into the "when, where and why" PR elects more women, it's complicated.

Why "PR Elects More Women"

First, PR doesn't elect more women: Voters do.

Why does that happen more often in PR countries? Well, the first step is women have to be put on the ballot in order for voters to choose them.

And here is the crux of the problem and the solution: The party nomination process. The party culture. The party gatekeepers.

Winner-take-all systems make it more difficult for women to break in

In a winner-take-all electoral system, with single member ridings, each party nominates just one candidate to run in each riding.

When: a) there is a bias in the party towards males

b) the incumbent MP in the riding is a male c) the candidate who ran in the previous elections is male Any of those factors make it very difficult for a new woman to win the nomination. How PR systems help women get nominated

PR systems mean multi-member ridings of some sort. Meaning, most of the time, the parties must nominate more than one candidate. The parties can't just nominate women most often in ridings which are strongholds for other parties, like they do in our winner-take-all system. As long as the party has a minimum level of support in a region/district they will now win seats in every area of the country. With PR, if more women are on the ballot in Canada, more women will get elected.

As soon as the parties must nominate more than one candidate, it allows new candidates who are women to enter the system without having to fight to displace the male incumbents. It also makes the a party look very bad if a party has a list and 80% of the names on the list are men.

Or with STV, if the party is running two candidates and both of them are men.

An offering of all men (or all white people, or anything else that so obviously doesn't reflect the diversity of the area) is hard to miss. In a winner-take-all system, with only one name on the ballot, there is simply not the same embarrassment factor, or the same pressure to fix it. Each riding is its own little silo. These are the kinds of "natural incentives" Women for Fair Voting mean when they say "Fair Voting elects more women, naturally."

Single Transferable Vote, Mixed Member Proportional and List PR

Let's dig a little deeper now into the different kinds of PR, and how they might elect more women. First, women's groups tend to support List PR systems. In fact, their submissions often call for a LONG list. In this submission to the Ontario Citizens' Assembly, you can see a chart showing that the PR countries with the most women have Party List systems (some also have quotas). If the parties choose to make sure their lists are half women - alternating male/female, which is called "zippered" - then women have a good chance of being elected.

In many countries, the lists are "closed" - once the list of candidates is set by the party (in democratic nominations, we hope) the candidates are elected according to the order of the lists. Voters cannot choose individuals.

Or the lists can be "flexible", which means to alter the order in which candidates are elected from the list, a very popular candidate would have to receive a certain percentage of individual votes to override the list order. These systems in effect are close to closed list. Imagine you are a voter faced with a long list. A good number of people will just choose the party name, leaving the order of the list unaltered. The third option is open list - meaning, the most popular candidates from the party's list get elected - the order of the list should make no difference. If a party puts up a diverse slate of men and women where voters have full choice, and voters in that country are not biased towards male candidates, women would be just as likely as men to get elected from an open list. It's easy to see why the top countries to elect more women use List PR.

Newsflash: We're not getting List PR in Canada

We're not getting a nation-wide, or province-wide, party list system. All the systems for Canada are rather unique because they keep local representation.

This means a PR system for Canada won't be nearly as proportional as somewhere like Denmark.

It also means that when we talk about electing more women "naturally", we need to deal with what is on the table: MMP and STV (or a hybrid).

Both MMP and STV - the most commonly recommended for Canada - are relatively uncommon voting systems. If you look at the 80 or 90 countries which use PR, MMP is used in 8. STV is used to elect the lower house in two countries (Ireland and Malta) and SNTV (a variation of STV) is used in four others (for a total of 6).

STV is also used in the Australian Senate, in two Australian territories (ACT and Tasmania) and for local elections in New Zealand and Scotland.

Neither system provides a large pool of data to compare rates of women being elected. From the places that do use these systems, some are so culturally different from Canada that it makes no sense to generalize.

But from what we've covered about how PR elects women already, and the countries which do use MMP or STV, I can deduce a few things.

MMP and Women

Women's groups in Canada have tended to support MMP mainly, I think, because it involves a party list.

Important fact number one: The Law Commission of Canada recommended 2/3 of the seats be elected using first-past-the-post, only 1/3 list. And most MMP advocates won't go above 38% top up seats.

So right away, with MMP for Canada, in 2/3 of the contests women will face the same barriers they do today - nominations in single member constituencies.

The potential to elect more women initially lies with 1/3 of seats filled by the open lists. (Closed lists are not on the table for Canada - everyone wants a system where we can choose from among candidates if we wish).

This is exactly what happened in New Zealand, which uses MMP. New Zealand has more list seats in their MMP model than Canada - almost half the seats are filled by closed party lists.

In their first election with MMP, in 1996, the percentage of women jumped significantly - from 21% to 29%. Now it sits at around 30% (Canada is 26%) and it hasn't moved in 15 years.

One interesting thing that did happen was that predictable, almost all of the new women were initially getting elected on the list side. But over time that percentage dropped so that fewer were getting elected on the list and more in the the local constituencies, until the percentage elected on both sides is now about equal, at 31%.

Detailed history in NZ:

1996: 15% women in constituencies, 44% women on lists; 1999: 21% / 40% 2002: 26% / 31% 2005: 20% / 40% 2008: 24% / 37% 2011: 30% / 37% 2014: 32% / 30%

So did MMP help improve the representation of women in NZ compared to first-past-the-post? Definitely.

Is there more now to do at the level of party nominations in NZ order to see any further improvement, even with 50% list seats? Yes.

How do other countries with MMP do? Scotland has an MMP system that is the most similar to what has been proposed for Canada (except it's a closed list). In Scotland, 42% of the seats are list seats. Our Law Commission recommended 33%.

Scotland currently sits at 34% women.

1999, the first election of the Scottish Parliament, saw an historic breakthrough for representation of women - 37%. More women were elected to the Scottish Parliament in one day than had been elected to represent Scotland in the UK House of Commons since women were first able to run in 1918. Wow! But since then, representation of women has stalled or fallen and womens' organizations are campaigning for 50% quotas.

What about other MMP countries?

Germany: 36.5% women (50% of seats are closed list). Hungary: 10%. Point being: PR opens the door. But culture matters and party nomination processes matter.

Single Transferable Vote and Women

Some women's groups actually opposed STV in BC. They would prefer to keep first-past-the-post - claiming "STV doesn't elect more women."

When STV came within 2 percent of winning the BC referendum (58% when the government set the bar at 60%) seeing women's groups take an anti-STV position is heartbreaking. Here's some fun questions to consider.

Which countries elect the most women to the European Parliament and what voting system do they use to do it?

1) Malta (STV) 67%

2) Finland (Open List PR) 62%

3) Ireland (STV) 55%

Wow. But STV can't elect women?

Now let's take a look at Australia. Here's an interesting comparison. Their lower house (the green line) uses a winner-take-all system (Alternative Vote). Their Senate is elected by STV. Same country, same voters. The percentage of women in the lower house is comparable to Canada. Their Senate is 10% higher (38%) and it keeps trending up.

In all fairness, the STV system in the Senate in Australia functions more like a party list system. Instead of ranking candidates, many voters choose to vote "above the line" - endorsing the party list.

Let's look at the Australian states and territories that use STV

Here we can see the real effect of STV using an STV system that looks similar to what we would see in Canada. See sample ballot for Canada here.

Tasmania uses STV to elect its lower house, with each district electing 5 members. Percentage of women elected: 36%

ACT (Australian Capital Territory) uses STV to elect its lower house. Percentage of women elected: 41.2%

Those are higher on average than Germany and New Zealand with MMP closed list and about 50% top up seats!

Neither of these territories use gender quotas. Both elect their Senate with winner-take-all voting - for a direct comparison of effects - and far fewer women are elected there compared to their lower houses with STV. It's also interesting to compare the Australian Territories that use AV to elect their lower houses those that use STV. See the STV territories circled in red. Overall, comparing the upper and lower houses that use AV vs the upper and lower houses that use STV, there are slightly more women elected with STV.

UPDATE: New 2016 figures show 44% women elected using PR-STV in the Tasmania House of Assembly. See chart below of progress over time.

As an interesting side note, STV was used in New York City in the 1920's. It was brought in - fought for by the left - to improve diversity of representation. What happened? It improved diversity of representation. Too many blacks were getting elected for the liking of some people so the elites got rid of it. In the 1920's getting more women elected wasn't a high priority but researchers have concluded that the same mechanism that elects more women elects more minorities.

So what was the opposition to BC-STV from women's groups about?

It appears on the surface to be about Ireland and Malta. While both recently beat the rest of Europe in electing women to the European Parliament with STV, their record of electing women at home is very poor. Malta has only 12.9% women. Ireland elected 23% women in it's last election - only because a quota was put in place tying party funding to nominating at least 30% women. Prior to that only 16% of their legislature was female.

Those two cases are very likely what has led some women's groups to conclude "STV does not elect more women." (Strangely they did not look at Hungary or Romania which use MMP and which elect 10% and 13.7% women respectively, and conclude that MMP does not elect women).

If STV does not elect women, feminists in Ireland must be taking note of their flawed electoral system and demanding a better PR system, right?

Actually, no.

The latest research from Ireland using a "feminist-institutionalist framework" concluded that like all most PR systems, STV provides the appropriate structure for electing women. The problem is the culture of the parties in Ireland (emphasis mine):

"the research finds internal party cultures, systems of party competition, and electoral preferences for incumbency mainly account for the dearth of women representatives in Ireland, not the specific mechanics of STV. If anything, STV’s multi-member nature facilitates the election of new women, which would not have been the case under a single-seat system."

"Thus, as a whole, the literature suggests we cannot conclude any universalised theory as to STV’s impact on women’s representation – rather, its effect on female seat-holding is highly contextualised, depending on the specific environment in which it operates."

The article concludes that the key to women’s improved representation in both jurisdictions lies in reforming candidate selection processes within party organisations.

Applying some common sense to a complicated topic in a Canadian context

So what can we make of all this variation? What can we expect in Canada?

First, Canadian voters have shown that when women are on the ballot, they will select a woman as often as a man. So with Canada's culture, that brings us back to the party nomination processes.

With MMP, 2/3 of the seats will still be elected in single member constituencies- presenting the same barrier for women.

If the parties are pro-active with making sure half their regional lists are women - which I expect some of the parties who genuinely want to improve diversity and look good will do and some will not - we can expect to see more women elected on the lists, increasing the total percentage of women in Parliament.

More may need to be done to encourage the candidates to nominate women, but MMP would provide a good start.

Now consider this: Unlike MMP, with STV, single member ridings are history.

Multi-member districts of various sizes provide an excellent opportunity to nominate and elect more women.

With STV, parties usually only run as many candidates as they think they can win seats.

Imagine there are 4, 5 or 6 seats in a district. The larger parties will likely run two or three candidates each (I think it would be rare they would run four, but possibly).

If they run 2 candidates, that "natural incentive" kicks in to make one candidate a man, and one a woman - so instantly, in many of those ridings 50% of the candidates are women. Running two men doesn't look very good.

If they run three candidates, if they don't run at least one woman out of the three, they would look very bad. So right away, in those ridings, at least 33% of the candidates are women.

In some parties which already do better recruiting women, two out of the three candidates could be women.

You could easily see how, if parties are motivated to appear fair and diverse, more women could be elected with STV than MMP. But here's the catch:

In Canada's rural areas, STV may mean 2 or 3 seat districts. In these cases, most of the parties might only run one candidate. In these cases, women will face the same uphill battle to win the nomination as they do in single member ridings. So the number of smaller districts in an STV model will have an impact on how many women are on the ballot.


The representation of women has been climbing in Canada, although very slowly. We lag behind PR countries because our winner-take-all electoral system is a barrier to getting more women on the ballot in winnable ridings.

Although we won't be getting a Party List PR system with a long zippered list which appears to make it the easiest to increase gender representation the most, both STV and MMP have great potential to significantly increase the representation of women in the House.

Well known social activist Judy Rebick concluded the same thing in her response to the opposition to STV from women's groups:

"Nevertheless, it's true that one of the country's most respected feminists, Doris Anderson, who was a pioneer supporter of PR, did oppose STV in the last B.C. referendum before she passed away. At first Doris persuaded me that it was not a good system for electing women but I changed my mind once I went out to B.C. and talked to voters who understood the system as one that gave them more power to shape the legislature they want.

Also, it is not a vote to choose which system of PR would be best. The Citizen's Assembly, a representative selection of B.C. citizens half of whom were women, already made that determination. So now the vote is either for PR or against it.

Doris had more confidence in our ability to pressure the parties to have balanced lists. I prefer to trust the voters to vote for women and minorities. If we lived in a country where voters were less likely to vote for women than men, then MMP might be better but that is not the case in Canada. Voters are as likely as or more likely to vote for a woman than a man. In fact, study after study has shown that the primary barrier to the election of women is the nomination process that is run exclusively by the parties today."

And in the end, it's all going to come down to how the parties react to the new system, and what they are willing to do to achieve equality.

bottom of page