Skip to content

Proportional Representation Definition Example Essay


Full Representation:

Proportional Systems Promote Inclusion, Deliberation and Better Policy

The Center for Voting and Democracy


President Bill Clinton began his second term in 1997 with an optimistic inaugural address in which he urged Americans to "keep our old democracy forever young." Among our nation's foremost challenges, he warned, will be "the divide of race," but that: "Our rich texture of racial, religion and political diversity will be a godsend in the 21st century. Great rewards will come to those who can live together, learn together, work together, forge new ties that bind together." Few Americans would dispute this vision, but it is not merely individual attitudes that govern the tolerance of diversity in our communities. Institutions and their rules play a major role in relations among people, and one of the most significant rules in a community is the one determining how citizens can win and sustain legislative representation and a fair share of power in a competitive electoral environment. Just as consumer choice and buying power are the foundation of a free market economy, citizen choices and voting power are at the foundation of a responsive and inclusive democracy. The rules governing citizens' choices and voting power have a great impact on who runs, who votes and who wins. Unfortunately, most American cities use antiquated "winner-take-all" rules that too often divide us and undercut accountability. Despite the president's charge to keep our democracy young, we all too easily accept these electoral rules simply because we inherited them. Indeed, President Clinton himself in 1993 cut short a healthy national debate about our general use of winner-take-all elections when he withdrew his nomination of University of Pennsylvania law professor Lani Guinier to head the civil rights division of the Justice Department. To increase turnout and diversify representation Guinier had proposed consideration of proportional and semi-proportional representation voting systems and even more challenging ideas about ensuring minority influence in legislative bodies. Even though most mature democracies use proportional systems and even though the Justice Department under Presidents Reagan and Bush had upheld their adoption in many communities, she was demonized as a "quota queen" and never given a chance to explain her writings in a congressional hearing. Recent city elections show that Guinier's concerns about representation and participation are valid. Turnout continued its plunge in elections around the country this year. In New York City's Democratic primary -- a primary that decided mayoral nominees and the de facto winners for most city council seats -- turnout among registered Democrats was only 18%. Detroit's mayoral primary turnout was 17%; in Charlotte, it was 6.4%. General election turnout was under 40% in Miami and New York City and under 30% in Boston. Incumbents generally won in walkover elections, with one party holding near-monopoly control in many city councils. Our city halls are more racially diverse than they were a generation ago -- in part because of demographic changes, in part because of implementation of the Voting Rights Act and in part because of decreasing racism. Now the Supreme Court has put severe limitations on traditional methods of increasing representation of black, Latino, Asian and Native American voters through redistricting wards to have majorities of targeted minority voters. Many cities may have less diverse representatives after the next redistricting in 2001. Regardless, they could face expensive legal battles from plaintiffs on both sides of the controversy. But even in ward elections drawn to encourage representation of racial minorities, winner-take-all rules put geographic straitjackets on diversity and restrict accountability by limiting competition. In contrast, non-winner-take-all voting systems -- historically called "proportional representation," but perhaps better understood as "full representation" -- promote a more modern, cosmopolitan vision of a city. Representatives are more likely to emerge from communities of interest than personal ambition. At the same time, the major political forces are more likely to support candidates that represent these different communities such that a political force's slate of candidates fully represents the "big tent" of voters from whom it seeks support. As a result, city councils are more likely to represent a "gorgeous mosaic" of overlapping interests and groups. Cities become all the stronger and more stable by giving diverse communities real incentives to participate and realistic access to the making of public policy. Electing the candidate who wins the most votes in a given area reflects a crude understanding of elections. True, winner-take-all elections were how the early democracies held elections, but their serious limitations have led most democracies to reject them. There is a range of proportional voting systems -- some candidate-based and some party-based, some with a mix of wards, some in relatively small multi-seat districts and some citywide ---- but nearly every political jurisdiction in the United States rather blindly follows traditional winner-take-all models. As we head toward the 21st century, it is high time that we re-examine our 18th-century electoral rules.


What is Proportional Representation?

Proportional representation (PR) is a principle of representative democracy, not a particular voting mechanism. Most mature democracies use forms of PR, although they vary widely in the threshold of votes necessary to win representation and the relative role of political parties: important differences that mean PR cannot be judged by its performance in any one nation or city. Of the 36 nations with at least two million people and a top rating from the human rights organization Freedom House, 30 use PR systems for their central national legislature; only the United States, Canada and Jamaica exclusively use winner-take-all elections for all national elections. PR can have a positive impact on campaign conduct, the fairness of representation, voter turnout, the role of money in politics and governance. The principle of proportional representation is that groups of like-minded voters (partisan or non-partisan) should win seats in legislative assemblies in proportion to their share of the popular vote. With PR the majority wins its right to decide, but a minority wins its fair share at the table of representation. Thus, in a city council election for nine seats, a political force with at least ten percent of voters throughout the city should earn one seat. A force with 51% support should earn five seats, and so on. Another way to understand PR is that most voters will elect a candidate of their choice: the more voters who have the opportunity to elect candidates of choice, the "fuller" the results will be. In contrast, U.S.-style winner-take-all elections allow a majority (or even a simple plurality) of voters in a given geographically-defined district -- a district usually created to achieve certain political results -- to win all the representation for that area. In an at-large, winner-take-all election, one group of voters can elect nearly all the winners. Adoption of single-member ward elections may break up a citywide majority, but it simply transfers distorted representation down to a neighborhood level. When one winner "takes all" in a ward election, 51% of voters (and less when there are more than two strong candidates) win the right to speak for the other 49%. Proportional representation systems can be party-based, as in much of Europe and in South Africa in its 1994 elections. They also can be completely non-partisan, as in city council elections in Cambridge, Massachusetts, and Peoria, Illinois. Historically, the National Civic League has most strongly supported "the single transferable vote" -- a system also known as "preference voting" and "choice voting" because voters rank candidates in order of preference. Choice voting was included in the League's model city charter for much of this century, and at one time was used for city council elections in New York, Cincinnati and Cleveland. Other non-winner-take-all systems currently used in the United States are cumulative voting and limited voting (see sidebar).


General Benefits of Proportional Representation

Comparative political scientists have attributed a number of benefits to use of proportional representation (PR). Among them are:

* Voter participation: Voter turnout is generally 10% to 15% higher in nations that have PR than in similar nations using plurality elections. This difference is logical. In the United States, a majority of legislative elections are not competitive. The average margin of victory in U.S. House elections is consistently over 30%. Usually, the lower the level of election, the lower the level of competition and participation. One-third of state legislative elections consistently do not draw even two major party candidate. Most cities have little real competition in council races; in 1997, only one of New York City's 51 city council seats was closer than a 10% victory margin, and all 45 Democratic winners received "landslide" percentages of over 60%. The sad reality is that voters in non-competitive wards -- whether in the majority or in the minority -- might better use their time and resources to send a check to candidates they like in more competitive races rather than vote in their own. In PR systems, winning fair representation is dependent on voter turnout. When nearly every vote will help a party win more seats -- regardless of a party's level of support in a particular area -- voters have more incentive to participate, and parties have every incentive to mobilize their supporters. Perhaps just as importantly, parties and other electoral organizations have every incentive to keep their supporters informed in order to hold onto their support; studies show that informed citizens are more likely to vote.

* Fair minority representation: When winning seats does not require a majority of the vote, minorities of all kinds by definition have a better opportunity to elect candidates of their choice. Indeed, the history of proportional representation around the world and when used in the United States is an excellent one for representation of minorities. And fair minority representation means better majority representation because together different minorities can constitute a majority. In addition to winning a fair share of seats, minorities also have greater opportunities to negotiate for influence, since they have more options for whom to vote. When South Africa held elections using PR in 1994, the two leading parties ran multi-racial slates with messages of inclusion. When New Zealand had its first PR election in 1996, the first Asian citizen was elected, and Pacific Islanders and indigenous Maori tripled their representation. In addition, a Maori-backed party formed a coalition government with the governing party -- a party that in recent years had had few Maori representatives, in a way analogous to the Republican Party's relationship with blacks in the United States. PR has the twin benefits of encouraging minority communities to mobilize and giving them access to power. From 1925 to 1955, Cincinnati used the choice voting form of PR to elect a nine-seat city council. In 1929, when blacks were barely 10% of the population, a black independent candidate ran a strong campaign. In the next election, he was added to the Republican party's slate and was elected. In 1947, when blacks were 15% of the population, a former president of the Cincinnati NAACP ran in large part to defend the choice voting system that was under attack from Republicans seeking to restore their old domination of the council. In a recognition that any substantial group of voters could not be ignored, the other major slate (the Charter-Democrats) added him to their slate in 1949. He was elected, resulting in black representatives holding two of nine seats for the next four years. For the remaining choice voting elections, both parties competed for the black vote.

* More women in office: The percentage of women elected to office in the United States -- only 11% of the U.S. Congress -- is scandalously low, particularly in light of the relative strength of the American women's movement compared to other nations and particularly when studies show that women legislators do often provide substantively different representation. Although the connection is less conclusive for local elections, studies show that women in state legislative elections win seats in significantly higher percentages in multi-seat districts than in one-seat districts -- double in some states with a mix of systems. The major reasons for this difference are that women are more likely to run and voters are more likely to seek gender balance when there is more than one seat to fill. PR systems give women additional leverage to force the major parties to support more women candidates because women have an opportunity to vote for smaller, more women-friendly parties. A threat to do so by women supporters of the major parties in Sweden in 1994 led to an increase of women in the legislature to 41%. New Zealand and Germany are among a growing number of democracies that use systems with a mix of district and PR seats. Instructively, women are three times more likely to win seats elected by PR than elected in one-seat districts. In 1996 in New Zealand, women won 45% of PR seats and 15% of districts seats; in 1994 in Germany, women won 39% of PR seats and only 13% of district seats.

* Elimination of gerrymandering: Drawing district lines to influence who wins has taken place virtually from the first redistricting -- the term "gerrymander" refers to a Massachusetts district plan drawn in 1815. But gerrymandering has become far more potent in an era of powerful computers, more detailed census information and better techniques for measuring voter preferences. As one example, Democrats in control of the redistricting process in Texas in 1991 placed the eight Republican incumbents in districts that were among the most conservative in the nation. These incumbents were easily re-elected in 1992, but Democrats won 21 of the remaining 22 seats with only 50% of the statewide vote. Only one race was won by less than 10%, and the three open seats went to state legislators serving on redistricting committees. Congresswoman Eddie Bernice Johnson, the primary architect of the plan, admitted in 1997 that the redistricting process "is not one of kindness. It is not one of sharing. It is a power grab." Although political intentions can be removed from the redistricting process, as in Iowa's process, political results are unavoidable -- some districts inevitably will be non-competitive. Gerrymandering of any sort is much more difficult with PR systems. The fewer the percentage of votes that can be "wasted" on losing candidates -- 49% in a winner-take-all race, but less than 20% in a five-seat PR election -- the more likely the voters will be the ones who choose their representatives rather than legislators choosing their constituents through gerrymandering.

* Governance from the center with representation of the wings: The core principle of PR is that a majority should decide a policy issue after a debate of the whole. Like a town meeting, debate should include as many voices as possible without disrupting efficiency. Unlike many town meetings -- where policies are often debated and decided in one night -- this debate takes place in the context of a deliberative, legislative process in which issues are fully discussed and opportunities to negotiate are plentiful. One clear difference between PR and most winner-take-all elections is that in PR election, both major parties -- and perhaps smaller parties -- likely will win representation from the same geographic area. When the major parties co-exist in a given constituency, all constituents have access to representatives with both the governing party and with the main opposition party. No geographic area -- or type of area, as is now true of American urban areas that have few Republican representatives -- is likely to be isolated politically. The Chicago Tribune in 1995 editorialized for the return of the non-winner-take-all system of cumulative voting in three-seat districts for state legislative elections (see sidebar). Tribune editors wrote:

"[Cumulative voting] guaranteed the relative strengths of the two parties would be reflected in the legislature, but every region of the state would also have substantial representation in each party's caucus.... The intermeshing of political and regional interests has all but disappeared in the 15 years since cumulative voting was abolished.... [M]any partisans and political independents have looked back wistfully at the era of cumulative voting. They acknowledge that it produced some of the best and brightest in Illinois politics."

The Tribune's remarks point to two other important facts. First, even a limited modification of winner-take-all voting -- Illinois' three-seat districts still left a high threshold of 25% necessary to win seats -- is likely to have important benefits. Second, "third" parties and independent political forces often play a constructive role in governance and campaigns. A two-party system can be very polarizing, with each side playing "zero sum" confrontational tactics founded on the fact that loss of support for one party has nowhere to go but to the other. Having more choices across the spectrum can break down that polarization, and allow a governance that more consistently reflects majority interests. A particularly revealing April 1994 essay in World Politics describes the "Proportionate Influence Vision" of democracy, in which "elections are designed to produce legislatures that reflect the preferences of all citizens." The article contrasts this vision with the "Majority Control Vision" -- one in which "democratic elections are designed to create strong, single-party majority governments that are essentially unconstrained by other parties in the policy-making process." In their statistical comparison of 12 democracies in Europe regarding how citizen preferences are translated into public policy, they concluded that "The governments in the Proportionate Influence systems are on average significantly closer to their median voter than are governments in the Majority Control and Mixed systems....If voters are presented with a wide range of choices and electoral outcomes are proportional, governments tend to be closer to the median." In short, governance is more likely to take place at the center, but fair representation of the wings provides an ongoing mechanism to shift this center and transform governance. Opposition voices will be heard, and their ideas are far more likely to be debated. If their ideas draw growing attention, the major parties will adjust accordingly in order to hold onto their supporters.

* Expanding discourse: Winner-take-all elections -- especially in the present era of attack ads and focus groups -- can make it extremely difficult to have reasoned political debate on certain contentious issues. These issues can take on great symbolic weight for swing voters, who ironically gain the most electoral influence in our system by being among the relatively few who are detached from regular support of either party. At this point in our political history, for example, it is unlikely that a non-incumbent could run a credible campaign for president or most statewide offices with a position against the death penalty -- certainly a reasonable position, even if arguable -- which has come to represent being "tough" on crime. It is difficult for candidates to take nuanced positions on a range of issues, from drug policy to abortion rights to welfare reform. This freezing of debate makes it all the harder to change policy in the future. When legislative candidates can be assured of winning despite garnering less than 50% of the vote, it is far easier to have a full and serious dialogue on such issues in campaigns and in legislatures -- and thus in the public realm as a whole, since the major media often works within the confines of the "legitimate" positions of currently elected officials. The value of calmer, more reasoned voices perhaps can be symbolized by the movie "Twelve Angry Men," in which Henry Fonda's quiet skepticism ultimately results in reversal of the verdict of the eleven other jurors. This opening of discourse is important not only at a national and state level, within given legislative districts. One of the reasons that many districts stay "safe" for one party is that political debate within these districts can be stifled. No one takes campaigns in the district seriously, and the second-place party often writes off the district and puts few resources there. Franked mail from the legislator offers only one viewpoint. Opposition views are simply not heard or largely ignored.


How American Electoral Rules Developed

The American founders were at the forefront of intellectual and scientific thought in the 18th century. The Constitution was carefully crafted, based on a mix of reasoned debate, empirical study and states' competing interests. Yet the Constitution is silent on methods of election for the U.S. Congress. Proportional representation was not discussed for the very simple reason that no mechanism had been developed to allow voters in the minority to elect candidates. The principle was touted -- indeed, John Adams wrote that "[American legislatures] should be an exact portrait, in miniature, of the people at large, as it should think, feel, reason, and act like them." But the only voting systems available were both winner-take-all: elections in multi-seat districts -- usually at-large -- and ones in single-member districts. Most states used statewide elections to elect their U.S. House delegations during the 1790s. Gradually, more and more states moved to district elections. One reason was to guarantee local representation, but for many the move was driven by an interest in better representing diversity. Statewide elections tended to allow one political party's candidates to win all the seats, diluting the votes of those in the minority. Many of these elections were quite close, but one party would win all. In 1832 and 1834, for example, the Democrats swept New Jersey's six at-large U.S. House seats despite barely 1,000 votes separating the top Democrat from the lowest Republican in each election. Reformers sought district elections to diversify representation. In retrospect, proportional representation systems would have been the sensible approach, but the first articles detailing mechanisms of PR were not published until the mid-1840s and not widely circulated until John Stuart Mill's advocacy in the 1860s. The most influential early advocate of PR in the United States was Charles Buckalew, a U.S. Senator from Pennsylva-nia, 1865-1871. Sen. Buckalew's pro-posals gained significant sup-port in Congress, and he played a central role in the adoption of cumu-lative voting in several cities and in Illinois for state legislative elec-tions (see sidebar). But earlier in his career he had been an advocate of district elections. In an 1867 speech Buckalew sheds light on the motivations of district reformers and what might have happened if they had known about PR systems:

"Our experience in this State and in other States is not in favor of carrying the idea of single districts very far. I drew the amendment to the Constitution of our State [of Pennsylvania] by which your city is bro-ken into [sin-gle member] districts. [Ap-plause] What was the idea of that amendment? It was that one political interest should not absorb the whole sixteen or eighteen representatives you send to the Legislature; that a little shift-ing majority one way or the other should not cast that large number of votes on one side or the other at Harrisburg. "The idea was to break up the politi-cal commu-nity, and allow the different political interests which compose it, by choosing in single districts, to be repre-sented in the Legislature of the State. Unfortu-nate-ly, when that arrangement was made for your city (and for Pitts-burgh also, to which it will soon apply), this just, equal, almost perfect system of voting [proportional voting], which I have spoken of tonight, was un-known; it had not then been an-nounced abroad or considered here, and we did what best we could."


Cities Swinging Between At-Large and Wards

Although most congressional elections have used districts since 1842, when the first law requiring district elections was adopted, cities have swung back and forth between at-large and ward elections. As discussed by Buckalew, major cities largely moved to ward elections in the 19th century, but by the turn of the century, many leaders in the progressive movement sought citywide, at-large elections in order to break up political "machines" that dominated many cities. The shift back to at-large elections occurred after proportional representation systems were developed, and indeed many leading reformers advocated PR rather than winner-take-all elections. The National Municipal League, Walter Lippman, A. Philip Randolph, Murray Seasongood, Fiorello La Guardia and League of Women Voters founder Carrie Chapman Catt were among many influential advocates of PR, more specifically the choice voting system that was adopted in numerous cities at that time. Unfortunately, PR proved more controversial than other elements of the municipal reform package-- perhaps because of its more decisive impact on democratizing power. Although choice voting was adopted in several major cities and upheld in 24 of the first 26 repeal attempts around the nation, it encountered increasing resistance. New York City's repeal of choice voting in 1947, after two previously failed attempts, set off a wave of repeals that by 1961 left Cambridge, Mass. as the only city with the system. The very strength of choice voting was a weakness in the post-Cold War period: opponents could pick on unpopular minorities like leftists and blacks to convince a majority of voters to reverse their previous support for the system. Black voters in cities like New York, Cincinnati and Toledo voted to keep choice voting, but were out-numbered. Since the 1960s the national pendulum has swung back toward ward elections, but this time without nearly as much debate over PR. The new wave of reformers saw the early 20th century municipal reformers as elitists, more concerned about downtown business interests than those of neighborhoods and minorities. Residential segregation by race in many cities made ward elections an obvious solution for those interested in racial diversity on councils. The allure of guaranteed representation of all parts of a city was powerful, even if historical memory about the problems of ward elections in cities was short and neglect of the problems with district elections at other levels of government was troubling. Indeed, ward elections have been disappointing to many in such cities as Oakland and Boston, but more cities keep adopting wards out of frustration with winner-take-all, at-large systems.


The Modern Case for Proportional Representation in Cities

Proportional representation allows cities to have the best of ward elections and the best of at-large elections. In contrast to at-large elections, diverse voters can win their fair share of representation. Campaigns are less expensive because it takes fewer votes to win, and candidates can choose to focus their campaigns on particular constituencies. In contrast to ward elections, however, those seeking representation are not required to be geographically concentrated. Representatives can work side by side with the same constituents, and citywide policy is less likely to be left to mayors and city managers. There are particularly good reasons to consider PR systems as we head into a new century of increasing diversity and residential dispersion of minorities in cities. Several cities had bitter and expensive battles over drawing single-member district lines after the 1990 census. In cities like New York, Oakland, Los Angeles and Chicago, blacks, Latinos, and whites fought over redistricting; the resulting court battle in Chicago cost taxpayers $10 million to defend a ward plan despite the fact that its intended impact was to protect incumbents and stifle competition. Battles over district lines promise to be only more contentious after the year 2000, both because of rising diversity (with Asian Americans having a significantly greater presence in many cities), decreasing residential segregation and recent Supreme Court rulings that make "creative" redistricting all the more difficult. Already some minority leaders are re-thinking reliance on wards. When faced with the choice between ward elections and choice voting in San Francisco in 1996, the Mexican-American Legal Defense and Education Fund supported choice voting. Austin's at-large system is under review, and several leading black political leaders have expressed support for choice voting rather than wards. While fair minority representation may be the most immediate reason to consider proportional systems, there are additional compelling arguments. One is that cities need innovative leadership. Too many cities have static elections in which voters have little ability to vote for change -- or if they do vote for change, it is greater than they might want. Ward elections also can keep most representatives focused on ward issues, leaving citywide policy-making more to mayors and city managers who by definition cannot be fully representative of the community. Proportional systems certainly can open a closed political system. In 1935, Democrats backed by Tammany Hall won 62 of 65 city council seats in New York City. After adopting choice voting in 1936, Democrats barely won a majority. Although maintaining their majority in five choice voting elections, they faced serious competition from four other parties and from reformers from within. After restoring ward elections, the Democrats won all but one seat in the first elections with districts in 1949. Democrats have dominated council elections ever since. Many municipal reformers for understandable reasons also are seeking campaign finance reforms. But taking the money out is only an initial step. One also must put the people in, and the full representation provided by PR systems enables people to come together, organize, voice their interests and have a fair chance to win and sustain representation. They create a new space for community organization and independent representation.


A Growing Movement

Grassroots activity in support of proportional representation has grown dramatically in recent years. At a national level, Rep. Cynthia McKinney (D-GA) has introduced the Voters' Choice Act to allow PR systems for congressional elections. At a news conference when the bill was first introduced in 1995, speakers included the directors of the National Women's Political Caucus, U.S. Term Limits and the Committee for the Study of the American Electorate. Grassroots activists and civic leaders are raising the idea of PR in numerous states and cities. The black caucus in Georgia's state legislature has introduced bills to adopt PR for its congressional elections and held hearings around the state in the fall of 1997. An elections task force in North Carolina gave bi-partisan support to a bill to allow localities to adopt PR systems. Texas in 1995 approved allowing school districts to use PR systems, and more than 40 Texas localities have adopted cumulative voting to settle voting rights suits. The ACLU of Washington has adopted a policy in support of lifting all legal restrictions on using PR for local, state and federal elections. Boston's leading black elected officials support PR for city council elections, the Center for Voting and Democracy has been asked to testify about PR before charter commissions and task forces in such localities as Cincinnati, Detroit, Miami Beach, Nassau County and San Francisco. Activists are planning initiative campaigns for PR for state or local elections in Alaska, Arizona, California, Colorado, Hawaii, Illinois, Maine, Massachusetts, Michigan, Minnesota, New Mexico, New York, Ohio, Oregon, Rhode Island, Washington and Washington, D.C. Such initiative campaigns have real opportunities for success. In Cincinnati in 1988 and 1991, under-funded initiatives to restore choice voting for city council elections each gained 45% of the vote. After a two-year study in San Francisco, a task force recommended that choice voting replace the city's winner-take-all, at-large system. The Board of Supervisors in July 1996 voted 10-1 to place choice voting on the November ballot. It also voted 7-4 to put single-member districts on the ballot. San Francisco would have been the first American city to adopt a PR system by popular vote since the 1950s. The community coalition that rallied around choice voting indicate where support may be found for future PR efforts. Choice voting picked up early support in communities of color -- several leaders in these communities recognized that their voting power would be enhanced by choice voting more than by ward elections. Choice voting won the endorsement of the Democratic Party, the San Francisco Examiner, MALDEF, NOW, the largest Bay Area labor unions (including SEIU Locals 790 and 250, the ILWU and HERE), Mayor Willie Brown and leading city organizations representing tenants, environmentalists, Asians, gays and lesbians. In the election, choice voting was defeated 56%-44%, and wards were adopted by 57%-43%. Exit polls revealed that choice voting won more than 75% support from black votes and a higher percentage of support from Latinos, Asian Americans and self-identified liberals than did the ward proposal. Wards won due to greater support from moderates and conservatives who were less influenced by the mostly liberal-leaning endorsements. The quickly-organized campaign for choice voting was able to win over many community leaders and organizations, but its budget of less than $30,000 was not enough to reach enough of the city's 600,000 eligible voters in a short campaign for an idea that was new to most San Franciscans and received limited media coverage.


Building a Democracy for the 21st Century

In the language of Alvin Toffler, we have entered a third wave, an information era in which old technologies and institutions may be quickly superseded by new institutions. Yet we grasp desperately onto a conception of representation that has decreasing relevance to our society. Our communities of interest are increasingly non-geographic, but more cities keep moving to council representation based solely on geography -- a move that tends to decrease the citywide influence of the one elected institution in which representation of diversity and real deliberation is possible. The growing diversity of our cities is forced into wards that represent diversity well only when communities stay segregated and racial and ethnic groups stay uniform. Proportional representation holds the promise of representing existing diversity while at the same time encouraging new political forces to develop, voice their interests and earn a place at the table. PR certainly is a way out of the legal and political battles over redistricting, but more fundamentally, it is about providing "universal coverage" for minority representation in a manner analogous to how Social Security protects low-income seniors through helping all seniors. "Everybody wins" sounds too good to be true, but it is the logic of a proportional system. With all substantial political forces winning a fair share of representation and with parties in power likely to reach out to include candidates from these forces, policy-making will more naturally reflect the united will of the community. Any efforts to bring people together in a community to solve problems will be reinforced by ensuring that most of these people having strong representation in elected government. New rules are never the answer in themselves. But they create the foundation from which to build, the form into which people can provide content. As they confront sinking participation, struggles over a shrinking tax base and controversies over fair representation, cities have a great opportunity going into a new century: the opportunity to consider a full range of democratic reforms to enable their people to debate and make policy to build healthy communities. The fundamentally fair level playing field of full and proportional representation will be an essential part of any reform package.


The Center for Voting and Democracy 6930 Carroll Ave. Suite 901 Takoma Park, MD 20912 (301) 270-4616,,

Politics watch

Return to the Blog

Proportional Representation:

An argument for representative democracy, and a refutation of its opposition

By Ryan Ringer

I. Introduction
II. The Logic Behind It
III. The Arguments Against
IV. The Arguments For
V. Some Obvious Injustices
VI. What PR Should Be
VII. The Options
VIII. Conclusion


An idea which is floated around with increasing frequency in the political debate is proportional representation - the idea that a party's representation in parliament ought to be roughly equivalent to the number of votes it receives - and it's not difficult to see why. Voter cynicism and apathy is at an all time high; people think that their vote doesn't mean anything, as most ridings in the country are "safe seats" in one way or another, meaning that people can vote for whomever they wish, but the outcome is predetermined. Imagine a Liberal-minded person in Medicine Hat, a Conservative-type in Toronto--Danforth, a federalist in Quebec City, a socialist in Calgary Southwest, or a Green supporter, well, anywhere. The point is, in many if not most ridings, a vote for the party of your choice means absolutely nothing - and people are catching onto this fact. They are catching onto the fact that Canada is one of the only industrial democracies left - the list includes only the United States, Britain and ourselves - which still uses a voting system which, while appropriate for its time, has become antiquated. They are catching onto the fact that it is broken.

The brokenness of our current system is even being unintentionally acknowledged by Paul Martin and more intentionally by Jack Layton, both of whom are urging potential voters to vote strategically - Paul Martin saying vote Liberal to stop the Tories, Jack Layton saying vote NDP to do the same. Buzz Hargrove's endorsement of Paul Martin came with the condition that people should vote NDP in ridings where that party is ahead of the Liberals, and vice versa.

Strategic voting is the order of the day - and it is, simply put, anathema to democracy. Nobody should be forced into the choice of supporting a party they detest in order to prevent a party they detest even more from winning. That is not democracy, not really. Voters should not be forced to play games with their ballot, basing their vote on guesswork and mathematical approximations of who stands the best chance of defeating their feared candidate. In democracy, voters should be allowed to vote for the candidate of their choice, without fear of any nasty surprises, like waking up the next morning with a newly minted MP from their nightmare party, which could have been prevented had they only guessed better. Voting should not be a game of guesswork, it should be a matter of principle.


From a logical standpoint, it makes sense, which is why so many people wonder why it is not already the status quo. It seems elementary that a party's representation in the legislature should be determined by, and proportionate to the number of votes they received.

It seems like basic logic that a party which only receives 37% of the vote - as Jean Chretien's Liberals did in 1997 and Bob Rae's New Democrats did in 1990 - should not be rewarded with literally 100% of the power, completely unchecked and unaccountable to anyone until the next election.

It would seem that it ought to be a matter of course that Brian Mulroney should not have been allowed to impose a hated and reviled tax - the GST - on Canadians in 1991 when he had only received a mandate from 45% of the people in the previous election, and the two parties for whom the other 55% had voted opposed it.

It most certainly seems logical that a party which loses the popular vote should not win the election, and yet Joe Clark (1979), John Diefenbaker (1957), Glen Clark (1996) and, incidentally George W. Bush down south (2000), did just that: with their closest opponents actually receiving more votes than they did, they still formed the government - and in the case of Glen Clark, formed a majority government!

Fundamentally, there is nothing flawed with this logic. Conversely, there is a great deal flawed with "logic" which states that a party which receives support from fewer than two in five people should have the unchecked and absolute power to do whatever they please - this is what our system allows, however, and it allows it routinely, so much so that it is mundane, masking the abuse of democracy that it truly is. From a purely logical point of view, there seems something deeply flawed with such a system.

Yet there are, incredibly, people who oppose such logic. Why they do so varies - many of them value stability over democracy; others fear a hostile takeover of government by fringe lunatics; others still are merely partisan hacks who oppose any changes to the current system on the grounds that it would reduce the power of their party, which has benefited from a system which disenfranchises entire swaths of society, and thus are loathe to change it, offering up any argument they can muster against it. This last group is particularly bothersome; the first two can be dealt with through a calm refutation of their arguments. The last group, of which I am sad to say I suspect our prime minister of being a member, can only be dealt with in the way one must always deal with party hacks - by simply exposing them for what they are, acknowledging and accepting that they will always put party above principle, and moving on.


There are many arguments against it. The goal here is to refute them all, at least the most prominent ones.

1) Proportional representation usually guarantees minority governments.

This is not so much an argument against proportional representation as it is a statement of fact. The argument against comes from the value judgment, "minority governments are bad". To this the reply is simple: Canadians, if given a choice, would prefer to be represented by a minority government, regardless of whether they support it, than by a majority government they oppose. If offered a choice between a Paul Martin majority of a Stephen Harper minority, Canadians overwhelmingly choose the latter, and vice versa. Majority governments, quite simply, do not allow for any meaningful opposition. The Conservatives, for example, with a mere 40% of the popular vote, could form a fairly sizable majority government, and proceed to introduce legislation that 60% of the population would find anathema, for instance, to criminalize abortion - and nobody would be able to do a thing to stop them. There is a fatal flaw in a system which purports to be democratic, yet would allow that to take place. Given the choice between minority or majority, most Canadians would clearly prefer the former - just ask anybody who has ever voted for the losing side in an election where the winner did not win a majority of the popular vote, yet received over 50% of the seats.

2) Proportional representation leads to instability.

More of a sub-part to the previous argument, this is the argument which states that the minority parliaments generated would necessarily be unstable. Of course, this is not true. The experience of every industrial country which uses proportional representation - with the exception of Israel and Italy - is quite the opposite. Israel can be discounted out of hand, as their instability likely has more to do with the fact that the country is under constant threat of terrorism by Muslim fanatics. As for Italy, it is the exception which proves the rule, the one single example where political instability is par for the course - and yet they still seem to be governed well enough to be included in the G8.

Simply put, this so-called "instability" is a phantom; it does not exist in any substantive form. Opponents of PR will often attempt to obfuscate the facts by drawing attention away from the relevant reality and concentrating on side-issues. For example, they are quick to point out, in light of the recent political bantering in Germany, that PR led to months of uncertainty, and that - horror of horrors - the parties actually had to cooperate (something PR opponents dread). But the fact of the matter is, they did cooperate, they did form a stable government, and Angela Merkel has emerged as the Chancellor of a grand coalition, which will govern Germany from the stable centre. All is well. And if they want to concentrate on Germany, perhaps they can be reminded that Germany, prior to 2005, had not had a federal election since 2002. Hardly the "elections every few months" nightmare scenario PR opponents like to use to scare election-wary Canadians into opposing it. Then there's Japan; recently, the parliament there was dissolved. They had an election, and the governing Liberal Democrats were re-elected, unsurprisingly, under a PR system. The reason the government collapsed, of course, was not because of PR - it was because prime minister Junichiro Koizumi's own party turned on him and voted against his proposed privatization measure. Hardly a function of PR, that was a function of something which must seem foreign to Canadians - a break in party discipline.

Some point out that this recent minority government is an example of why they are bad, because of all the political intrigue, instability and deal-making (though as a political junkie I find such things hard to call "bad"). They seem to dismiss without even a second thought the minority governments of Bill Davis, David Peterson, Lester Pearson, Mackenzie King, Pierre Trudeau, John Hamm, and many others, once again preferring to concentrate on the exceptions, rather than the rule. The recently-deceased minority government was an extremely delicate situation, which is unlikely to happen again, even under PR. It was the result of series of flukes which would be extremely unlikely to repeat themselves, among them: the late Chuck Cadman's election as an Independent, the exceptional nature of a parliament literally divided 50/50 between Liberal/NDP and Conservative/BQ voting blocs after the election, the removal of Carolyn Parrish, the defection of David Kilgour, the one-issue bigotry of Pat O'Brien, cancer-stricken MPs, the Peter-Belinda-Paul triangle, the sponsorship scandal revelations, the Tories promising to support the budget but then withdrawing that support when their poll numbers shot up, Gurmant Grewal's tape-recorder, and virtually every ancient parliamentary procedure being dusted off and tried at least once in an effort to get the upper hand on political opponents, to be specific. The odds of those conditions occurring again are infinitesimal, and the odds of those conditions producing the exact same result for a second time are pretty much nil. Let's face it, the 38th parliament was extremely interesting, but it was hardly normal - even countries with PR don't experience such drama routinely. Which is exactly the point.

It is also important to keep in mind that, in our system, minority governments are the exception, not the rule. As such, politicians are acutely aware that a minority government is just one, long pre-writ campaign. It is an interlude between Round 1 and Round 2. As such, they are not expected to last very long, as the strategists are just biding their time until the perfect time to bring down the government and force an election campaign, during which one of the parties tries to win an artificial majority; and if they can't, we're back to the polls again soon enough. The period between 1957 and 1968 actually saw 6 federal elections. (57, 58, 62, 63, 65, 68). Only two of those (58 and 68) resulted in majority parliaments. Surely a system which allows for such instability is flawed, or so one would expect the PR opponents to say! The reason for the instability is not the minority parliaments. Minority governments are not, by their nature, unstable - our politicians choose to make them so, because they know that the ultimate goal is preparing for the next campaign, during which they hope to win a majority. If minority governments were the rule, not the exception - as is the case in countries with PR - then the politicians would be presented with a very different scenario: either cooperate and compromise, or face a never-ending string of election campaigns.

The thing about politicians is, they hate elections more than the people do, and it is not a difficult task to imagine why. Every time they go to the polls, their jobs are on the line - unless of course their name is Monte Solberg and they routinely win in Medicine Hat with 70% of the popular vote without even trying. Needless to say, if they were faced with the very real threat of losing their jobs, they would find a way to get along. Paul Martin would not be so uncompromising; Jack Layton would not be so demanding; Stephen Harper would not be so belligerent. Gilles Duceppe wants to break up the country - bad example. But under PR, the Bloc Quebecois' influence would be severely reduced, anyway. The point is, they would be forced to cooperate, whether they like it or not - which is what the voters put them there to do, not fling mud at each other and call each other names in the hope that next time around they can win an artificial majority. (Though I'm sure they could still find some way to accommodate monthly Brison-MacKay grudge matches.) Voters put them there to cooperate with each other. PR necessitates that; the present first-past-the-post (FPTP) system discourages it.

3) Proportional representation leads to the hijacking of parliament by small parties

Once again a continuation of the "minority parliaments are bad" way of thinking, this is quite a curious position coming from proponents of a system which, as a matter of routine, allows the hijacking of parliament by a single party which received support from fewer than 2 in 5 Canadians. It is also curious considering that it was our current system which produced a parliament in which Stephen Harper and Gilles Duceppe's parties - between them only receiving 42% of the popular vote, as compared to 53% for Paul Martin and Jack Layton's combined total - actually did hijack parliament by shutting it down for several days last spring. So the skew works both ways - the Tory/BQ bloc, representing 42% of Canadians, actually had the numbers to outvote the Liberal/NDP bloc, representing 53% of Canadians. Again, it is our current system which allowed for that ridiculousness, not proportional representation - proportional representation, in fact, would have prevented that from happening, and allowed the Liberals and NDP to work in a stable coalition, as opposed to wondering day in and day out whether or not today was the day the government would fall.

Now, the fear here is this: the Communist Party wins 0.8% of the vote. (I use the Communists because they seem to be the most common boogiemen for this scenario when it is presented by the PR opponents - and I thought we were over the red scare.) They thus win 0.8% of the seats - in our current parliament, that would mean 2 seats. This scenario also has the Liberals winning 154 seats - just two shy (after the speaker is elected) of a majority in the House. Enter the Commies, who pledge to prop up the Liberals, but only if the Libs bend to their every whim and do everything they say. In this scenario, the Liberals are apparently completely spineless, as they agree to these terms, and suddenly, Canada goes Red, the maple leaf is replaced by a sickle and hammer, and the capitalist dogs are sent to the gulags. This scenario also assumes that the Liberals (or whatever party is dealing with them) are not only quite evil, but also very, very stupid. It assumes that the Liberals would not, say, go to the Conservatives, or the NDP, or the Greens, or some other less Cold War-ish party for support; it assumes that the Conservatives, NDP, Greens, or the unnamed-but-definitely-not-crazy party would be completely unwilling to give this support, or that their demands would be more untenable to the Liberals than the demands of the Communist party. (The logic there actually contradicts itself - the assertion is that the commies are a fringe party and thus have crazy demands, yet it also asserts that the Conservatives would be even crazier than them.) And it assumes that the governing party - again the Liberals, just for example - would rather turn Canada into Soviet Canuckistan than simply dissolve parliament and face the people, letting them pass judgment on the Communist Party's attempt to take over Canada, and the NDP, Conservative, Green, and unnamed-but-not-crazy party's refusal to negotiate with the Liberals. To sum it up, the entire argument is based on all sorts of bizarre, ridiculous, and hyperbolic assumptions and contortions of reality.

One concern is legitimate - should a party with 0.8% of the vote be represented? I would personally say no - at least 1% support should be required, in my opinion - and most agree with me to some extent, and in fact, most countries impose some sort of threshold which parties must meet in order to qualify for representation in parliament, much like a certain threshold must be met in order to qualify for federal funding currently. This is generally the position supported by most advocates of PR; usually the cap is anywhere from 1-10%, though generally more like 2-5%. This would prevent parliament from facing the Weimar Germany scenario, where endless fringe parties kept getting elected. It is, admittedly, quite practical, and probably the system that would be used, should Canada opt for PR, and should that PR include the mixed-member proportional system (more on that later).

The hijacking of the House of Commons is something that occurs regularly under our own system. Allowing other voices to be heard instead of simply a ruling party which does not represent a majority of Canadians is not hijacking - it's democracy.

4) Proportional representation means parties do not govern for principle, but for politics

Perhaps I am a severely jaded individual, but it would seem to me that parties already govern for politics and not for principle, the reason being that they have to attract as big a tent as possible in order to reach the 40% threshold and win a fake majority, and then once in power, must keep those 40% happy by not doing anything drastic. Our current system reflects this - parties which flip-flop on every issue and govern by not doing anything meaningful are rewarded. Parties which do unpopular but necessary things are punished with crushing electoral defeats. In other words, parties which play politics win, and parties with principles lose.

It may be an idealistic notion, but parties should not have to fear being disproportionately decimated at the polls if they act on principle. Mulroney's GST was, in hindsight, necessary to balance the budget and pull the country's economy out of a tailspin - but the Tories were reduced to 2 seats for it. Now, if the party had only won 0.5% of the popular vote - which is the proportion of seats they won - this would seem justified. But the PC Party in 1993 actually won almost as many votes as the Reform Party, which won 52 seats. Our current system amplifies the severity of an electoral defeat to such an extreme that parties are terrified of alienating even a few of their voters, and thus are loathe to take principles positions. Politicians who actually do as they promise - like Mike Harris - are the exception, not the rule. Harris' Tories, although led by Ernie Eves, received about 35% of the vote to the McGuinty Liberals' 45%. A mere 5% swing from Liberal to Tory would have resulted in a near-tie, yet under our current system, this difference resulted in a huge electoral defeat for the Tories, with the Liberals taking 72 seats to the Tories' 24. The message was clear - if you do as you promise, govern from principle, and make unpopular but necessary decisions, you will be punished, and punished harshly. No wonder the Tories elected John Tory as their new leader, a good man to be sure, but one who has been slow to actually take stands on issues that differentiate himself from McGuinty. Principle over politics indeed!

Now, would this be fixed by proportional representation? Perhaps, or perhaps not. The point is, if the complaint is that PR could result in government by politics rather than by principle, then the same complaint must be turned on our current system, as well. One thing is for certain - under PR, the Tories would have been punished, but not reduced to a mere 24 seats, impotent to do anything meaningful for the following four years. They would only have been punished in proportion to how angry the voters were. Considering they garnered 35% of the vote - around what the governing federal Liberals seem to be polling lately - they were not nearly as hated as the system made it seem.

5) Proportional representation leads to backroom deals

It is true that proportional representation often leads to minority governments, which leads to deal-making. What is not elaborated upon is why this is a bad thing. In a democracy, one would expect that, if a majority does not support one particular way of doing things, then at least a compromise can be hammered out. The parties would get together, hammer out stands on policy issues which satisfy everyone involved, and then govern. Regardless of where these deals take place - in backrooms and broom closets or in front of TV cameras - they would then have to be presented to the legislature, in full view of the public, and debated, sent to committee, and put to a vote. It's not as if a parliament would be elected and then our leaders would make secret decisions that we would know nothing about. This is still a democracy, and the public would still have a right to know about the policy, and voice their opinions on it.

6) Proportional representation means that you can't hold anyone accountable

This is a concern. Basically, the thinking goes that, if two or three parties are working together in a coalition, then people do not know which party to punish for making unpopular decisions, and reward for making popular ones. There's just too much confusion.

The solution to this problem is two-fold. The first is to pay attention - for example, the Liberals and the NDP recently hammered out a budget deal. The budget, popular as it was, was claimed by both parties - in other words, Paul Martin and Jack Layton both wanted to take credit for it. PR opponents would argue that that's just too confusing. But let's be frank: is that really so confusing? Of course the Liberals are going to take credit for it, that is the nature of the beast - they want credit for the NDP's popular ideas. But anyone who is at least a casual observer could plainly see that it was the NDP, not the Liberals, who were responsible for that budget. That politicians will contort the truth to make themselves look better is a given no matter which political system is used. The key is in the voters actually paying enough attention to notice when they are being played for fools. And of course, it happens under majority governments as well - how many people are actually aware that the Liberals' debt reduction efforts, and Paul Martin's talk of the "democratic deficit" were both ideas the Liberals adopted from the Reform party?

The second part is to remember that, in cases where a coalition government presents a united front, and it thus does become difficult to distinguish who is responsible for what policy position, they are a united government. A coalition which presents a united front can be seen as one singular unit. Thus, if the Liberals, NDP and Greens formed a coalition government and did something unpopular that the voter wanted to punish them for, but that voter could not tell if it was the Liberals, NDP or Greens who were responsible for that position - and they are all blaming each other, not helping at all - then the solution is simple: vote them all out. In presenting a united front, a coalition government is making a choice; it is saying that it takes responsibility for its actions. In a sense, they are acting as a single party, and assigning blame and voting against them would be just as easy as going into a ballot box and checking the box marked "Conservative", or "Libertarian", or "Some Other Party".

This problem, in other words, does not need to be a problem, if voters pay attention, and if they know how to vote against someone they don't like.

That just about covers all of the major arguments against proportional representation. With that finished, it is time to examine the system itself.


Many of the arguments for proportional representation have been explicit throughout this essay. Outlining them at length again would be a waste of space. The merits of PR are just such that they speak for themselves, and should seem self-evident, and already covered in the logic argument. In summary they are:

1) Proportional representation promotes compromise between parties.
2) Proportional representation does not allow a single party representing less than 2 in 5 Canadians to govern as if it has support from 5 in 5 Canadians.
3) Proportional representation reduces voter apathy by making voters feel as if their vote actually counts towards something, particularly if they are voting for a party which is unpopular in their riding.
4) Proportional representation allows people to vote for whom they truly desire to win, as opposed to turning voting into an exercise of strategy, mathematics and predictions, in order to vote against what they oppose.
5) Proportional representation allows parties the freedom to take stands on issues without having to worry about suffering a disproportionately crushing electoral defeat.
6) Proportional representation discourages regionalism, which FPTP actively promotes, by encouraging parties to seek support by playing to a single region, which allows them to take all of that region's seats. PR, on the other hand, promotes nationalism, and allows representation from areas which would normally not have representation - for example, if PR were in place, the Tories would have some representation from Quebec, and the Liberals in Alberta and Saskatchewan (besides the mere two members they have now). It also puts separatism into perspective, reducing the power of the Bloc Quebecois and showing that Quebec is not as monolithically in support of separation as they appear to be.


The first past the post system has led to some incredibly unfair election results. Some key offenders include:

1) 1993


PartyLeader% of Votes% of Seats# of Seats
LiberalsJean Chretien41.2%60.0%177
BQLucien Bouchard13.5%18.3%54
ReformPreston Manning18.7%17.6%52
NDPAudrey McLaughlin6.9%3.1%9
PCKim Campbell16.0%0.7%2
+ 1 Independent

This election is stunning in its injustice, notable for wiping Canada's oldest political party off the map and setting it in a spiral towards its eventual death - even though it won millions of votes, and in terms of popular vote, placed third out of five.

The Liberals with barely over 40% of the vote took 60% of the seats and thus, 100% of the power. But that is the case with almost every election in our present system. What is truly exceptional are the four other parties.

The Bloc Quebecois was able to sweep the province of Quebec, winning 54 seats, enough to become the official opposition. They won 49% of the vote in the province of Quebec, yet took 72% of the seats, vastly over representing the separatists in relation to the federalists. Despite winning fewer votes than both the Reform Party and the PC Party, they still won as many seats as both of those parties combined. This is a clear-cut case of regionalism allowing a party to do better than a national party - namely, the PCs.

The Reform Party is essentially the same scenario, but in the west instead of Quebec. Actually coming second in the popular vote with 18.7%, they still won fewer seats than the Bloc Quebecois.

The NDP suffered their worst defeat ever - along with the Tories - in this election, taking a mere 6.9% of the vote. Their punishment was both a loss of seats and a loss of official party status, which requires at least 4% of the seats in parliament. They took almost 7% of the popular vote, but popular vote is not linked with seats, and they lost party status.

The Progressive Conservatives must have been shell-shocked on election night of 1993. I was too young to remember, but I imagine the feeling would have been similar for many Tories to the crash of '29. The worst electoral defeat in Canadian history reduced the governing Tories to a mere 2 seats in the House - that is, less than 1% of the seats. The PC defeat really tells the story of this election's unfairness. First of all, they were a national party, unlike the Reform and Bloc parties, receiving support from all across the country. This hurt them more than it helped them, however, as in our system, support must be concentrated in a few key areas, rewarding regionalism as a result. Second, despite winning over two million votes, compared to two and a half million for Reform and five and a half for the Liberals, they were utterly impotent when it came to representing those two million Canadians in parliament, having lost official party status, their leader, and their prestige. Third, despite actually winning more votes than the official opposition, they became the fifth party. Brian Mulroney said of the defeat, in derision of Kim Campbell and her campaign, "You don't come fifth." Actually, the Tories finished a very strong third; it is our electoral system which made them come fifth.

2) 1997


PartyLeader% of Votes% of SeatsSeats
LiberalsJean Chretien38.5%51.5%155
ReformPreston Manning19.4%19.9%60
BQGilles Duceppe10.7%14.6%44
NDPAlexa McDonough11.1%7.0%21
PCJean Charest18.8%6.6%20
+ 1 Independent

This particular election is notable for several reasons, not the least of which being the fact that, at 38.5% of the vote, the Liberals formed a majority government - that is, held absolute power - with the lowest proportion of the popular vote ever.

The Liberal Party, despite intense unpopularity, benefited from the FPTP system and was able to win many ridings where the vast majority of people opposed them. Their majority was narrow, but a narrow majority does not matter - with 155 of 301 seats, they could not be outvoted. So with support from a mere 38.5% of Canadians, they proceeded to govern as if they were given support from a majority of the people.

The Reform Party once again benefited from its western regionalism. Despite only marginally increasing its popular vote (up from 18.7% to 19.4), they won 8 more seats than they had before. Ironically, this sort of regionalism contributed to western alienation more than it helped it - which was Reform's raison d'etre. With the west voting as a bloc for a single regional party, alienation became even more pronounced. Of course, Reform would not have swept the west as completely as it did under a PR system, which would have given the Liberals, PCs and NDP a fair number of seats there, too, allowing the west to be a part of the consensus, as opposed to being on the outside looking in. On the other hand, through a fluke the Reform Party's seat total was actually within 0.5% of their popular vote total, which is incredible even in PR systems. However, a significant amount of that support came from Ontario voters, and Reform was not represented at all in Ontario, thanks once again to FPTP.

The Bloc Quebecois followed suit, and with a mere 38% of the popular vote in the province of Quebec, won 57% of the seats there, once again vastly over representing the separatists. The Bloc's national total was actually the lowest of the five parties, yet they came a very easy third.

The NDP actually benefited from a bit of regionalism on this outing, as well. With only 11.1% of the vote, they won a fair number of Maritime seats due to the favourite daughter status of leader Alexa McDonough. Their result of 21 seats was quite the recovery from 1993, and at 11% of the vote and 7% of the seats, were more or less fairly represented.

PC Party, despite experiencing a similar recovery, once again were the victims of an unfair voting system. Once again their attempt to have a national message was punished, once again their third-place status in the popular vote was rewarded with fifth-place status in the House, and their two and a half million voters were vastly underrepresented. Despite having increased their vote total to almost exactly that of what the Reform Party received in 1993 - which won Reform 52 seats - the Tories still only won 20 seats. And despite having a nearly-identical popular vote total with Reform in this election, the Tories received only one third the number of seats. Despite having strong support all across the country, they were once again relegated to fifth party status.

3) 1979


PartyLeader% of Votes% of SeatsSeats
PCJoe Clark35.9%48.2%136
LiberalsPierre Trudeau40.1%40.4%114
NDPEd Broadbent17.9%9.2%26
Social CreditFabien Roy4.6%2.1%6

Let's rewind for a moment, back to 1979. The election, as is typical, gave the winning party a much larger seat total than was representative of the votes it received, and gave the third and fourth parties extremely low seat totals compared to the number of votes they received. But what is interesting about this election is that the Liberals, despite losing the election, actually received about 5% more in the popular vote than the Tories did. The Tories won a very strong minority government, without actually winning the most votes. Canadians like to deride the Americans for allowing George Bush to become president despite losing the popular vote, but they are likely unaware that this can happen in our own system, as well. Other examples of this happening are 1957 (John Diefenbaker), and 1996 in BC (Glen Clark, who actually won a majority).


Now that FPTP has become thoroughly discredited, an alternative must be proposed. It is not a difficult task to see what is wrong with the current system, but what kind of system could we change it to? There are many options, and we must consider a number of criteria.

The first is that the system be roughly proportionate. Being a few percentage points off in one direction of another is forgivable, as no system is perfect, but the make-up of the legislature should more or less reflect the votes of Canadians.

The second is that the system not discriminate against minority voices - a key part of liberal democracy, after all, is allowing minorities to be heard and counted. So small parties should be given at least a fair chance. Therefore, it should not be too difficult for small parties with small but significant support bases to win seats, but neither should it be easy for radical fringe groups to win seats to attempt to hijack the political process. A balance must be struck.

The third is that the system eliminate strategic voting, that is, voting against a party rather than for one. No Canadian likes to vote strategically, and if possible, they should not have to.

The fourth and final criteria is that the system be practical and workable. It should not as a normal function produce legislatures which are unmanageable or unstable. This goes hand-in-hand with number two, above. This is extremely important, for if a system does not work, it is not worth implementing.


There are many types of proportional representation used by countries, many of them adding their own tweaks to them to suit their needs. This is a broad overview of some forms of PR. There are more than many people think. There are also many variations on the types listed.

1) Instant Runoff Voting (IRV)
Examples: Australian House of Representatives, some Canadian political party leadership votes

In Instant Runoff Voting, voters mark their ballots in order of preference. For example, there are seven candidates, a Liberal, Conservative, New Democrat, Green, Libertarian and Christian Heritage, and an Independent. Instead of simply marking an "X" beside one candidate, the voter ranks the candidates in order of preference, in this case, from 1 to 7. The ballot would look something like this:

4Alison Independent
2Bart Liberal
1George Libertarian
7Harriet Christian Heritage
5Jack New Democrat
6Mario Conservative
3Rob Green

On the first count, the first-choices are added up. If no single candidate wins over 50% of the vote, the candidate with the lowest support is eliminated, and their second choices are applied to other candidate. The process of elimination and redistribution continues until a candidate has over 50% of the vote. The way in which the votes are counted can differ from system to system, but the ballot always looks the same.

This can tend to produce some complicated counting sessions if the ballot is large. It remains difficult for small parties to actually win seats, and so is not very proportional, though more so than our current system. It also does little to address strategic voting, as the candidate with the lowest amount of support is the first eliminated, so someone might be tempted to vote for the last-place candidate in order to have their second choice votes counted before anyone else's. It would tend to produce quite stable legislatures.

A way to counter the strategic voting problem would be to have voters mark only two preferences; all but the two candidates in first and second place would be immediately eliminated, and the second choices of every eliminated candidate's voters would all be distributed at the same time. However, this could result in voters being discouraged from voting for smaller parties, instead voting for a larger yet still third-party candidate, hoping that they would win and go onto the final round. It would also tend to make it even more difficult for small parties to win.

The strategic voting problem is seen in Australia, where the parties actually produce advertisements telling their supporters how to rank the candidates, as the party strategists have worked out a strategic way to, they hope, elect the most candidates possible. Many Australian voters follow their chosen party's voting instructions exactly.

Given that this system does little to address strategic voting, if not making the problem worse, while not addressing proportionality problems and makes it difficult for small parties to win seats, I would have to advise against it.

2) Runoff Voting
Examples: Louisiana, France, some Canadian political party leadership votes

This is essentially the same principle as IRV, except the voter does not rank candidates on the ballot, but rather makes a single vote.

In the system used in France and Louisiana, if no candidate receives over 50% of the vote, the top two candidates go into a runoff election, held at a later date. Everyone then votes again, and whichever candidates wins the most votes wins the election. This is used in French presidential elections, and became famous a few years ago when it inadvertently produced a contest between the conservative Jacques Chirac, and the far-right nationalist Jean-Marie Le Pen. Le Pen lost in a landslide due to overwhelming opposition, as everyone except for Le Pen's own supporters voted for Chirac, despite Chirac being a political opponent, because of Le Pen's frightening social policy. (If that sounds familiar, Canadians, it's not; it's incomparable. It would be like making a choice between Stephen Harper and Ernst Zundel. Not a difficult choice.)

The reason for this anomaly was because the Socialist candidate, Lionel Jospin, came a very close third behind Le Pen, because the left's votes were so divided. Fear of this sort of situation happening would be reason enough for many people to vote strategically, so this does not really address the problem of strategic voting.

Alternatively, Canadian political parties often use this system, but it takes much longer. The first ballot is cast mostly to remove the fringe candidates - anyone under a certain level of support, usually 5% - is disqualified immediately. The second ballot is then cast, removing the candidate with the lowest number of votes. A third ballot follows, removing another candidate, and on until one candidate has over 50% of the vote. For example, at the PC convention in 2003, the first ballot was cast which would have removed fringe candidate Craig Chandler had he not already removed himself. The second ballot eliminated Scott Brison, who threw his support behind Jim Prentice, which was enough to move Mr. Prentice into second place. The third ballot eliminated David Orchard, in third place, who in turn gave his support to Peter MacKay, giving MacKay enough support to breach the 50% line. While this system works well for delegated conventions, it is difficult and cumbersome - though not impossible - to implement for the population at large.

3) Party Lists (Closed List or Open List)
Examples: Israel, The Netherlands, Finland

This system of PR works in a very simple way. Voters cast a vote for a party - not an individual candidate - and their vote is counted towards that party's total. The totals are then tallied and seats are distributed based on the percentage of the popular vote. So for example, in Israel the Knesset has 120 seats. A party winning 23% of the vote would be entitled to 23% of the seats, or roughly 27 or 28 seats (the actual number being determined mathematically in comparison to all the other results).

In this system, there is usually some sort of threshold which a party must meet in order to qualify for representation - such as a party must win at least 4% of the vote to qualify for seats. The actual occupants of those seats are determined by party lists, drawn up before the election. So each party fields a list of candidates - presumably with their leader at the top of that list - and those members take seats in that order. In the case of a party winning 28 out of 120 seats, the first 28 people on that list would be elected. This is known as closed list.

An alternative to that is open list. In an open list system, the party fields a list of candidates, but the voters, as opposed to the party, determine how they want the list to be structured, casting their vote for a party, but also voting for a member of the party list, to determine his or her placement on the list when all is said and done.

In terms of proportionality, this system provides a near-completely accurate portrayal of the actual voting preferences of the voters in the legislature. It is very precise. As such, it also reduces strategic voting, since people can vote for a party confident that their vote is not going to waste, unless that party's support is so low that it cannot meet the threshold. It is very easy for small parties to win seats, but impossible for fringe parties to do so - the definition of "small" and "fringe" being set by the threshold. It does tend to produce minority parliaments, but that does not necessarily mean that the system will be unstable, as once again, fringe parties would not be able to win seats. The system's stability would be entirely predicated upon the ability of the parties to work with one-another - so for example, a legislature made up of 1/3 Communists, 1/3 Anarchists, and 1/3 Fascists would not be able to work well together, but one made up of 1/3 Socialists, 1/3 Liberals and 1/3 Conservatives could likely find some way to compromise.

A downside to this system is that it makes Independent representation difficult if not downright impossible; since people vote for parties instead of candidates, Independents do not really exist. However, Independents are difficult to elect in our current system, anyway, so the loss would likely not be noticed. Also, it may not work for Canada - a country as large and diverse as Canada needs local representation. It is not a small, unitary state like Israel or Finland, but a vast federation with two official languages, ten provinces, three territories, and many unique regions. This system could likely not be employed in Canada, unless it was applied to only one house of parliament (the Commons or the Senate), while the other was used for local representation. Thus, some people propose keeping the House of Commons as it is, but electing 100 Senators using this system. This would actually solve problems of stability, as the Senate is not a confidence body, that is, the government is answerable only to the Lower House. The Senate, with diverse representation and views, would merely act as a check on the power of the Commons, which is routinely controlled by a single party, ensuring that unpopular legislation could not be passed, but not at the peril of the fall of the government.

The closed list system also opens up many debates about precisely who controls the lists - the party membership or the leadership? Are prime list spots earned, or political rewards? This can of course be easily solved by an open list system.

And argument against party lists is often put forward that it discriminates based on race, gender, sexual orientation, etc. The thinking goes that the party will take into account such factors when generating lists, and so employ a type of affirmative action with regards to who gets on the lists, ensuring that there are a suitable number of women and minorities. This is based on the false premise that the list system necessitates this - it does not. The parties would be left to their own discretion - much like they are now - as to who their candidates would be. Presently, for example, Canadian political parties would be completely free to pass party policy requiring that 50% of all nominated candidates be women. But just because they are able to do so does not mean that they would. In Germany, in fact, only the Green party employs such a system.

4) Mixed-Member Proportional (MMP)
Examples: New Zealand, Germany, Japan, Scotland

This seems to be the type of PR which most people advocate and many people assume this is the form that it would take if it existed in Canada, and hence, what many opponents and advocates base their arguments around. The system works by merging the system we have now of electing a member for each riding, with a party list system.

Japan uses this system; in the 2005 election, 300 members were elected to the lower house (the Diet) from single-member districts, and 180 members were elected using a party list system, as outlined above. The result was that the governing LDP took 219 of the 300 single-member districts, like our ridings (a situation not unfamiliar to Canadian voters), yet with only 38.2% of the popular vote, was only entitled to 77 of the 180 PR seats, for a total of 296 out of 480 seats, producing a majority government.

As can be seen from that example, this system can and does produce majority governments, and is far more likely to do so than a direct party list system, it does increase the likelihood of minority governments. Japan is perhaps a bad example - the LDP has been ruling that country for decades, almost uninterrupted, even more dominant than the Liberals in Canada.

In Germany, where this system is also used, the recent election saw the CDU/CSU of Angela Merkel win 150 of 299 seats in constituencies (ridings), with barely over 40% of the vote, so a bare majority without an actual majority of the vote. However, unlike in Canada, Merkel's power is not absolute, and her party is not free to do as it pleases. This is because there are an additional 315 seats distributed based on party lists, and the CDU/CSU won only 76 of those, for a total of 226 seats out of 614. (The party lists also took into account a system known as "overhang", where a party is actually entitled to fewer seats from the list seats as a result of doing so well in constituencies.) Also, the libertarian (though the term used in European politics is "liberal") FDP and the Green Party, along with the Left Party, while winning few if any seats in constituencies, still won 61, 50 and 51 PR seats, respectively, and the SPD won 222, which has forced the CDU/CSU and SPD to work together in what Germans call a "Grand Coalition" to lead a centrist government.

In Canada, this system would work in a very similar way. For example, Canada would have 300 ridings, about what it has now, which would elect 300 MPs. However, on top of that, more MPs would be elected from lists using PR, making the House more proportional. Exactly how proportional is determined almost entirely by the number of extra seats which would be added. An extra 100 seats would add a degree of proportionality; and extra 200 an even greater degree, an extra 300 an extreme degree, and over 300 would put greater emphasis on proportionality than on ridings.

The problem, unfortunately, is then created that there are some MPs which legitimately serve the people who elected them, and others who merely serve the party and do as their leader tells them. ("Party hacks.") This was a concern voiced by vocal PR opponent and Liberal MP Derek Lee, who said bluntly that he would not consider list MPs to be his equals, as he represents people directly, and they represent the party. To an extent I can understand why he would think that - many people would indeed be likely to view list MPs as somehow less legitimate.

But look at that from a detached point of view. First of all, Derek Lee himself - at the risk of employing ad hominem - is quite the Liberal partisan himself. Anyone who has ever seen him on a political panel knows full well that he will support his own party tooth and nail, and one gets the impression that he would - and does - vote against the will and the interests of his constituents if he is told to by his leader. Now, I cannot attack Lee too vociferously. For one, he is one of the few MPs in this country who consistently wins legitimate majorities in his riding - that is, over 50% of the vote. And for another, he is not alone. But he is useful for demonstrating a point. There are many politicians like him, and they are elected not under PR, but under our system. Under the auspices of representing their constituents, what they really do is warm seats in the House, cheer when their colleagues cheer, and even if they oppose a piece of legislation, will vote for it anyway because they are told to. This is ignoring the good work that MPs do in committee and in their ridings, of course, but when it comes right down to it, the most important thing an MPs has is his or her vote, and the vast majority of MPs, when it comes right down to it, vote the party line without dissent. This is, once again, under our current system, and not MMP PR.

To dissect that complaint even further, the claim that list MPs do not represent anyone is faulty - they represent the people who voted for them. There are entire swaths of the Canadian population who do not agree with the views of the Liberals or the Conservatives, and usually, they are quite underrepresented, if they are represented at all. The people who vote for these parties would most certainly feel that their views were being represented by the list MPs elected to parliament. It is just a different kind of representation.

And interesting way of dealing with this problem, which I am not aware has been tried anywhere to this date, is to allow the list seats - however many - to be made up of defeated candidates. Candidates who lose by the smallest margins are the first on the list, and it goes in descending order. This means that, just because a candidate loses their seat by 0.1%, they cannot go into parliament to represent the still-substantial number of people who voted for them.

This system eliminates the need for strategic voting, at least for the party lists; people could still vote strategically for constituencies. (Voters are usually allowed to cast two votes, one for the constituency, and one for a party.) It is either somewhat proportionate or extremely proportionate, depending on the ratio of constituency members to list members. It tends to produce minority governments, but usually strong ones, so it is unlikely to be unstable. Finally, it is fairly easy for smaller parties to be elected, but not fringe parties, with once again the difference between the two being set by a threshold for representation. This system also allows the election of political Independents at the constituency level.

5) Single Transferable Vote (STV)
Examples: Republic of Ireland, Australian Senate

This system is so arcane and complicated that I am loathe to explain it. The BC Citizens' Assembly on Electoral Reform produced a video which explains the system much better than I could. (URL: Essentially, the voter ranks their choices by preference - like in IRV - but instead of electing a single member, each electoral district or riding elects multiple MPs, usually anywhere from 3 to 8. This would of course result in either a vastly increased House of Commons, or much larger electoral districts, most likely both.

The math is more complex than most people are used to, but the long and short of it is that voter choice is maximized - the reason the BC Citizens' Assembly chose it as their preferred system. There is no need for strategic voting at all, and it tends to produce very proportionate legislatures, it is fairly easy for smaller parties to be elected, and it is also easier for Independents to win seats. Its main weakness is that the stability may be undermined due to the fact that politicians from the same party are often forced to compete with each other. In a riding that elects 5 MPs, for instance, a party would tend to field at least 5 candidates, hoping to win as many of the seats as possible. This means that there are five candidates from the same party, and they are all competing with each other.

This also creates a problem of voter confusion. It is difficult for many voters to keep track of their 5 or so candidates. 20, 30 or 40 candidates per riding would be too daunting a task for many voters. On the other hand, this system does maximize voter choice, allowing voters to cast their ballots for a party, a candidate, or both. They can rank their ballots so that they vote for a Liberal, a Conservative, two Greens, then another Liberal, followed by an Independent, then an NDPer, followed by another Conservative, etc. In general, the voters in countries with this system tend to love it, because it gives them the greatest choice. The politicians, of course, hate it because it forces parties to compete with each other, which as any party operative will tell you, can be very damaging, even more so than an election loss sometimes. Even Ed Broadbent, who has passionately pursued electoral reform, does not like this system for that very reason.

It could definitely be termed the best and the worst of PR - on the one hand, it has every good aspect of PR, including voter choice, proportionality, and a near-elimination of strategic voting. On the other hand, it can be unstable due to party in-fighting, is very difficult to understand for some people, and can lead to confusion.

6) Multi-Vote District (MVD)
Examples: None; This system remains theoretical; has been used in a video game

Also known as Shared Candidate Democracy, this system remains completely theoretical. To my knowledge, it has not been implemented in any notable way.

In this system, multiple candidates from each district are elected, and they each cast as many votes as they received in the election in the law-making body. There would usually be a limit to how many candidates can be elected, usually by a run-off, instant run-off, or threshold. Essentially, this means that each MP would be "worth" a certain amount, based on the number of people who voted for him or her.

The benefits of this system include its simplicity and lack of strategic voting (voters simply cast their ballot for their preferred candidate), its proportionality (it is the most directly proportionate of any as-yet proposed system), and representation of minority voices without the minority voices overruling the majority. Voters could feel less disenfranchised, as their representative would actually be casting their vote on behalf of them in the legislature, and the number of votes that legislator had to cast would be directly proportionate to the number they received in the election.

The drawbacks are, once again, mostly a matter of complexity. Computing vote results could be an issue, as could the variability in the number of seats in a legislative body. It could also add so much complexity to the legislature that it would be difficult to get things done. However, since most systems introduce a certain amount of complexity, and since this system is so proportionate, the complexity is worth it.

A variant on this allows for fraction voting - that is, giving each candidate a percent of your vote (A gets 23%, B gets 22%, C gets 45%, and D gets 10%).

Once again, this system has no known applications.

That is, this system has no known applications outside of the video game world. I personally have thought this system a good idea since I first played a strategy role-playing console game known as "Disgaea", which has a small degree of politics involved in it. The game includes a Senate which votes based on a similar principle, only instead of each Senator casting a number of votes equal to those who voted for them, they cast a number of votes equal to their character level, or in non-gaming terms, how strong they are. Obviously, this author does not advocate a system where parliamentarians are forced into mortal combat with each other in order to prove who is the strongest, as fun and endlessly amusing as that would be for observers. Still, it makes one think when a video game employs a more sophisticated voting system than many in the world.

7) Preferential Ballot
Examples: None of which the author is aware; is used for ranking sports players

This system is actually very simple. The voter marks either "approve" or "disapprove" beside each candidate. (Neither approval nor disapproval is expressed through simply not voting for or against that particular candidate.) Each approval is counted as a positive vote - the candidate's vote total goes up by one - and each disapproval is counted as a negative vote - the candidate's vote total goes down by one. when all is said and done, the candidate with the most votes wins.

As this system is completely theoretical, and I admittedly know very little about sports, I cannot upon it with regards to the criteria I have laid out. It is easy to imagine it creating a degree of proportionality - especially if used in conjunction with the MMP system. It would definitely help combat strategic voting, as a "Stop Whomever" vote could be cast by a simple "disapprove". It would likely produce fairly stable parliaments, and finally, it would allow for smaller parties and Independents to campaign and receive votes from people who are not concerned about "wasting" their vote, as they can approve, disapprove or remain neutral on each and every candidate.

I admittedly just dreamed up this system when I heard a similar system was used for choosing sports stars in some game or another - I remain, as ever, sports illiterate - and thought it would be interesting when applied to politics. I would fully endorse more study being done on how it would impact voting.


The merits of proportional representation truly do speak for themselves. Reducing voter apathy, having votes count, allowing disenfranchised people to have a voice, and parties having a voice proportionate to the amount of support they receive from the people seem like very logical and democratic concepts.

As has been pointed out, the opponents of PR have many arguments, but these arguments do not hold when subjected to scrutiny. When it comes right down to it, the only real argument for a continuation of our present system is a preference for forcing majority governments on populations that did not truly vote for them, for the sake of stability, or just for partisan reasons. In this case, the arguments for our present system are simply anti-democratic and utterly elitist; opponents of PR looking at proponents with disdain, seeing mere "losers" who are sore and want the system changed. It is about more than that. PR advocate are not sore losers - we are completely willing to accept a loss, if that loss is legitimate and democratic. We have nothing against majority governments - if a majority of the population actually votes for that government. There is nothing shameful in losing, but there is something shameful about losing because of a system stacked in favour of the winners, and the shame lies not with the losers themselves, but with the system itself, the system which calls itself democratic, fair and just, but is truly nothing more than a stick used to beat down dissent and silence opposition, opposition that a majority of people support.

There are many types of PR, and I am not in this particular article advocating any particular type. Canada is in the unique position of badly needing reform of some kind to both the lower house, and the upper (a whole other argument in itself). It would be very much possible to implement two completely different voting systems for both houses - such as is the case in Australia, with the House of Representatives using IRV, and the Senate using STV. Certainly some system could be worked out in which the best of all worlds was represented.

Our political systems were created over the past several hundred years. They have evolved constantly since then. Canada's parliament, in its current incarnation, has existed since 1867. Britain's has existed longer, and it has reformed itself - such as the elimination of rotten boroughs. There is no particular reason why our political institutions must remain static. Canada is far behind most of the industrialized world in terms of modernized voting systems.

Canada is a world leader in many aspects, and it is time we stepped up and reformed our democratic institutions, to make them more democratic, and show the world that Canada can also be a leader in democracy.

Copyright 2005, Ryan Ringer