Gerry Kristianson is a political scientist who has served as a CIDA-sponsored supervisor or observer in several East European elections. He was among a group of international observers at the 1999 Russian election. This is a revised version of a paper presented at the Making Votes Count Conference held in Vancouver in May 2000.

It is commonplace for Canadians to cite our federal and provincial electoral systems as models that should be copied by others.  There probably is not a parliamentarian, past or present, who has not told an audience of students about the virtues of our democratic process, and especially about the way in which our “first past the post” voting system helps ensure strong and decisive majority governments, thereby avoiding the instability of European multi-party systems and coalition governments based on proportional representation. However as few of the new democracies seem to be interested in adopting our electoral system, this article suggests we may want to pause and rethink how good the system really is. 

Our electoral system has many admirable traits.  Having observed or supervised elections in various parts of Eastern Europe, Asia and South America, and having been either a candidate or a campaign worker in more federal, provincial and local elections than I care to remember, I can testify personally to the comfort that comes with knowing that there is no doubt about the impartiality of the people running the process, the secrecy of the ballot, or the accuracy of the count.  We take for granted things that are not assured in many other countries.  We do not have to worry about whether the ballots being counted are the ones that were cast.  No one in Canada has to be afraid that soldiers or police will seize the ballot boxes at the close of polls in a phoney attempt to ensure their “security” until the count can begin.  We do not have to fear that losing incumbents will refuse to leave office, claiming that the election process was fraudulent because they do not like the outcome.

Canadians can take justifiable pride that we were pioneers in the development of free and fair elections.  We were amongst the first to adopt the secret ballot and to extend the franchise to all adult citizens (with some notable exceptions such as Asians and aboriginal Canadians).  It is a testimonial to our system that many emerging democracies have come to Elections Canada and its provincial counterparts for help in setting up their electoral machinery.

These positive features of Canadian representative government should not be allowed to mask the fact that we do not necessarily live in the best of all electoral worlds.  When considering whether there is a need to modernize our electoral system, it is worth asking why there seems to be little enthusiasm elsewhere to adopt the most fundamental element of our election process, the plurality voting system.  We also should ask whether there are positive lessons to be learned from some of the many experiments in democracy that are being conducted in other countries.

The Limitations of the Plurality System

Although Canada and Canadians have been playing a prominent role in efforts to help governments in Eastern Europe and elsewhere develop democratic institutions, few countries seem interested in adopting the most fundamental aspect of our electoral system – the simple plurality or “first-past-the-post” ballot – as a means of selecting legislators.  In virtually every part of the world where new electoral systems are being established, some form of proportional representation is being adopted.  Systems are being put in place to ensure that legislative seats are allocated in proportion to the number of votes received by each political party, rather than simply going to the candidate who wins the most votes.  

Having watched government officials in a number of countries squirm as foreign observers criticized their failure to meet international democratic standards, I can only hope that none of my observer colleagues from other OSCE countries decide to visit British Columbia during our next general election!

When one considers the outcome of recent elections in Canada it is not hard to understand why people who have the opportunity to start fresh might reject our system in favour of some form of proportional representation—or even some variant of the preferential or single-transferable ballot.  When examining the choices available to them, people might well be influenced by the fact that our electorate almost never gets what it voted for.  British Columbia’s current NDP government won a majority of seats in the last election, despite the fact that the Liberals got more votes.  The federal Liberals got a working majority in the House of Commons in 1997 with less than 39% of the votes.  On the other hand, the Progressive Conservatives, while receiving 18% of the votes got only 7% of the seats.  The NDP’s 11% share got it only 7% of the seats, while the Bloc actually got more seats than its electoral support justified.  Reform was the only party to get a number of seats proportional to its share of the popular vote.

These results were not an exception. The fact that our voting system frequently distorts the result in order to give the winning party more seats than is justified by its share of the vote has been cited as one of its virtues.  “Strong” government, by which is meant government in which the executive has unshakeable control of the legislature, has been seen as more important than ensuring that everyone’s vote counts equally.

This disparity between votes and seats is, of course, a reflection of two factors — the uneven distribution of support for different political parties and variations in size between constituencies.  One cannot do anything about the former.  Canadians have a right to choose their place of residence and to decide which party or candidate they support.  It is possible to ensure that individual electoral districts are similar in size, but in British Columbia, at least, no government has ever shown the will to do so.  In fact, the province’s election law provides for a disparity of up to 25% between seats — a number that appears to place the province in breach of Canada’s international obligations as a member of the Organization for Security and Co-operation in Europe.  

During an OSCE conference in Copenhagen in 1990, as part of our efforts to encourage democratic development in Eastern Europe, Canada signed a declaration that emphasized “the central role of elections in securing the citizen’s right to participate in the government of his or her country.”  We agreed to “guarantee universal and equal suffrage to adult citizens.”  As a member of the OSCE we agreed that “the principle of equality requires that one’s vote be given equivalent weight to other voters in order to ensure equal representation.”  We accepted a definition of equality as meaning that in plurality systems like our own, “the size of the electorate among constituencies should not vary by more than approximately ten percent.”

The above statements come from a document that was used by OSCE observer teams in Russia to judge whether that country was meeting the required standards during the recent presidential election campaign. I have to confess that when it arrived on my e-mail in the central Siberian city of Krasnoyarsk I was more than a little embarrassed to find that I had to judge the Russians against a standard that my home province could not meet.

Proportional Representation as an Alternative Voting System

Because of the disparities that seem an inevitable result of first-past-the-post voting, many countries have adopted systems designed to ensure a direct relationship between the number of votes received by a party and its proportion of legislative seats.  I have some personal familiarity with such systems having observed at first hand elections using some variant of proportional representation in Bosnia, Slovakia, Russia and Guyana.  I have watched the votes being cast, been present during the counting process, and observed the process by which seats are allocated.

Simplicity often has been cited as one of the virtues of our plurality system.  The voter marks an X beside the name of the favoured candidate and the person with the most votes wins.  However, from the voter’s perspective, a proportional ballot need not be any more complicated. In its most basic form it can be identical, except that individual names of candidates are removed and only the list of party designations left on the paper. Given evidence that for most voters this is the most important piece of information anyway, the only substantive difference is that the votes get counted on a national or provincial basis rather than constituency-by-constituency.

The counting process for this basic kind of proportional representation ballot is no more complicated than counting plurality ballots. You just sort the ballots into piles for each party and count the number in each pile.  It is true that after the votes have been counted, proportional systems, instead of simply declaring the name of the candidate with the most votes, must conduct the additional step of allocating seats on the basis of voting shares.  There are a number of different ways of doing this, but we can safely assume that all the voters really care about is the bottom line.

Variants of Proportional Representation

Proportional ballots can be made more complicated in order to give the voter greater control over the order in which individual candidates will be declared elected from the party lists.  One has to ask, however, whether this is necessarily a problem?  Given their experience in municipal elections, where they are faced with a long list of names, the right to vote for more than one candidate, and no party designation on the ballot, I think we can assume that most Canadian voters would not have much trouble dealing with even the most complicated proportional ballot.

What are some of the variants of proportional representation? Some proportional systems allow the voter to vote for a party and then to express one or more preferences within a party list.  In Slovakia, for example, the voter can give up to four names a special preference by circling them.  Any candidate receiving more than 10% of the preference votes expressed for his party is assigned a seat first, starting with the one with the highest number.  

Some proportional systems allow the voter even more power to determine the order in which individual candidates will be elected.  The voter selects a party list and then votes preferentially within the list.

Other proportional systems do not give the voter the luxury of helping to determine who actually gets to hold a seat.  In Russia, for example, the names of successful individuals are simply selected from the list in the order in which the parties have presented them.  Unlike Slovakia, the lists of names do not appear on the ballots, although they appear on large posters at the entrance to the polling station.

Most proportional systems are based on giving the voter a choice between party lists, but some also allow a vote for individual independent candidates. In municipal elections in Bosnia, for example, the ballot lists parties, coalitions and independent candidates.

Whatever their individual differences, however, all proportional systems have in common the intent of trying to ensure that everyone’s vote contributes to the election of a legislative representative and that parties hold legislative power in direct proportion to their level of voter support.

By their nature, proportional systems cannot be based on single member constituencies.  Elected members do not acquire specific geographic responsibilities unless assigned to them by their parties after election.  In most examples, a single list applies to the whole country or jurisdiction, although there are variants in which one counts the votes proportionally for lists that apply to a region.  In this case, the only requirement is that the number of representatives from each district must be proportional to the size of the whole electorate.  Such a system could be used in a place like British Columbia, based on natural regions like Vancouver Island, the lower Mainland, and do on. 

Despite the basic attempt to ensure a direct relationship between the number of votes received by a party and its share of legislative seats, most proportional systems require a minimum level of support before a party can participate in the sharing of seats.  This number usually is 5%, but can be less.  For example, in elections for the Russian Duma, if the cumulative total of votes cast for all parties passing the 5% threshold is less than 50% of the total votes cast, then parties gaining 3% also share in the allocation of seats.  In Slovakia, if no party makes the 5% cut-off, then the barrier is lowered to 4% for everyone.

The minimum vote requirement can be criticized as discriminatory in the same way that one can criticize a legislature’s use of a minimum size rule to determine “official” party status.  On the other hand, the minimum can be defended as a means of encouraging coalition building.  In Slovakia, for example, the election law allows parties to submit separate lists but to register as a coalition so that they can add their votes together for the purpose of allocating seats.  However, since the Slovak law also states that both parties have to achieve the 5% threshold, there was no incentive to take advantage of this provision during the 1998 election.  On the other hand, having recognized the problem posed by the 5% barrier, two groups of parties decided to register under a single new name that included the word “coalition”.

A desire to combine the benefits of legislative responsibility for a specified geographic area with the virtues of proportional representation has led to some mixed systems.  In Russia, for example, half of the Duma’s 450 seats are allocated on the basis of proportional representation, as described above, and half are single member constituencies with the seat going, as in Canada, to the candidate who gets the most votes. In the latter case, the winner does not need an absolute majority, but does have to get more votes then have been cast for “none of the above”.  The ability to vote against all of the candidates on the list provides an interesting added dimension to the Russian election process.  In the December 1999 Russian election voters rejected all candidates in eight seats—thereby requiring by-elections.  

The Russian attempt to combine the virtues of proportional and plurality systems does seem to have had the effect of facilitating the election of formally independent candidates.  Judging from the results, many Russian voters seem to take advantage of the mixed system to cast one vote for a party list and the second for a candidate not affiliated with that or any other group.  

I am particularly intrigued by the Russian provision of a place on the ballot on which to reject all candidates. In the single-member races for legislative and executive positions in Russia, the winning candidate has to get more votes than are marked for “none of the above”.  A cynic might suggest that it would be dangerous to introduce such a provision in Canada.  Given current attitudes towards politics and government, we might find that no one could get elected!

If we are not going to move to proportional representation, we might want to consider election rules which require majority support before one can be elected.  While the example is less relevant to Canadian circumstances, Russia, along with a number of European democracies has adopted the run-off system for its presidential elections.  The winning candidate must have an absolute majority of support.   A second election is held between the two leading candidates if no one gets a majority on the first round.   Since a second round of voting has obvious expense implications, one can accomplish the same objective by moving to the preferential or single transferable ballot, as used in British Columbia during the 1952 and 1953 elections.

A run-off requirement also was being imposed on a provincial level by-election during my recent visit to Russia.  In addition to the presidential vote, people in the Krasnoyarsk suburb of Leninski were being asked to vote for someone to fill a vacancy in the regional legislature.  Since no one got an absolute majority on March 26, a run-off election between the two leading candidates had to be held.

While turn-out has not been a particular problem in Canadian provincial and federal elections, it is interesting to note that many foreign jurisdictions require a 50% turn-out before an election can be considered valid.  In Russia, for example, this rule applies to all elections.  Such a provision would have an interesting impact on our municipal elections.

In addition to the form of the ballot, there are other areas where we might consider change based on examples elsewhere.

Even though our system of appointing impartial returning officers has worked reasonably well I rather like the way in which a number of Eastern European jurisdictions require the equivalent of our returning officer to work with a formal committee or commission composed of party or candidate representatives as well as non-partisan people.  This helps to ensure that no one feels isolated from decisions affecting the mechanics of the electoral process.

We also might learn from efforts elsewhere to ensure that every citizen has the opportunity to vote. Given the mobile nature of our society, there will always be people whose names do not appear on the current voters list.  Many jurisdictions are more flexible than either BC or Canada in allowing voters to register, as in the Russian case, “to the beginning of the calculation of votes”, in other words, until the polls close.  Voters are routinely added to the list throughout voting day, upon presentation of appropriate identification.  It may seem somewhat perverse that the requirements of what used to be totalitarian police states, such as the need for every citizen to carry photo identification, actually facilitates the democratic election process.

It also must be said that the use of proportional representation facilitates election-day registration in the sense that when voting for party lists rather than for individual constituency candidates, the place of local residence becomes less relevant to determining a person’s right to vote.

The use of portable ballot boxes is another election innovation that might be considered.  In both Slovakia and Russia I have accompanied election officials as they carried a small ballot box to the homes of elderly or handicapped voters.  At the close of polls, after the total number of ballots in these boxes had been reconciled with the list of voters, the contents were mixed with those from the larger polling station boxes so that voting secrecy was assured.

In closing, let me repeat that I do not intend this discussion of alternatives to our voting system to imply that our current electoral systems at both provincial and federal levels do not have some strong and important virtues. But it is these very strengths that should allow us to consider the need for change.  We need to keep asking ourselves the fundamental question.  As they set out to create a new democratic process, why is it that so few countries want to adopt our ballot system?