Surveys have long shown that, when asked, many Canadians—a plurality in many provinces at various times—say they believe their provincial government should exercise more power (see Figure 1 and Table 1). This finding has two plausible interpretations. It could mean that Canadians really would prefer more power in the hands of their province. Or it could be a manifestation of provincial pride or attachment. Those with stronger provincial identities would be more likely to tell a pollster that they would like their provincial government to exercise more power, but in fact they may either support the status quo or have no real opinion about the actual distribution of powers.

To shed light on these issues we recently conducted two provincial public opinion surveys. We used more specific and nuanced questions than have been asked in the past in order to develop a better appreciation of Canadians’ views on federalism. Our data were collected immediately following the provincial election campaigns last February and March in Alberta and April and May in British Columbia. (We did not choose to begin in Alberta and BC, but were lucky enough that these were the first provinces to hold provincial elections following our securing of funding for the study. We hope to continue our study in other provinces during future provincial election campaigns.) We run our surveys to coincide with elections because it is during these periods when issues of federalism and accountability are likely to be at the forefront of people’s minds as they pay closer attention to politics and are forced to think about governmental responsibility in light of competing arguments.


Alberta and BC clearly are key players in Confederation. Both are large, both have traditions of some alienation from the center, and both have had prominent federal-provincial disputes. They have also traditionally been among the most devolutionist provinces. If we find a lack of real commitment to strong provincial governments in these provinces, this will be strong evidence for the idea that what previous polls have picked up is provincial pride rather than a genuine desire for devolution.

Funding for the study was provided by the Social Sciences and Humanities Research Council of Canada’s strategic initiative on Federalism and Federations. The survey was administered by the Institute for Social Research at York University. People in BC and Alberta who had responded to the Canadian Election Study (CES) of 2000 were re-contacted for this study. This means that we have a great deal of additional data on these respondents. We will make use of that in future analysis, but here we simply report the results from our own new questions, not those of the CES. Because these are respondents who had already agreed to answer the CES questions, they are likely to be more politicized than average Canadians. All frequencies are available through the website of the Canadian Public Opinion Archive at Queen’s University.

Our findings create a dilemma for policymakers and political actors in Canada: Canadians want both orders of government to be involved in important policy issues, yet they are very much opposed to the disputes that will inevitably occur when more actors are involved.

We begin the discussion of our results with questions about co-operation and intergovernmental disputes. We asked respondents: “Which government should be in charge of trying to fix problems with the health care system: the federal government, the provincial government, or should they have to work together on this?” Eighty-six per cent of respondents in BC and 78 per cent of respondents in Alberta chose the “work together” option. We also asked the same question in regards to the environment in BC, and 82 per cent favoured co-operation. In Alberta, where control over energy policy has been a long-time rallying cry for provincialist sentiment, we asked: “Which government should be in charge of trying to deal with the rising cost of energy: the federal government, the provincial government, or should they have to work together on this?” Amazingly, 78 per cent of Albertans wanted both governments to be involved (see Table 2).

But perhaps these questions are too easy. Why not have both governments involved if the question doesn’t highlight any costs that may be associated with this cooperation? We assume respondents may not immediately connect the obvious benefits of collaboration with potential costs. This became apparent in a wording experiment we conducted. We presented all respondents with two possible models of federalism. Half of our respondents were asked: “Thinking about how governments make decisions, which of the following do you think would be best for Canada: One, the federal government should have the final say on some things, the provincial governments on others, and they should both stay out of each other’s way; or two, both levels of government should work together and cooperate on most issues.” This treatment includes the word “cooperate” in the second option, and it is favoured by respondents by an overwhelming 78 per cent to 18 per cent.

However, half of our sample received a more neutrally worded second option: “… or two, both levels of government should work most things out together?” Two-thirds (67 per cent vs. 27 per cent) still opted for the collaborative model, which is strong evidence for the proposition that most Canadians oppose a classical model of federalism. However, the fact that fewer respondents choose the co-decisional model when the “work things out together” wording is used rather than the “cooperate” wording does suggest that potential costs, even if only vaguely hinted at, do cause some citizens to have some misgivings about joint policy-making if conflict becomes a major problem. At the moment, however, the link does not seem to be prominent in the consciousness of Canadians, so this and other questions in our survey indicate an overwhelming preference for collaborative, cooperative policy-making.

The flip side of this preference for cooperation is that most respondents do not like intergovernmental disputes. This became clear when we asked questions that tried to mimic some of the arguments for and against different models of federalism. We asked BC respondents two questions. “Now, thinking about the relationship between the federal and provincial governments, which of these two statements better reflects your own view: One, when governments fight with each other, we get better policies because of the discussions; or, two, when governments fight with each other, nothing ever gets done because neither one wants to take responsibility for things.” Three-quarters thought “nothing ever gets done,” with only one in five respondents seeing the benefits of debate. An explanation for this view is found in the answers to a subsequent forced-choice question: “One, when governments fight with each other, they usually have good reasons to do so and are just trying to do what is best for the people they represent; or two, when governments fight with each other, these disputes are usually just petty political bickering.” Only 27 per cent believe governments have good reasons for fighting with each other, while two-thirds conclude that intergovernmental conflict reflects cynical politicians’ attempts to manipulate the intergovernmental context to their advantage.

In Alberta we asked a series of agree/disagree questions rather than the forced-choice questions discussed above. These kinds of questions inevitably have a bias—known as “the acquiescence effect”—toward the “agree” response, and therefore the percentages should not be thought of as absolutes but are best interpreted relative to one another. Figure 2 reports the percentage of respondents who agreed with the various statements. What is striking, given our understanding of the strength of provincial identity in Alberta, is how few Albertans feel strongly that one government acting alone should be able to make a quick decision and how few see benefits to governments debating with one another. On the other hand, there is very strong agreement with the sentiment that governments should work together and compromise on most things, and a reasonably high level of fatigue with intergovernmental disputes. The challenge for policymakers—and what may cause them to despair at the unhelpfulness of public opinion—is that the two latter attitudes appear to contradict one another, a theme we will return to below.

The inner workings of government
Keep track of who’s doing what to get federal policy made. In The Functionary.
The Functionary
Our newsletter about the public service. Nominated for a Digital Publishing Award.

One of the driving concerns of our study is the relationship between federalism and accountability. It is possible that collaboration blurs the lines of public accountability because it allows governments to pass the buck when policies go sour, just as the present pattern of ad hoc intergovernmental cooperation and blame likewise blurs lines of accountability. We designed another wording experiment to find out whether citizens are sensitive to this possibility. While half of our respondents were asked the question discussed above about what happens when governments fight the other half were asked the same question but with more collaborative wording: “Which of these two statements better reflects your own view: one, when governments work together, we get better policies because of the discussions; or, two, when governments work together, nothing ever gets done because neither one wants to take responsibility for things.” Despite the fact that earlier questions revealed a huge majority preferring collaboration to unilateral action, even under this more positive wording 43 per cent of respondents believe that “nothing gets done” when governments work together. Admittedly, some of this must reflect general cynicism, but it does indicate that Canadians cannot be labelled naïve collaborative federalists: a substantial number appear to be sensitive to the difficulties of collaboration, such as the difficulty in getting things done.

There is other evidence that the public sees problems with accountability in a federal system. When respondents were asked whether they agreed or disagreed that “It is often difficult to figure out which level of government is responsible for what,” only 17 per cent disagreed. This is clear evidence that, for the vast majority of citizens, the current system involves some compromises in accountability. On the other hand, of the small number who said they had no trouble figuring federalism out, very few thought that “nothing ever gets done because neither one wants to take responsibility.” This contrasts with the large majority who do have some trouble figuring out where to give credit and lay blame: a majority of these people throw up their hands and say that “nothing gets done.” In short, those who can’t figure out the lines of accountability are more likely to think that nothing gets done. This suggests that a more collaborative model of federalism will appeal to most citizens only if it reduces, rather than exacerbates, the difficulty of determining which government (or both) is responsible for important policy outputs.

More specific questions on responsibility for health care give an indication about how well citizens can or cannot lay blame in a muddy intergovernmental context. We asked: “In the last few years, some people have come to think there are problems with the health care system in this province. Do you think these problems are mostly the fault of the provincial government, the federal government, both the federal and provincial governments, or are these problems not the fault of government at all?” We expected a greater willingness to blame the provincial government in BC than in Alberta, given the general unpopularity of the BC NDP and the fact that nurses were engaged in a job action during the campaign. However, in both provinces, less than one in four (22 per cent in BC, 24 per cent in Alberta) were willing to say that problems were “mostly” the province’s fault. Exactly half thought that both governments were to blame, and this proportion was the same even among voters who supported the incumbent governments. Most importantly, though, it appears as though Canadians are not dupes when it comes to intergovernmental rhetoric: despite far greater provincial blame-shifting to the federal government than vice versa on this issue, fewer than one in ten said the federal government was mostly at fault. The vast middle of the electorate sensibly blames both, or is so confused that it cannot tell where to fix blame.

On the surface, one might claim that Canadians’ views combine a naĂŻve intergovernmental pacifism (“please don’t fight!”) with a commitment to processes that are likely to encourage more conflict and cloud accountability (“everyone should be involved in everything”). However, it is possible to reconcile these two apparently contradictory tendencies through collaborative, co-decisional, integrative problem-solving processes, the kind one is more likely to associate with Germany than Canada. This model stands in contrast to the present one, where governments work together on an ad hoc, issue-by-issue basis only when it suits their interests, but in which a lack of an organized and permanent institutional home for decision-making means that governments are free to blame each other for policy failings whenever it suits their short-term interests.

Without necessarily realizing it, the Canadian public may in fact have a coherent and realistic vision of how federalism should work in Canada, one quite consistent with a more collaborative style of federalism. Or, at least, they may be receptive to moving in that direction. At present, we think that many Canadians are federalists only in the instrumental sense of supporting federalism as a check on the unrestrained power of either level of government. Yet even this minimalist view ought to point toward more advanced evolved processes than the division of powers constructed a century and half ago, albeit for logistical and religious reasons that made perfect sense at that time. The data suggest that Canadians, like policy experts, may recognize that the world has changed and that interdependence is inevitable. In Canada, however, there are no permanent processes to manage this interdependence.

We are not in a position to answer the more vexing question: can the processes of federalism in Canada be reshaped in a way that encourages governments at different levels to cooperate in policy development and service delivery? Can governments with different bases of support, different priorities, and sometimes deeply opposed philosophical commitments be provided with incentives to work together and then be credited or blamed according to their particular impact on the ultimate success or failure of a given policy?

And what about Quebec? Our surveys were conducted in Alberta and BC just after provincial election campaigns and we have not yet been able to replicate our questions in Quebec. On the other hand, Quebecers are among the most surveyed people in the world and may be the most surveyed in regards to their attitudes toward federalism. As Figure 1 showed, Quebecers are fairly evenly divided between the status quo and devolution on the question of the distribution of powers. However, when asked in 1997 which is more important, greater cooperation between the federal and Quebec governments or devolution, a very strong plurality opted for “more cooperation.” That Quebecers could so easily be pushed off their “devolutionist” position by a response choice as mushy as “more cooperation” says a great deal about how weak Canadians’ support is for the jurisdictional integrity and independence of their provincial government, even in Quebec. This in no way should be interpreted as suggesting that most Quebecers won’t instinctively side with the Quebec in a dispute with the federal government. But this “instinctive side-taking” is of the emotional and affective sort and does not directly imply a real desire to strengthen the powers of the provincial government. Quebecers, almost as much as their fellow citizens in the rest of Canada, care about results.

On many policy issues, the public either does not know what it wants or refuses to accept real-world constraints on its desires. Many commentators apparently believe that the arcane issues of federalism fall into this category of public policy issue about which polls have nothing useful to say to policy-makers. We disagree. We believe that the Canadians in our survey are actually giving realistic political direction about how governments should work. Canadians outside Quebec have little attachment to particular divisions of powers in the BNA Act and show little support for governments asserting their sovereignty in particular policy areas. They have little respect for the classical federal principle and little interest in attempting to implement “watertight jurisdictions.” If Canadians are federalists, they seem to be instrumental and protective federalists: they take for granted that there are two constitutional orders of government and want both to be involved in most policy areas in order to check one another. On the other hand, they also want to know who to hold responsible for policy. In short, they want a governmental system that provides them with effective policy and clear accountability. Political platforms built around massive devolution where the federal government is shut out of major policy areas are likely to have little public appeal, even in a province like Alberta.

Although Canadians do want both governments to be involved in most areas, and thus would not like to live in a unitary state, their preference is for a collaborative and cooperative intergovernmental model in which governments work together to come to agreement on most things. What they seem to want is a German-style, co-decisional model of federalism. That is a model of federalism envisioned by the spirit of the Social Union Framework Agreement, and it would be troubling to those who support more devolution, more autonomy, limits on the federal spending power, and greater respect for the 1867 division of powers. Canadians do worry about the disputes that could arise when both governments are involved in policy areas, and because they do we see more robust collaborative processes as a way for both governments to simultaneously take responsibility for decisions and avoid the most intense public disputes and blame-shifting, a game that few Canadians can track with accuracy or make sense of.

There are three possible futures for federalism in Canada. First, some have called for a return to a more classical model of watertight jurisdictions, where there is a closer connection between raising revenue and spending it on programs. Since responsibility would be easier for citizens to figure out, accountability would be strengthened. But Canadians do not appear to be anywhere near ready to abandon the advantage of having each government act as a watchdog on the other. Second, processes might be pushed in the opposite direction, toward a more collaborative model with formal intergovernmental institutions for decision-making. Governments at different levels would do much of their work together, which would likely require a significant change in political culture and the way our federal processes operate. Our evidence suggests that this model is strongly preferred by Canadians, though few Canadians likely realize that a loss of accountability is possible under such a model. If they did understand that, they might well reconsider the views they communicated in our surveys, which strongly supported this collaborative model. Even so, we think the results of our surveys show that Canadians prefer either of these two options to the third—the status quo—where conflict and uncertain jurisdictions blur the lines of accountability without fostering the more productive collaboration and joint accountability of the second option.

Fred Cutler and Matthew Mendelsohn’s detailed polling survey results are available at

Photo: Shutterstock

Matthew Mendelsohn
Matthew Mendelsohn is a professor and co-founder of First Policy Response at Ryerson University and a Senior Advisor to BCG’s Global Public Sector Practice. From 2016 to 2020, he served as Deputy Secretary to the Cabinet, leading the Prime Minister’s Results & Delivery Unit and the Government of Canada’s Impact & Innovation Unit. Prior to his cabinet role, Mendelsohn was the founding Director of the Mowat Centre, a public policy think tank at the School of Public Policy & Governance at the University of Toronto. Some of his former credentials include: Deputy Minister with the Ontario government, chief architect of the 2015 Liberal election platform, and member of Prime Minister Trudeau’s transition team. Mendelsohn received his B.A. from McGill University and Ph.D. from l’Université de Montréal and was a tenured faculty member in the Department of Political Studies at Queen’s University.
Fred Cutler is Assistant Professor in the Political Science Department at UBC.

You are welcome to republish this Policy Options article online or in print periodicals, under a Creative Commons/No Derivatives licence.

Creative Commons License

More like this