On October 10, 2007 people in Ontario, Canada will go to the polls for the 2007 Ontario provincial election. At the same time, they will vote for or against an electoral reform proposal that would change the method of voting for future provincial elections.

How the Referendum Came About

The Citizens' Assembly on Electoral Reform was established by the provincial government to review Ontario's existing Single Member Plurality (aka 'first past the post') riding-based electoral system, in use since 1792, to see if it should be retained or replaced. The Assembly was formed of 'ordinary citizens' — 51 men and 52 women, (ages 19 to 78) one from each of the 103 ridings, plus chairperson George Thomson. People of different ethnicities, including aboriginal peoples, were included in the Assembly selection process. Of the 104, 27 were born outside of Canada.

The members of the Assembly met to learn about the electoral systems and process and how the votes are counted. They studied other systems used in Canada and elsewhere in the world. Using a combination of private study sessions and 41 public consultations, the Assembly heard from Ontarians and learned what it needed to know to make an informed choice. Input was not limited to current voters — a special assembly of high school students, as future voters, also had their say.

How Alternatives were Evaluated

In evaluating different systems, the Assembly had a mandate to use this list of criteria:

  • accountability
  • effective parliament - government and opposition parties are included, and can work together to form a functional legislature.
  • effective parties - the existence of strong political parties with different agendas is encouraged.
  • fairness of representation - all types of voters are represented; all votes are equal; parties are represented in proportion to the votes that they receive.
  • legitimacy - voters can have confidence in the process and its results
  • stable and effective government
  • stronger voter participation - voters will be encouraged to vote; all votes will "matter" even in strongly held local ridings.
  • voter choice - voters have a range of choice on the ballot.

The Assembly added "simple and practical" to this list. Voters need to be able to understand the system.

What was Recommended

After consideration, the Assembly decided to recommend Mixed Member Proportional (MMP) as a new electoral system. Why? The Assembly felt that of the systems it considered, MMP best met the criteria above. They boiled those down to the main benefits below:

Greater voter choice

The Assembly says that MMP provides greater voter choice by letting voters cast two separate votes on their election ballot, one for their individual constituency representative and one for a political party. Voters may choose to exercise only one of these two votes, if they so desire.

Details: The first vote cast would be for the riding's Member of Provincial Parliament, or MPP), using the plurality ('first past the post') system to determine the winner, as happens today. The second vote would be for a political party, which might be the same party as, or different than, the party their choice for MPP represents. The party vote is not cast for any specific individual, though the individuals from each party who could be selected as a result are known and published in a list beforehand.

Commentary: This concept partly frees voters to make the best individual choice for their riding, while still supporting the political party that they prefer, should those two differ. Voters who say "I like Candidate A, but I can't stand Party X that he represents." can thus vote for Candidate A, and also for Party Y. Of course, it could also lead to voters trying to guess who will form the government and vote locally that way, hoping to score a cabinet minister.

Fairer election results

The Assembly felt that MMP produces fairer results - the number of seats allocated to each political party better represents their share of the popular vote.


After the riding results are known, parties are assigned "list seats" in a manner that approximates the overall attribution of party votes. In the Ontario proposal, there will be 90 ridings, and 39 list seats, for a total of 129 MPPs. The 39 list seats are divided amongst the parties in a way that makes their combined representation (local and list MPPs) roughly represent their share of the party vote. Parties cannot lose riding MPPs this way, they can only gain additional seats as needed to make up their proportion of the vote.


Past Ontario elections have seen a party win a majority of seats with less than 50% of the popular vote. Infamously, the 1990 Ontario provincial election saw a huge protest vote. To everyone's surprise, the New Democratic Party capture a majority (74 of 130 seats) with only 37.6% of the popular vote. Under an MMP system, a minority would likely have resulted instead, which would very likely have been more representative of Ontarian's desires.

Stronger representation

MMP preserves the strong local representation of the current system. Each geographic area remains part of a (slightly larger) riding, which still has its local MPP to focus on local issues and represent the riding. MMP also allows each geographic region to be represented in government through the list members, who can be assigned to represent areas that did not elect local MPPs from the party that forms the government.

Stability of government

The Assembly acknowledges that minority or coalition governments are more likely under MMP than the current system, but does not believe that this is destabilizing. Its report points out that Germany, which uses MMP for the Bundestag, has had 16 elections since 1948, while Ontario under the current Single Member Plurality system has also had 16 elections in that time.

The Assembly also speculates that abrupt shifts in party power are less likely in the MMP system than under Single Member Plurality, where voters often "punish" one party by throwing support to its opposition.

Effectiveness of parliament

The Assembly notes that MMP produces a less adversarial system where the parties cooperate more and do more work in all-party committees. Opposition party members are less likely to be "waiting" for the next election, and more likely to be doing real work for the voters.

What Else was Considered

The Assembly considered Single Transferrable Vote as its second choice, but members felt that it would work poorly for Ontario, with its dense urban centers and large and sparsely populated Northern Ontario ridings. (STV is arguably better suited to municipal elections for urban centers.) In selecting alternative systems, the Assembly voted 75 to 25 in favour of recommending MMP over STV, with one spoiled ballot and two absentees. In then recommending the change, the Assembly voted 86 to 16 in favour of recommending MMP over the current system, with one absentee.

Proposed Implementation of MMP

The general details of MMP can be covered in Mixed Member Proportional. As to its proposed implementation in Ontario:

There will be two ways for MPPs to be elected, riding and list. Riding will work as it always has, but the current 103 riding seats (increasing to 107 on October 10, 2007) will decrease to 90. 39 list seats will be added, for a total of 129 seats (still 1 less than the 130 MPPS of the 1990s.) With 129 MPPs, Ontario will still have far fewer representatives for its population than any other Canadian province or territory.

The proposed proportion of riding seats to list seats is 70:30. This is a relatively higher proportion of riding seats to list seats than other regions that use MMP, where the proportions are usually more like 50:50.

These proportions should ensure that situations will not normally arise where a party has more local seats than its party vote apportions. In MMP terms, this situation is called "overhang". Some MMP systems temporarily create more seats to balance the overhang. The Assembly's proposed MMP system does not. If an overhang occurs, the list seats will be allocated in the way that closest matches the party vote, but the overhang will inflate the number of seats allocated to the party in question versus their popular vote, and reduce the other parties' share accordingly. The Assembly feels that this situation will rarely occur.

The official parties will nominate local candidates, and also a party list, which will be published will before the election. The party is required to make the way these names were chosen public. The lists should ideally represent a mix of regions, gender, and ethnicity. The names of local candidates and list candidates may overlap. Elections Ontario will publish the lists widely.

These lists are closed, which means that the order of names is fixed and voters cannot select a specific list candidate or rearrange the name order in any way. One expected benefit of this is increased representation for women and ethnic groups in the legislature.

When seats are apportioned, each party first gets it locally elected members. Then, if it is entitled to more MPPs under MMP than it elected based on its percentage of the party vote, those MPPS will be selected in order from the party list, skipping anyone who was elected already as a riding MPP. Several formulas exist for apportioning the list seats. The Assembly has recommended the Hare formula which has the virtue of relative simplicity.

The Assembly's recommended implementation of MMP requires a party to get at least 3% of the total vote to be entitled a list seat in the provincial legislature. In past elections, the closest a "small party" came to 3% of the popular vote was the Green party, with 2.8% of the vote in 2003. Any party, regardless of popular vote, can carry a specific riding if its candidate wins a plurality in that riding.

Any vacancies in local ridings between elections would be filled by by-election, as usual. Any vacancies in list ridings would be filled with the next eligible name from the party list. If a list member were to quit his or her party, they might (or might not) have to resign the list seat. The Assembly makes no recommendation on this point.

Criticisms of MMP

A number of criticisms have been levelled at MMP. In the main, I think that these counter-arguments come from a position of mistrust in both politicians and voters. I will not attempt to argue for their side here. Instead, I will report them as I understand them, and lightly rebut them where I feel it is warranted. Some prominent political experts do echo these objections, though. The leader of the Progressive Conservative Party of Ontario, John Tory, is one of those who cites many of the arguments below in discussing MMP, though he has not taken an official stand against it to date. Ontario Energy Minister Dwight Duncan is on record as opposing MMP.

Fewer local ridings

The number of ridings would drop from 107 to 90. However, the reduction from 130 to 103 ridings in 2000, a reduction of 21%, which was hardly noticed by the average Ontarian. This further 16% reduction will be balanced by 39 new MPPs who will represent all Ontarians. I personally find this criticism to be spurious.

Politicians chosen by other politicians

The NO MMP web site suggests that list members will be chosen by politicians, and implies that this is necessarily nefarious and underhanded. While this could happen, voters are unlikely to be supportive of parties who do this, unless every party acts in this way. In jurisdictions currently using MMP, the party lists are created in very public fashions, often by open party votes. I personally find this criticism to be reactionary.

Back-door deal-making

The NO MMP web site suggests that parties will dicker to form a coalition. This is undoubtedly true, just as it is today when minority governments form. There's no clear reason to think that this horse-trading will be any less public than it is today. Such is the power of the free press. One can as easily view MMP as fostering party cooperation as opposed to skullduggery, as NO MMP chooses to do. It is also valid to argue that the real back-door dealing happens within the parties, which are (of course) loose coalitions of other individuals, groups, and special interests.

More politicians

The number of MPPs would rise by 22, with corresponding expenses in salary and staff. However, this is still 1 less than the 130 MPPs which sat from 1987 to 1999. (The Mike Harris government reduced the legislature with its Fewer Politicians Act.) It seems to me that 22 MPPs and staff is a small price to pay for the benefits of proportional representation.

Confusing balloting and vote-counting

MMP is more complex than Single Member Plurality. There are two votes to count, and the voting must be tallied in two stages: first the riding votes are decided, and then the list votes are apportioned. It will take longer, so clear and final results may not be available right away, possibly until the next day. However I trust that the broadcast networks will still confidently predict the broad results before bedtime.

A weaker, indecisive Ontario

This argument, being opinion, is difficult to refute. It will likely result in more minority governments, and may even result in different coalitions governing between elections. On the other hand, the balance of power is more likely to accurately reflect the will of Ontarians, with large policy shifts coming less often, and shockers like the 1990 NDP victory less likely.

Fringe parties holding the balance of power

It's possible that, say, the Green party could hold the key to a government's survival of a non-confidence motion. However, a 3% popular vote threshold should keep most one-note parties from gaining a seat. Even the Green party has yet to break that barrier under current voting rules.

The need to lean on others for support can be a spur to cooperation rather than a sinister threat. Some Canadians feel that the early-1970s federal Liberal minority, supported by the NDP, was among Canada's most productive parliaments. We can also look to the current federal Conservative minority, which has successfully brokered deals with various opposition parties to keep itself afloat long after pundits predicted.

Voter dislike of specific list members

The NO MMP web site makes a big deal of the fact that if you don't like someone at the top of a party list, there's no way you can keep them from being elected if the party gets sufficient support to earn a list seat. That may be true, but then again there's no way to prevent the election of the 89 local MPPs who aren't in your riding, either, many of whom you may like even less! The parties will want their list members to attract voters, not to repel them. If a voter finds him or herself unable to support any party's list, that voter can simply choose not to exercise that half of the ballot.

Support for MMP

CBC news, in its report on Minister Duncan's views, notes that several ministers of Ontario's government (Attorney General Michael Bryant, Municipal Affairs Minister John Gerretsen, and Health Minister George Smitherman) support MMP. Other support crosses party lines. Former NDP leaders Ed Broadbent, Stephen Lewis, and Bob Rae (now a Liberal) support MMP. Current Liberal MP Carolyn Bennett, Conservative senator Hugh Segal, and Toronto mayor David Miller are in favour. Former Green Party of Canada leader Jim Harris, current Green Party leader Elizabeth May, and current Green Party of Ontario leader Frank de Jong are supporters. Each feels that the results would be more fair, and more effective.

Segal's speech supporting the proposal from the Conservative point of view is particularly interesting. He says, in part:

We have a rare chance here in Ontario to show leadership and take the simple position that every vote counts; that governments should not be chosen by accident; that the electoral system should not guarantee distortion and misrepresentation of how the voters actually voted! We have the chance to modernize, democratize, legitimize and upgrade the democratic process in Canada’s largest, most diverse province.

And then there's Dalton

Premier Dalton McGuinty has made no official statement about the referendum. It is generally assumed that he is aware of it.

For the Proposal to Succeed

Basically, the proposal needs to gather a metric butt load of votes in its favour during the upcoming referendum. The proposal must win at least 60% of voter support overall, as well as approval by a majority of voters in at least 60% of individual electoral districts (ridings). With little public visibility so far, and only weeks to go, this seems unlikely. However, an NDP majority in 1990 was unlikely too, yet it did happen. I hope that any and all Ontarians reading this writeup will vote in the referendum on October 10th.


Of the 107 ridings in Ontario's 2007 election, 60% — 64 ridings — needed a straight majority (50% + 1) to approve MMP. In fact, only five ridings did, all in the provincial capital of Toronto (Davenport, Beaches — East York (by 80 votes), Parkdale — High Park, Toronto — Danforth, and Toronto — Trinity-Spadina). The proposal received 36.8% per cent approval from voters province-wide, with 1,548,234 votes in favour and 2,661,781 votes opposed. This result is almost identical to the approval rating from Prince Edward Island in that province's 2005 referendum on electoral reform.

Many voters reportedly arrived at the polling stations still unaware of the referendum and uninformed about their choices. Overall turnout was only 52.6% of eligible voters, a record low.

Sources and Sites

Log in or registerto write something here or to contact authors.