The normal timing of elections causes us to think of Canadian politics in terms of four- or five-year cycles. A party is elected to form a government and, four or five years later, it is either re-elected or defeated. But we have seen throughout this book that there are other and perhaps more meaningful ways to think about Canadian electoral politics. Canada’s most politically successful prime ministers — Macdonald, Laurier, King (followed by St. Laurent), Trudeau (following Pearson), and Chrétien (with Martin) — built electoral dynasties that endured for long periods of time. Mackenzie King served a total of 22 years as prime minister before handing power over to his successor, Louis St. Laurent, who served another nine years. Jean Chrétien served 11 years before handing over the job to Paul Martin who, for most of that time, had been his finance minister. As we have seen over the course of this book, there have been five party/leader combinations that would clearly qualify as “dynasties” over the long sweep of Canadian political history (see Figure 14.1).
These periods of electoral dominance, which in Canada are often identified with the fortunes of individual political leaders, are truly long ones by most comparative standards. In Britain — perhaps the most useful comparison because of institutional similarities — only a few leaders have approached these degrees of political longevity. Churchill, for example, served two terms in office spanning a total of nine years, but these were separated by one stunning electoral defeat (1945). Margaret Thatcher won three consecutive elections, serving a total of 11 years as prime minister. While Gladstone served longer (14 years), his tenure was spread across four interrupted terms in office. American presidents, of course, are now constitutionally limited to two four-year terms, but even before the enactment of the 22nd amendment, only Franklin D. Roosevelt served longer than eight years. The first question we pose therefore is: what explains these patterns of enduring political success enjoyed by Canadian political leaders, and by the parties that they led?
The interludes that we encounter periodically over the course of Canadian political history have been, with a few exceptions, short and sharp. Alexander Mackenzie and R.B. Bennett each served one full term as prime minister, but their administrations ended in electoral defeat, explained to some degree by the economic circumstances of their times. Arthur Meighen’s two brief turns as prime minister, first as Robert Borden’s successor for 18 months prior to the election of 1921, and later for three months in 1926, were brief interludes within an electoral cycle, since Meighen failed on both occasions to win the ensuing election. Joe Clark’s nine-month interlude began with his minority victory in the 1979 election and ended with his defeat in Parliament and in the 1980 election that followed. The stark contrast between these brief interludes and the successful political dynasties tells much of the story of Canadian federal politics as it has evolved over nearly a century and a half. A second question therefore is: can we explain how and why these brief interludes occurred?
Library and Archives Canada.
Mackenzie King at ceremony celebrating the diamond jubilee of Confederation, Ottawa, 1927.
Dynasties and Interludes in Canadian politics
Part of the answer to these complex questions can be found in the cases that do not fit neatly into our typology of “dynasties and interludes.” Lester Pearson, for example, served six years as prime minister, and his tenure in the office is highly regarded by historians today, in part because of the significant policy achievements of his administration.1 But Pearson endured one of the worst electoral defeats in Canadian history at the hands of John Diefenbaker in 1958, and he never succeeded in obtaining a majority of parliamentary seats in the two elections that he won (1963 and 1965). Yet, unlike Borden, he did manage to pass the reins of power along to a successor who did preside over one of the more enduring political dynasties of recent times. Do we think then of a “Pearson interlude” or a “Pearson/Trudeau dynasty,” as is suggested by the configuration employed in Figure 14.1 and as we have argued in Chapters 5 and 6 of this book? Neither term captures with total accuracy the high and low points, the successes and failures, of the Pearson and Trudeau eras of Canadian politics. But, with the benefit of historical hindsight, we can now see that the longevity of the Pearson/Trudeau dynasty rivals those of Macdonald and Laurier, even though neither Pearson nor Trudeau enjoyed unqualified political success over the course of their political careers.
The careers of two other political leaders who fell short of establishing dynasties also can help us to formulate part of the answer to the questions posed above. John Diefenbaker and Brian Mulroney led the Progressive Conservative Party to two of its greatest electoral victories, in 1958 and 1984 respectively, but neither was able to found a political dynasty based on this initial success. Diefenbaker’s six-year tenure as prime minister ended ignominiously with his defeat in the 1963 election and his ouster from the party leadership in 1967. Mulroney’s two terms culminated in the near destruction of his party in the disastrous 1993 election (see Chapter 11). Both leaders left a political legacy, however, that endured long after their time in office had ended. Diefenbaker bequeathed to his successors a solidly Conservative West and Mulroney a nationalist Quebec. In part, the huge and unwieldy electoral coalitions that these two leaders built explains both their initial success and their ultimate failure. In the longer term, holding these diverse coalitions together proved more difficult than constructing them in the first place.
One remarkable feature of Canadian electoral politics as seen from the perspective of Canadian history is the endurance of our main political parties. In contrast to the fundamental changes that periodically take place in the party systems of many other countries, elections in Canada continue to be fought largely between two parties — Liberal and Conservative — in name the same parties that contested elections in Canada in the nineteenth century. Of course, these parties have undergone many changes over their long history, and they have had to adapt to changing political circumstances on numerous occasions. But they have survived.
True, the Liberal Party of today is not the same as the “Clear Grits” of its formative years, or even the Liberal Party of Pierre Trudeau. And the Conservatives have undergone even more changes over their long history — from John A. Macdonald’s “Liberal-Conservatives” to Robert Borden’s “Unionists,” and John Bracken’s “Progressive Conservatives,” to mention only a few of the party’s many twists and turns. While it appeared after the 1993 calamity that the party might not survive at all, the new Conservative Party of Canada, with its minority victories in the 2006 and 2008 elections, appears to be putting at least some of the old pieces back together. Thus, today’s elections continue to be a contest between Liberals and Conservatives for power. Other parties such as the NDP, Bloc Québécois, and perhaps the Greens will continue to be a significant factor in elections, as “third” parties so often have been in the past. And future governments, like the present one, may well command only a minority of the seats in the House of Commons,2 but are all but certain to be led by either a Liberal or a Conservative prime minister.
One reason for the continued dominance of the two “old” parties can be found in the fact that Canadian elections are conducted using the single-member district, plurality wins, electoral system, which makes it extremely difficult for smaller parties, other than regional ones, to win a large enough number of seats to challenge either of the two main parties. As Alan Cairns argued in an oft-cited essay written in 1968, the structure of the Canadian electoral system has implications not only for the competitiveness of parties, but also for the nature of the parties themselves.3
A major change in the shape of Canada’s party system could still be precipitated by a change to a more proportional electoral formula — a proposed reform that continues to be actively debated.4 Had electoral reform come about during the period after the 1993 election, the new party system which that election produced might well have persisted in some form. But, as parties such as Reform, and its successor, the Canadian Alliance, learned over the course of the next two elections, their prospects of becoming truly competitive with the Liberals remained sharply limited under the existing electoral regime. With the 2003 merger of the Alliance with the remnants of the Progressive Conservatives, the party system began to revert, at first slowly, then more rapidly, toward its former shape. Of the new parties spawned by the collapse of the Mulroney coalition, only the Bloc remains.
The first-past-the-post electoral system has served both to sustain dynasties, and to help create interludes. Both the Trudeau and the Chrétien dynasties were supported by large numbers of Liberal members from Quebec, to a greater extent than the Liberal vote totals in that province might have warranted, had they been translated proportionally into seats. To an even greater extent, Chrétien and his government benefited from a virtual monopoly on parliamentary seats from Ontario, despite the fact that, in 2000 for example, the party won only a bare majority of the votes cast in the province. Conversely, the electoral system in 1984 magnified the scope of the Liberal defeat, and gave the Mulroney PCs so many seats (211) that a new Conservative dynasty appeared imminent. In 1979, the Clark Conservatives won the election despite having a lower share of the popular vote than the Liberals, a fact the party appeared to forget when they allowed themselves to be defeated in Parliament a short time later, thereby creating one of the shortest interludes in Canadian history.
However, the electoral system, important as it is in both shaping and constraining the fortunes of political parties, does not provide a complete explanation for the persistence of the Liberal and Conservative parties in the federal arena. Like their American counterparts, which have also endured for more than a century and a half, Canada’s two main political parties have learned to change and adapt in order to survive. Leadership conventions, policy conferences, and (when in power) Royal Commissions are among the institutional devices that the parties have utilized to restructure and reposition themselves in the political arena from time to time.
Both major parties have also proven adept at stealing ideas and policies from the minor parties and from each other. King’s Liberals took the steam out of the Progressive movement in the 1920s by adopting many of its policies. Today, elements of the populist rhetoric first associated with Social Credit and later with the Reform Party can be found in the policies embraced by the new Conservative Party. It was a Royal Commission appointed by the Liberal government of Pierre Trudeau (the Macdonald Commission) that proposed a free trade agreement with the United States. But it was the Conservative government of Brian Mulroney that negotiated the agreement and adopted the policy as its own.
Students of the Canadian political process observed long ago that Canadian political parties differ from those in other countries in significant ways. Unlike many European political parties, they are not strongly ideological. Although the terms “left” and “right” are widely used by political commentators and other elites in Canada, they do not resonate with much of the electorate. In the 2000 Canadian Election Study, which asked respondents to categorize their political views as “left” or “right,” by far the largest component of the electorate was found in the political centre (39 percent), and another large grouping of respondents (29 percent) rejected the concepts entirely or were unsure how to apply them to their own views (see Figure 14.2). Fewer than a third of the survey respondents (31 percent) chose to describe themselves in ideological terms. Of these, 18 percent placed themselves on the political right, and 13 percent on the left. This observation that Canadian voters are not particularly ideological in their orientation to politics is not new. Lambert et al. reported a similar finding from studies conducted in the 1980s.5
Restoring the competitiveness of the party system was always more likely to come about through the emergence of a party and leader that were capable of mounting an appeal to the political centre. This was essentially the key to the success of the Mulroney Conservatives in 1984 and 1988 and, to some extent, of the Diefenbaker Tories in the 1950s. Of course, that has also been the preferred strategy of the Liberals throughout much of their modern history. It was, after all, King whose formula for political success was to campaign from the left, but govern from the right. The 2006 election campaign saw the Conservatives attempt to move away from an emphasis on divisive ideological issues, and to pursue a more traditional centrist electoral strategy, an effort rewarded with a minority victory, thus encouraging the party to continue along this strategic path.
Much the same might be said of other “cleavage” variables based on the many different ethnic, linguistic, and social groups that comprise the Canadian population. While factors such as religion, social class, or ethnicity have been present from time to time in Canadian party politics, such forces have also long been noticeably weaker in Canada than in many other Western countries. In comparing Canada with Britain, the United States, and Australia in the 1960s, Robert Alford noted the comparative weakness of social variables in explaining Canadian voting patterns, and Richard Rose, in a study comparing Canada with a number of European countries, drew similar conclusions 10 years later.6
Rather than dividing the electorate along clear and relatively stable lines of social cleavage, Canadian parties often compete for much of the same policy space and many of the same voters. While regional and linguistic divisions are often highly relevant to politics in Canada, the key to national political success has always lain in bridging these differences rather than in exacerbating them. Parties based solely on linguistic or regional grievances can win votes from time to time, as the Progressives demonstrated in 1921 or the Bloc since 1993. But theirs is a strategy of registering political protest or regional identity, not of winning control of a national government.
Ideological Orientations in the Canadian Electorate
Canadian Election Study 2000, campaign wave. N=2691. The survey question asked: “Some people talk about polictical parties being on the left or the right . . . Would you personally say that you are on the left, on the right, in the centre, or are you not sure?”
Office-seeking parties tend to play down social and ideological cleavages, to form electoral alliances across social, regional, and linguistic divides, and to concentrate on framing their electoral appeals around general policy questions of national concern.7 Their electoral support tends to be based less on the long-term loyalties of particular social groups than on potentially unstable coalitions that can change over time. Such parties only rarely present voters with a clear choice between starkly different policy alternatives. Instead, they search constantly for electorally successful strategies, or repackage variations of established responses to ongoing problems. They organize around leaders rather than around political principles and ideologies, and they expect the leader to work out the various compromises required for the party to compete effectively in elections. Leaders who have been skillful at this style of politics have often enjoyed continued electoral success.
This style of party politics, as it has existed in Canada in various forms over many generations, has been repeatedly referred to by political scientists and other observers as the brokerage model. While the term has been employed by many scholars writing about Canadian political parties over the years, it was first stated explicitly by J.A. Corry in his 1946 textbook on Canadian politics.8 It is now, perhaps more than any other single theoretical understanding of the Canadian party system, the conventional view. The reasons for the persistence of brokerage politics are easy to understand when placed alongside the realities of the Canadian political environment. The risk of ethnic and linguistic conflict and the sensitivity of regional inequalities create a situation in which any national party would be destined to fail if it tried to mobilize the electorate around any single cleavage.
The political reality of such risks has been demonstrated on many occasions over the course of Canadian history and in many different ways — Diefenbaker’s weakness in Quebec, for example, or Trudeau’s alienation of the West. In more recent years, the failure of Reform, and its successor party the Canadian Alliance, to significantly expand its electoral base beyond the West illustrates the continued existence of these risks.
The brokerage model has many implications for the conduct of electoral politics in Canada, not all of which have been viewed favourably by voters or by scholars. Among the consequences of brokerage politics are the seeming lack of principled differences between the parties, the inconsistencies of many of their policies over relatively short periods of time, and the parochialism of their organizational structures, which frequently emphasize patronage and the settlement of competing claims.
Brokerage politics also places a great deal of emphasis on the role of the leader, whose job it is to act as the “chief broker” among the many different interests and social groups that make up the Canadian mosaic. Parties become identified, not with policies or programs, but with their leaders. A party that cannot win elections often blames the leader, and then transfers its hopes to a new leader who promises to lead it out of the political wilderness.9
This interpretation of Canadian party politics helps us to understand the main characteristics of “dynasties” and “interludes” as we have discussed them throughout this book. Leaders such as John A. Macdonald or Mackenzie King, who were particularly adept at the practice of brokerage politics, enjoyed not only electoral success but long tenure in the position of party leader. Others such as John Diefenbaker or Brian Mulroney, who enjoyed initial success in building broad electoral coalitions, failed to hold them together. Leaders such as Trudeau alternately succeeded and failed, but survived nevertheless. And certain other figures such as Arthur Meighen or Joe Clark, either for reasons of personality or circumstance, proved utterly unable to master the brokerage game. As the new Conservative Party under Stephen Harper continues its transition to the more traditional ground of brokerage politics, it remains to be seen whether the leader can continue to provide the glue to hold the party together and position it to build on its tenuous electoral successes of 2006 and 2008.
Conservative Party of Canada. Herman Cheung, photographer.
Stephen Harper speaking during the 2008 election campaign.
The brokerage model of party politics has implications for the ways in which issues are addressed by the parties, and the way in which they are presented and managed in election campaigns. The parties and the leaders who represent them are often wary about dealing with “big” issues in any other way than identifying them as general societal problems — unemployment, inflation, and crime, among others. When deciding which issues to emphasize in a campaign, and how to present them, the modern-day parties employ strategists such as pollsters, consultants, or advertising firms to gauge their electoral appeal. Earlier strategic techniques for vote-getting also allowed political parties to develop their electoral appeals, with political picnics, mainstreeting, or whistle-stopping providing the venues in which to test different campaign themes. Political parties frequently find it in their interest to change the issue focus from one election to the next, lest they become too closely identified with any one side of a particular policy controversy.
Any subject, large or small, discussed during an election campaign can be thought of as an “issue.” The ubiquitous use of the word issue disguises the fact that subjects of political discussion can take many different forms. Political issues fought over in elections can and sometimes do reflect the major questions of a particular place and time. In some of the earlier elections treated in this book, “the tariff” was a frequent subject of discussion (see Chapter 2). But, although it was often the subject of election oratory, elections in themselves did little to resolve the issue. The parties of the time recognized that farmers, manufacturers, and consumers had quite different interests in the matter, and they therefore tended to discuss the tariff issue in the most general terms. Political parties or candidates often find themselves in agreement in identifying the importance of a particular subject, but stop short of outlining a more detailed policy response. “Jobs, jobs, jobs” were, according to Brian Mulroney, the “three most important issues” in the 1984 election campaign. A Free Trade Agreement with the United States, which was never mentioned in that campaign, proved to be the path that he ultimately chose to work toward that goal after he had assumed office.
This book has examined the issues emphasized in election campaigns over the long sweep of Canadian history. In so doing, we have observed that there is considerable discontinuity in the framing of issues from one election to the next. Sometimes, this is because of the differing social, economic, or political problems which arose at a particular time. It was the high unemployment caused by the economic recession of the early 1980s that led Mulroney to place the main emphasis of his successful 1984 campaign on the creation of jobs. Similarly, the periods of high inflation in the 1970s injected that issue prominently into the 1974 campaign, leading the main contenders to agree on the definition of the problem, but to propose different solutions. We could cite Diefenbaker’s use of the pipeline issue in the 1957 campaign, the defence policy issues of 1963, or the “national unity” issue in the 1979 election, which was largely a response to the victory of the PQ in Quebec, as examples of the variability of issues over time.
Table 14.1 displays the “most important issue” cited by respondents to the Canadian Election Studies in four elections spanning four decades. The dominant concern with inflation found in the 1974 study was replaced by an emphasis on unemployment in the elections of the early 1980s and 90s, and that in turn gave way to health care as the most frequently cited issue in 2004.
The prominence of social issues in 2004 can be partly attributed to the general public satisfaction with the state of the economy at that time. Since the economic crisis of 2008, economic issues have returned to a more dominant position (see Chapter 13). The predominant impression here is the variation in specific issues within the continuity of the general issue types.
The choice of issue emphasis in a party’s campaign is also partly strategic. Parties choose issues on which they or their leaders sense that they have an electoral advantage, or on which they believe their opponents are vulnerable, and inject these into their campaigns. In the 2004 and 2006 elections, the Conservatives, sensing the vulnerability of the Liberals in the wake of the sponsorship scandal, emphasized “corruption” and “accountability” in their campaign rhetoric. The Liberals, in contrast, emphasized their commitment to social programs, particularly health care, implying that a Conservative government might place these programs at risk.
Most Important Issue in Four Elections, 1974–2004
Canadian Election Studies, 1974–2004. The question asked was: “What was the most important issue to you personally in this election?” First response only. Issues included were mentioned by at least 1 percent of respondents in one survey. Respondents mentioning other issues or citing no issues as important are included in percentages but not shown in table.
The 2008 election is a prime example of parties talking past one another in election campaigns. The Conservatives concentrated on portraying themselves as responsible economic managers, while the Liberals promoted the Green Shift program, designed primarily as a plan for environmental sustainability. As we have suggested in Chapter 13, it was partly the Conservatives’ ability to impose their definition of the issue agenda which allowed them to achieve a minority victory in 2008.
Canadian political parties are not completely free to pick and choose the issues that they place before the electorate. Quoting Siegfried at the beginning of this book, we noted his observation at the beginning of the twentieth century that electoral success in Canada all but requires a political party to address certain types of issues.10 While we would use slightly different language today to describe these, it is clear that leaders of brokerage parties have had to deal competently with all of these issue types in order to enjoy continued electoral success. The “prosperity of the country,” as Siegfried called it, questions of “national unity,” and the management of the public sector are recurring issue themes in many of the election campaigns that we have examined in this book. Siegfried also recognized the importance of leadership as an “issue,” a theme which likewise recurs in many modern-day election campaigns, just as it did in the earlier ones. Thus, while a comparison of issue themes across election campaigns shows us that there is considerable variability from one election to another, it also discloses the continuity of certain kinds of issues in national politics. “Inflation,” “unemployment,” and “budget deficits” appear as different issues in several campaigns. But they are all economic issues that address what Siegfried would have called “the prosperity of the country.”
Issue Priorities in Canadian Federal Elections, 1974–2008
Canadian Election Studies, 1974–2008.
See note on Table 14.1 for question format and wording.
Figure 14.3 shows the response to the open-ended question on “what was the most important issue to you personally in the [year of the] election?” in National Election Surveys since 1974. This shows that, until the late 1990s, economic issues dominated the public’s agenda. The issues identified as most important were not always the same specific economic problem or policy — they varied from the state of the economy generally, to inflation (in 1974), to unemployment (in 1979), to taxation (in 1980), to jobs (in 1984), to free trade (in 1988). Social welfare issues, primarily related to health care, have tended to predominate in some elections in the last decade, prominently in 2000, 2004, and 2006. And national unity, manifested in issues of bilingualism, federal–provincial relations, and Quebec, has been perennially important, at a lower but fairly constant level.
In elections over the course of Canadian history, parties and leaders that have held power for substantial periods of time were often seen by voters as the ones best able to deal with the economy and national unity, and, in the years since the Great Depression, social welfare issues.
While the specific campaign issues associated with each of these key areas may be short-term in nature, the areas themselves are of longer term importance. It has proven to be all but impossible to construct a foundation for continued political success in Canada without mastery of all three of these core issue areas. Allied with these in solidifying a continuing base of electoral support has been an image of competence in managing government, as well as an advantage in leadership defined more generally. In other words, a competent, trusted leader, who is able to provide voters with reassurance in each issue area, has always been a party’s best electoral asset. A combination of public preferences on these factors has allowed some Canadian political leaders to sustain themselves in power for long periods of time, providing the foundations for the five dynasties mentioned earlier.
Leaders who have failed to deal effectively with these three principal types of issues, whether because of their own inability to do so or because of the particular circumstances in which they found themselves, have at times won elections, but have not been able to establish “dynasties.” Put in today’s terms, a successful leader needs to be able to demonstrate to the electorate that he or she is best to manage the economy, hold the country together, and deal with welfare state issues such as health care, unemployment insurance, and pensions. A leader who can sustain such a favourable positioning has the potential to build a political dynasty as defined in this book. A prime minister who has the misfortune to preside over a period of economic recession, or who is forced to cut funding for essential public programs, will find electoral survival difficult. Likewise, a leader who is unable to appeal across the country’s regional and/or linguistic dividing lines is unlikely to be able to win a majority government.
Pierre Trudeau’s career as Liberal leader demonstrates many of these risks. At first a unifying figure, Trudeau’s Liberals won a solid majority in the 1968 election. But the accumulation of economic problems, and Trudeau’s handling of the FLQ crisis in 1970, nearly cost him re-election in 1972. Trudeau, like King after the 1925 election, eventually navigated through these difficulties and returned with a majority government in 1974. Once again, however, the combination of economic adversity and national unity hampered Trudeau’s chances of founding a new Liberal dynasty. His defeat in the 1979 election appeared at the time to herald the end of his political career, which would have, at that point, spanned 11 years in office. But Joe Clark’s mishandling of the minority government, together with his inability to make a breakthrough for his party in Quebec, combined to allow Trudeau and the Liberals to return to power in 1980. Although he succeeded in realizing his ambition to repatriate the Constitution and establish a charter of rights during that final term, Trudeau’s alienation of the West through the National Energy Program, along with the deep economic recession of the early 1980s, doomed the Liberals to defeat. His successor, John Turner, led the party to one of its worst electoral disasters in 1984.
Trudeau’s career illustrates many of the pitfalls of brokerage politics in Canada. During his early years in office, Trudeau did not play the brokerage game well, and he often publicly disdained many of its more obvious requisites. “Why should I sell your wheat?” he once told western farmers, before delivering a lecture on the difficulties of international commodity marketing. But Trudeau, who had come into politics from outside the Liberal Party, learned from these early mistakes, and gradually adapted to the political realities that confronted him. Had the economy been kinder, he might even have avoided outright defeat in 1979. Few western governments, including those in the United States, Britain, Germany, and France, were able to survive the economic shocks of the late 1970s. For most national leaders who were defeated at that time — Jimmy Carter, James Callaghan, Helmut Schmidt, Valéry Giscard d’Estaing — it meant the end of their political careers. But Trudeau returned after a short interlude in opposition, thanks to a bit of political good fortune and the mistakes made by Clark. We acknowledge that luck can play a role in politics. Had Mackenzie King won the election of 1930 instead of losing it, the Depression would almost certainly have ended his political career.
The Conservative government elected in 2006 and 2008 is not immune to these constraints, in spite of the apparent short-term weakness of their principal opponents. However, the question of “prosperity” may well be beyond its control. Economies have always responded to forces beyond the reach of governments. But in an age of increasing globalization, and with many of the key economic variables being influenced in greater measure by its NAFTA partners, any Canadian government is increasingly at risk of being unable to turn the economic issues of the day to its political advantage. Rightly or wrongly, governments of any political stripe are blamed for macroeconomic adversity. The recession of 2008, even though its causes lay predominantly outside Canada, brought with it higher unemployment, budget deficits, and continuing economic uncertainties. Few democratic governments survive these kinds of conditions. While there may no longer be references to “Bennett buggies” in everyday political conversation, the voters know who to blame. And opposition parties, whatever their particular ideological or policy orientation, take full advantage.
The other two issue areas that we have emphasized throughout this book likewise have elements that are difficult for governments to control or influence; however they may be more amenable to the intervention of political leaders than is the state of the Canadian economy. “National unity” has had multiple meanings over the course of Canadian history, but it has consistently been one of the core issue areas with which every prime minister must deal with some success in order to remain in office. Macdonald, Laurier, and King all demonstrated great sensitivity to this concern over the course of their long political careers. Trudeau and Chrétien, in spite of their attentiveness to federal–provincial matters, proved to be more polarizing figures in some parts of the country. Borden and Diefenbaker, for different reasons, failed to broker the key issues of national unity that arose during their times in office (see Chapters 2 and 5).
Since the 1970s, the rise of a more nationalist Quebec has presented a challenge (or an opportunity) to every federal prime minister who has occupied the office. The standoff between Trudeau’s and Lévesque’s competing visions of the country virtually redefined the politics of the late 1970s and early 1980s, and the aftershocks of Quebec’s first referendum in 1980 continued to reverberate throughout the next decade. The events surrounding the second referendum in 1995 constituted the greatest single crisis of Jean Chrétien’s stewardship. To some extent, Trudeau’s preoccupation with the issue of Quebec separatism, and with constitutional renewal in his later years, may also have led indirectly to his alienation of the West and his inability to manage economic issues effectively. The return of the Bourassa government in Quebec in 1985 opened the door for Brian Mulroney’s constitutional initiatives (see Chapters 10 and 11).
In all of these instances, the issue was thrust onto the national agenda by events having relatively little to do with federal politics. But the prime minister of the day was forced to deal with them regardless. In true brokerage fashion, the Harper government has given considerable attention to Quebec, having become convinced, as have many of its predecessors, that the key to electoral success lies in bridging Canada’s linguistic and regional divides rather than in exploiting them. Success on that issue front will largely determine whether Harper might be able to turn his current minority government into a lasting political dynasty.
The third critical issue area that we have emphasized throughout this book is social issues, by which we mean the management of the modern welfare state. This has also become a central concern of every prime minister regardless of party. Whether Canadians believed Brian Mulroney’s designation of Canada’s core social programs as a “sacred trust” is less important than the fact that he felt compelled to make such a declaration in the first place. It was not accidental that Stephen Harper in the 2006 election campaign identified a reduction of hospital waiting times as one of his five issue priorities.11
But, as an area of provincial jurisdiction, health care issues are not always responsive to federal policy initiatives, other than through the rather awkward mechanism of federal–provincial fiscal transfers. We might add to this observation the fact that some of the causes of recent pressures on the Canadian health care system derive not from a failure of policy but from more fundamental factors such as an aging population, technological advances in medicine, and rising health care costs. Governing parties are increasingly held accountable, not only for the performance of the economy, but also for their ability or inability to respond to the many pressures that threaten the large public sector programs on which Canadians have come to depend.
Other types of issues that may have little to do with public policy also sometimes intrude on elections. Following the period of minority government in 1972–74, Prime Minister Trudeau managed to persuade a significant number of voters that the “most important issue” in the 1974 election campaign was the restoration of a majority government. Given that the Liberals were the party most likely to be able to obtain such a majority, emphasis on this type of issue was in part strategic. But it also reflected to some degree the political uncertainty of the times. In 2008, the subtext of the Conservatives’ decision to call an early election was so that they could appeal to the electorate to give them the majority they had failed to achieve in 2006. This time, however, there was less attention paid to the issue, and some doubt among the electorate that the Conservatives should be trusted with a majority.
In the 2004 and 2006 federal elections, the Conservatives’ repeated emphasis on issues such as “corruption” and “accountability” served to deflect the issue agenda onto the question of integrity of those in government. A Liberal government, seriously wounded by scandal, was deemed to be vulnerable on these types of issues, and the opposition took full advantage of that vulnerability over the course of those two campaigns. This was certainly not the first time that scandal had overturned a Canadian government: the first such instance was in 1874 (see Chapter 2) when John A. Macdonald was relegated to the opposition benches. Scandals played an important part in the elections of 1925 and 1963, as well. When they arise, they tend to blunt the ability of a government seeking re-election to appeal on any of the standard issue dimensions, since they undermine the public’s trust that they will carry out their promises honestly.
A major consequence of brokerage politics is that no party is able to count on a large, stable base of supporters who will remain loyal to it through bad times as well as good. Uncertain about where the parties really stand on many fundamental issues, voters are often willing to desert one party or leader and transfer their support to another as circumstances change. Such flexible partisanship helps to account for much of the volatility that we see in federal elections in Canada, and provides fertile ground for the sudden rise of new parties and political actors. A brokerage party system spawns a highly volatile issue agenda and creates the conditions for potentially abrupt changes in party leader images. Since large segments of the electorate are sensitive to these short-term forces, the possibility of substantial swings in party support from one election to the next is always present. Yet despite this volatility, or perhaps because of their ability to cope with it, some leaders have managed to put together a string of victories in elections, and sometimes even to create favourable electoral ground for their successors.
Long-term feelings of partisanship on the part of the electorate have little to do with this process. Studies of Canadian voting behaviour have shown repeatedly that many Canadian voters do not have strong ties to political parties, and that there is a much greater element of volatility in Canadian federal politics than is disclosed by aggregate election results or by data on party preference at any given point in time. Following their solid election victory in 1993, the Liberals came perilously close to being reduced to a minority in the 1997 election (see Chapter 12).
Similarly, the 2000 victory was misread by many as clear evidence of continued Liberal dominance, when in fact the party remained highly vulnerable, as the events of 2004 would subsequently demonstrate (see Chapter 13). This pattern is not an artifact of a sudden patch of political volatility, or an observation that applies only to the recent period. Ever since 1965, when national survey data became available, we have seen an electorate in which there is considerable partisan instability even during periods when there has been little aggregate change in electoral outcomes.12 Herman Bakvis has similarly observed that the combination of weak party alignments and long periods of relative electoral stability represents a peculiarly Canadian “paradox.”13
Figure 14.4 shows the continued weakness of partisan ties, and the electoral unreliability even of supporters of the major parties. Even in 2008, perhaps their weakest point for many decades, almost a third of Canadian voters do identify in some way with the Liberal Party. However, only about five percent of all respondents considered themselves “very strong” Liberals. In fact, fewer than 15 percent of recent samples have considered themselves “very strong” party supporters, and at least another fifth of the electorate declare no partisan attachment at all. Added together, there are more weak partisans or non-partisans in the Canadian electorate than there are Liberals, and the percentage of Canadians who do not identify with any of the federal political parties has risen considerably in recent years.14 If we are searching for a solid base of partisan support within the Canadian electorate, it is increasingly difficult to find one. The Liberal Party is considerably weaker today in core support within the electorate than it was in the 1960s and 1970s, and cannot win an election through appeals to partisans alone.
Party Identification in the Canadian Electorate, 1974–2008
Support for the new Conservative Party in turn, while at near parity with the Liberals in 2008, is much lower than the Progressive Conservatives enjoyed during the 1980s under Brian Mulroney, and now rests at about the levels enjoyed by the PCs under Clark in the late 1970s.15 Core support for the new party is still, even after its two election victories, less than the sum of its former parts — Reform/Alliance and the Progressive Conservatives.
While the Liberals since 1993 seem to have settled at a lower level of support in the electorate than they previously enjoyed, the Conservatives also appear stalled at relatively low levels of core support. This, of course, did not prevent the Conservatives under Stephen Harper from winning the 2006 and 2008 elections, and it does not mean that they cannot win another under these same conditions. It means simply that they, like the Liberals, would have to do so, in typical brokerage fashion, by appealing to at least some voters who identify with other parties, or who claim to be independent of partisan feelings
Vote Switching in the Canadian Electorate
Based on the 1984–1988 Carleton Panel Study. N = 1045. Excludes non-voters in both elections and newly eligible voters not voting.
Evidence from national surveys since 1965 confirms that there is considerable volatility in the electorate from one election to the next. This is precisely what we would expect from voters with weak partisan attachments and political parties that typically frame their electoral appeals around their leaders, or around issues on which they believe they can achieve a short-term advantage.
In many of the chapters in this book dealing with particular elections, we have included a figure showing the turnover of the electorate from one election to another where such data were available based on surveys conducted at the time. Figure 14.5 presents a snapshot of the electorate based on the 1984–88 period. We chose this period to illustrate the pattern of electoral turnover because it is fairly typical of a time in which a governing party was re-elected and also because panel data covering both elections exist for this period. Other periods might be slightly more volatile — or slightly less — but the model of voting behaviour that they suggest does not differ greatly from one election to another.
Turnout in Federal Elections, 1958–2008
Elections: Canada (www.elections.ca)
Across the two elections shown in Figure 14.5, we find slightly more than a quarter of the electorate moving its support from one party to another.16 Only half stay put. In any election, much of the electorate is up for grabs for the campaigning parties. The total amount of movement in the electorate is also affected by the participation rates of newly eligible voters, and by the extent of non-voting, both of which will vary somewhat from one election to another and both of which contribute to the overall rate of volatility.
It is particularly important to include voting turnout in our calculations, as turnout has been going down in Canada for a considerable period of time. In fact, turnout in federal elections began a steep decline in 1993 and then moved lower in subsequent elections, reaching a new historic low of just under 59 percent in 2008 (see Figure 14.6). While a considerable part of this decline can be explained by patterns of generational change and other demographic factors, there is little doubt that the lack of competitiveness in the party system after 1993 has also contributed to the withdrawal of potential voters from the electoral process.17 Citizens with no partisan leanings or weak partisan attachments are more difficult to mobilize in elections. The strong regional patterns which have been evident in recent years may also contribute to the ongoing tendency of many Canadians to withdraw from electoral participation. Thus, people in many parts of the country could readily surmise that their vote would have little influence on the outcome of the election, either nationally or in their own local constituencies.18
Turnout rebounded somewhat in the 2006 election, in part because of the context of that election and the fact that it was more competitive, at least at the aggregate level.19 But despite the change of government in 2006, federal politics in Canada remained highly regionalized, and patterns of competition were weak in many constituencies. When the 2008 election was held, turnout declined once again, to the point where almost a million fewer Canadians voted than in the previous election. Given that one of the most important factors driving turnout down in recent years has been the reluctance of newly eligible young voters to enter the electorate, it is doubtful that turnout in federal elections will recover to pre-1993 levels.
However, variations in turnout contribute to electoral volatility, regardless of the direction. If voters withdraw from the electorate, they deprive parties of sources of voting support that may previously have been important to them. If some voters re-enter the electorate after one or more elections of non-participation, they introduce into the calculus a new short-term source of variation. Thus, a party such as the Conservatives could win an election by gaining the support of some voters who identify with other parties, by appealing to those who have not previously voted because of a perceived lack of choice, or by gaining advantage when previous supporters of other parties choose not to vote. In reality, every election involves some combination of these forces, as the patterns of switching displayed in Figure 14.5 clearly suggest. No matter which party is in power, electoral volatility remains a central fact of Canadian political life.
The image of Canadian politics portrayed throughout this book is that of a highly volatile or “dealigned” electorate to which parties and leaders must continually present new electoral appeals based on relatively short-term political calculations. As history shows us, that state of affairs can continue to sustain a government with relatively little apparent change if a leader proves adept at managing the many competing forces at work in the Canadian polity.
In considering the political dynasties that have been most successful over the course of Canadian history, it does not escape notice that four of the five (see Figure 14.1) have been Liberal dynasties. Following John A. Macdonald’s dominance of the Canadian political arena in the nineteenth century, no Conservative leader has succeeded in constructing an enduring period of political success, although several have won decisive electoral victories. The Liberal Party has been in power federally through much of Canada’s modern history, and has been extraordinarily successful at adapting to a variety of political and social changes that have occurred. The durability of the King/St. Laurent, Pearson/Trudeau, and Chrétien/Martin dynasties is often explained in terms of the leaders’ ability to read the political signals of their time and to adapt to new political circumstances as they arose. But these dynasties ended suddenly and unexpectedly in 1957, 1984, and 2006 respectively at the hands of the Conservatives. Does the current period resemble these previous situations, or might the political balance at last be shifting in the Conservatives’ direction?
One might argue that, despite the minority outcome of 2004 and their losses in 2006 and 2008, the present period still bears some resemblance to the interludes between other long periods of Liberal hegemony in Canadian federal politics. Even with the weight of the sponsorship scandal and an inept campaign, the Liberals nevertheless managed to win 103 seats and 30 percent of the vote in 2006, and they won seats in that election in every province except Alberta. Despite an unpopular leader, a disastrous issue choice, and another inept campaign, the Liberals managed to retain 77 seats in 2008. Surveys of the electorate still show that a substantial number of Canadians identify with the Liberal Party, in spite of these setbacks.
Analysts of the 1958 and 1984 landslide elections, or even of Joe Clark’s defeat of Trudeau in 1979, might easily have misread those events at the time as portending the demise of the Liberals. But in each instance they were back in power within a few years. In our attempt to track the course of changing political alignments, we sometimes ignore the ability of established political parties to adapt to new political circumstances. Parties that are primarily “power seeking,” as the Liberals have tended to be, are often particularly good at such adaptation.20 Those who have seen the Liberal Party come back time and again from electoral defeat throughout Canadian history would not be surprised to see yet another rebirth of “Canada’s natural governing party.”21
However, it is also possible that the present period is different than the past, and that the dominance of the Liberals over the period following Jean Chrétien’s 1993 election victory has been deceptive. At first, the post-1993 world seemed very different than the traditional brokerage model of party politics that Canadians had long become accustomed to, and that we have discussed throughout this book. After 1993, Canadian federal politics became segmented into parties and groups representing much narrower and more specific ideological, interest, or issue positions than had been the case in the past.22 Even the Liberals, with weaker representation from Quebec and the West, appeared increasingly to speak largely for the interests of Ontario or the major urban centres.
Now, with five further federal elections producing broadly similar electoral patterns, it becomes increasingly difficult to argue that 1993 was a temporary aberration or a transient phenomenon similar to those of 1962 or 1979. The Bloc Québécois has retained its appeal among Quebec voters, and the new Conservative Party of Canada has captured most of the electoral support won by two of its predecessors — Reform and the Canadian Alliance. The Liberals’ share of the vote has declined from a high of 41 percent in the 1993 and 2000 elections to lows of 30 percent in 2006 and 26 percent in 2008.
This newer and more highly regionalized alignment makes it increasingly difficult for any party to win a majority of seats across the country. Minority governments may well become the new norm, heralding more frequent elections and weaker, more unstable federal governments. Partisan attachment in such a system could harden, reinforcing regional alignments and increasing heretofore weak ideological differences between the parties. Such a situation could, in turn, cause portions of the eligible electorate to withdraw from politics altogether, in a vicious circle of turnout decline and political reaction.
Despite the surface plausibility of both of these interpretations, neither the “Liberal hegemony” hypothesis nor the “new party system” line of argument stands up well when examined in the light of the evidence presented throughout this book regarding the attitudes and attachments of Canadian voters and the ways in which successful Canadian political leaders have operated.
While nearly a third of Canadian voters still identify themselves as “Liberals,” such attachments are weak, and may well be growing weaker. The numbers of Canadians who do not identify with any of the federal political parties continues to increase. Given the right electoral circumstances, significant numbers of Canadian voters can still be moved by short-term appeals emanating from an election campaign or other sources. The volatility of public opinion polls over relatively short periods, such as an election campaign, further demonstrates this.
We have also seen that the ideological underpinnings of the Canadian party system are very weak (see Figure 14.2). This relative lack of ideological fervor does not appear to favour the prospects of those who believe that the future of the Canadian party system lies in presenting voters with more polarized choices. Neither a united right nor a more radical left would appear to be well placed to win the allegiance of a large cross-section of Canadian voters on any continuing basis. Stephen Harper fashioned his 2006 victory not by uniting the right but by deliberately positioning his new party closer to the centre of the ideological spectrum and appealing more directly to interests outside of his secure western base. It is therefore reasonable to expect more of the same in future elections.
However, in a dealigned electorate with weak party ties, the potential for sudden and unpredictable change is also high. In the past, leaders such as Diefenbaker, Trudeau, and Mulroney have burst onto the scene suddenly, and swept to landslide victories. The conditions under which this scenario might occur continue to be present in the Canadian electorate. A new party, political leader, or a compelling issue could easily ignite such a process of rapid change. It has happened before in the elections of 1911, 1930, 1958, 1984, and 1993.
Such a reversal of fortune can be precipitated by changing economic conditions, political miscalculations, or changes in the leadership of one or more parties. Given the nature of the Canadian electorate and the Canadian political environment as profiled throughout this book, there is no reason to believe that these traditional patterns will not occur again, perhaps in the not too distant future.
1. See Michael Bliss, Right Honourable Men (HarperCollins, 1995), Chapter 8.
2. Peter Russell, Two Cheers for Minority Government: The Evolution of Canadian Parliamentary Democracy (Toronto: Emond Montgomery, 2008).
3. Alan Cairns, “The Electoral System and the Party System in Canada,” Canadian Journal of Political Science 1 (1968), 55–80.
4. A large and active lobbying organization, Fair Vote Canada, continues to press for changes in the electoral system, as does an important women’s organization, Equal Voice. For some of the background of this issue, and a critical discussion of various reform proposals, see Henry Milner, ed., Making Every Vote Count (Toronto: Broadview Press, 1999). See also Voting Counts: Electoral Reform for Canada, the report of the Law Commission of Canada (Ottawa, 2004).
5. Ronald D. Lambert, James Curtis, Steven Brown, and Barry Kay, “In Search of Left/Right Beliefs in the Canadian Electorate,” Canadian Journal of Political Science 19 (1986), 542–63.
6. Robert Alford, Party and Society (Chicago: Rand McNally, 1963); Richard Rose, ed., Electoral Behavior: A Comparative Handbook (New York: Free Press, 1974).
7. See Kaare Strøm, “A Behavioral Theory of Competitive Political Parties,” American Journal of Political Science 34 (1990), 565–97.
8. J.A. Corry, Democratic Government and Politics (Toronto: University of Toronto Press, 1946). On the theory and practice of brokerage politics, see, among other sources, David Smith, “Party Government, Representation and National Integration in Canada,” in Peter Aucoin, ed., Party Government and Regional Representation in Canada, Vol. 36 of the Research Studies for The Royal Commission on the Economic Union and Development Prospects for Canada (Toronto: University of Toronto Press, 1985), and Janine Brodie and Jane Jenson, “Piercing the Smokescreen: Stability and Change in Brokerage Politics,” in A. Brian Tanguay and Alain-G. Gagnon, eds., Canadian Parties in Transition, second edition (Toronto: Nelson, 1996). See also Janine Brodie and Jane Jenson, Crisis, Challenge and Change: Party and Class in Canada Revisited (Ottawa: Carleton University Press, 1988).
9. The policy implications of brokerage politics are developed in some detail in Harold D. Clarke, Jane Jenson, Lawrence LeDuc, and Jon H Pammett, Absent Mandate: Canadian Electoral Politics in an Era of Restructuring (Toronto: Gage, 1996). See especially Chapters 1 and 2.
10. André Siegfried, The Race Question in Canada (London: Eveleigh Nash, 1907). As cited in Chapter 1.
11. The others were a 1 percent reduction in the GST, a $100/month child care tax credit, a federal accountability act, and an anti-crime package.
12. See Lawrence LeDuc, Harold D. Clarke, Jane Jenson, and Jon H. Pammett, “Partisan Instability in Canada: Evidence from a New Panel Study,” American Political Science Review 78 (1984), 470–84.
13. Herman Bakvis, “The Canadian Paradox: Party System Stability in the Face of a Weakly Aligned Electorate,” in Steven Wolinetz, ed., Parties and Party Systems in Liberal Democracies (London: Routledge, 1988).
14. In the 1974 Canadian National Election Study, for example, 12 percent of those sampled held no identification with any political party. See Harold D. Clarke, Jane Jenson, Lawrence LeDuc, and Jon H Pammett, Political Choice in Canada, abridged edition (Toronto: McGraw-Hill Ryerson, 1980), 111.
15. In 1979, for example, about 29 percent of the electorate identified their partisan allegiance as Progressive Conservative. In 1984, the comparable figure was about 40 percent. See Harold D. Clarke, Jane Jenson, Lawrence LeDuc, and Jon H. Pammett, Absent Mandate: Canadian Electoral Politics in an Era of Restructuring (Toronto: Gage, 1996), 51.
16. In 1993, for example, the turnover figure would have been considerably higher because of the collapse in support for the Conservatives and the shift of votes toward the new parties. In 1979, by contrast, only about 18 percent of the electorate changed their vote from that of 1974. For comparisons with other elections, see Harold D. Clarke, Jane Jenson, Lawrence LeDuc, and Jon H. Pammett, Absent Mandate: Interpreting Change in Canadian Elections (Toronto: Gage, 1991), 117–21.
17. André Blais et al., Anatomy of a Liberal Victory (Toronto: Broadview Press, 2002), 45–63. See also Jon H. Pammett and Lawrence LeDuc, Explaining the Turnout Decline in Canadian Federal Elections: A New Survey of Non-Voters (Ottawa: Elections Canada, 2002).
18. Jon H. Pammett and Lawrence LeDuc, “Behind the Turnout Decline,” in Jon H. Pammett and Christopher Dornan, eds., The Canadian General Election of 2004 (Toronto: Dundurn Press, 2004), Chapter 12.
19. On turnout in the 2006 election and its relationship to some of the longer term trends discussed here, see Lawrence LeDuc and Jon H. Pammett, “Voter Turnout in 2006: More Than Just the Weather,” in Jon H. Pammett and Christopher Dornan, eds., The Canadian General Election of 2006 (Toronto: Dundurn Press, 2006), Chapter 12.
20. Peter Mair, “Myths of Electoral Change and the Survival of Traditional Parties,” European Journal of Political Research 24 (1993), 121–33.
21. For a more detailed exposition of this argument, with references to many of the elections discussed in detail in this book, see Stephen Clarkson, The Big Red Machine: How the Liberal Party Dominates Canadian Politics (Vancouver: University of British Columbia Press, 2005). See also André Blais, “Accounting for the Electoral Success of the Liberal Party in Canada,” Canadian Journal of Political Science 38 (2005), 821–40.
22. This interpretation is explored in detail in R. Kenneth Carty, William Cross, and Lisa Young, Rebuilding Canadian Party Politics (Vancouver: University of British Columbia Press, 2000).