In his paper, Alain Noël, associate professor of political science at the Université de Montréal, argues that, contrary to the popular belief that the Social Union introduced a significant degree of collaboration into Canadian intergovernmental relations, the resulting arrangement actually reinforces the hierarchical relationship between governments. First, he points out that the final agreement signed in February 1999 was defined and promoted by Ottawa with little regard for the provinces’ common position agreed to in Saskatoon in 1998. Secondly, he argues that it institutionalises an intergovernmental regime in which the federal government decides on the policy direction and the provinces follow. Finally, he suggests that the Agreement is reflective of a system of governance that concentrates power at the top, seeks short-term solutions, and pays little attention to formal rules and due process.
Professor Noël also argues that, while Quebec’s refusal to sign the Agreement may have allowed the province some measure of asymmetry by allowing it to pursue its own course, it does not constitute asymmetrical federalism – as has been suggested – because it does not respect the federal principle. The brand of federalism exemplified by the Agreement clearly reveals that the ‘rules of the federal game’ can be changed without the consent of one of the Confederal partners – Quebec – and has reinforced the federal government’s use of its spending power in areas of provincial jurisdictions.
Even in terms of policy outcomes, Professor Noël is doubtful that the Social Union will lead to optimal social policy decisions. In his view, the structures and process created by the Agreement will have two important consequences on social policy development in Canada. The Social Union will encourage income and tax-based policies aimed at individuals at the expense of community and servicesbased alternatives. It also favours a “boutique programme” approach, as the federal government attempts to bypass the provinces, assist individuals directly, and get full credit for the intervention.
Our federation is evolving toward greater cooperation and consensus-building, while respecting the constitutional jurisdictions of each order of government, rather than toward extensive centralization in favour of the federal government or extensive decentralization in favour of the provincial governments.
Stéphane Dion, President of the Queen’s Privy Council for Canada and Minister of Intergovernmental Affairs, April 22, 19991
While sharing essentially the same concerns, the Government of Quebec does not intend to adhere to the federal/provincial/territorial approach to social policies. Furthermore, Quebec did not sign the Social Union Framework Agreement. Consequently, any reference to joint federal/provincial/territorial positions or provincial/territorial positions in this document do not include the Government of Quebec.
Preliminary footnote, in the Progress Report to Premiers of the Provincial/Territorial Council on Social Policy Renewal, August 19992
Canadian federalism is increasingly described as collaborative, to account for what is perceived as a remarkable expansion of non-hierarchical intergovernmental collaboration on a variety of issues ranging from internal trade to child benefits. Until recently, this new brand of federalism was more an outcome than an objective, more the unplanned and mixed result of various pragmatic arrangements than the product of a clear design for the federation. The February 4, 1999 Social Union Framework Agreement changed this situation fundamentally. For the first time, a broad agreement codified the new rules that would govern intergovernmental relations, in all areas of social policy but also, by extension, in a number of sectors not considered in the document.
The Quebec government was not part of this agreement. The new rules nevertheless apply to Quebec, and the collaborative process goes on as if, or almost as if, all agreed. Shared visions, agreements, agendas, objectives, consultations, outcome indicators, and progress reports regularly come out, usually with a footnote stating that the Government of Quebec shares “essentially the same concerns” but “does not intend to adhere to the federal/provincial/territorial approach” and is not included in the analysis or in the stated positions.
Most observers have portrayed this situation as less than ideal, but nevertheless satisfying in the circumstances. Some have even suggested that the outcome is helpful because it creates de facto asymmetry in Canadian federalism. I wish to argue, on the contrary, that collaborative federalism with a footnote is the worst possible outcome of a process that was not well engaged but did not require such a rapid and superficial conclusion. The Social Union Framework Agreement is detrimental because it is not a truly collaborative achievement, because it is not faithful to the federal principle and to Canadian tradition, and because it does not augur well for social policy in Canada.
The paper discusses these arguments in three parts. The first one argues that for collaboration to be non-hierarchical, it must produce cooperative solutions when there are significant differences in interests and perceptions. The social union negotiations constituted a good test in this respect, and their abrupt conclusion, defined and paid for by the federal government, fits oddly with the idea of non-hierarchical collaboration. It corresponds well, however, to an understanding of Canadian federalism as hegemonic cooperation, driven by the logic of what Donald Savoie has called court government. The second part addresses the question of asymmetry, to explain that the failure to include Quebec in the Social Union and the limited recognition given to the idea of shared sovereignty make the Framework Agreement somewhat asymmetric, but not very federal. Finally, the last part discusses the implications of the new institutional framework for social policy. Collaborative federalism with a footnote will make some policy choices easier than others, and it is likely to produce social policies defined more around persons and through income taxes than around places and through services. If there is such a thing as a Third Way, the Canadian version will bear the imprint of the federal Department of Finance.
Peut-être que j’hallucine
Peut-être que quelqu’un a mis dans mon gin
Deux trois kilos de narcotine
Si c’est pas le paradis
L’illusion est exquise
Dieu faites que jamais je ne dégrise
Le monde est rendu peace
Hey qu’on se le dise
Le monde est rendu peace
Marc Déry, Le monde est rendu peace, 1999
The idea of collaborative federalism was coined to characterize the evolution of intergovernmental relations in the second half of the 1990s. When it came to power in 1993, Jean Chrétien’s Liberal government turned the page on years of failed attempts at constitutional reform, to concentrate instead on jobs and “good government.” The close results of the Quebec 1995 referendum demonstrated the limits of this approach and convinced the federal government to seek “the renewal of the federation through non-constitutional means.”3 The approach was not entirely new, but it became more explicit and ambitious, and involved the different governments in a range of discussions that was unprecedented in scope and in depth. On a number of issues, going from fiscal and trade to labour market and social policy questions, positions were defined, reports were drafted, meetings were held, and compromises were reached or at least envisioned. Most of these compromises were unspectacular but meaningful responses to concrete problems, and they changed intergovernmental relations in an incremental fashion, even though not always in a clear, well-articulated direction.
An Agreement on Internal Trade was signed in 1994, another on a National Child Benefit System was reached in 1996 without Quebec, bilateral agreements on active labour market programs were obtained with most provinces (including Quebec), along with a host of arrangements or understandings regarding child well-being, health, benefits and services for persons with disabilities, other labour market matters, education, aboriginal affairs, the status of women, housing, and fiscal issues. Numerous intergovernmental meetings continue to take place, involving various levels of officials, and reaching well into specific and technical issues. For an external observer, it is becoming increasingly difficult to keep track of an evolution that is rapid, multi-faceted and fine-grained. Progress reports list a variety of minute achievements that do not add up easily into a coherent or meaningful whole. In many instances, the progress seems to lie more in the process than in tangible policy consequences.
The concept of collaborative federalism was introduced to make sense of this evolution. Harvey Lazar, who has produced the best analyses on this question, contrasts collaborative federalism with earlier forms of cooperative and executive federalism in Canada. Whereas the cooperative and executive federalism of the previous decades saw the different governments manage interdependence within a basically hierarchical framework, the collaborative federalism of the late 1990s would entail a joint management of interdependence embodying “a greater respect for the idea that the two orders of government should relate to one another on a non-hierarchical basis.”4 Lazar’s formulation – “a greater respect for the idea” – acknowledges that, in many ways, Canadian federalism remains hierarchical. The collaborative model is a project in the making, more advanced in some policy sectors than in others, and not fully accepted, even in Ottawa. Still, argues Lazar, overall the federal approach “is becoming more collaborative.”5
As with cooperative or executive federalism, the concept of collaborative federalism does not need to apply to all intergovernmental processes or arrangements. It is used to characterize a period, to describe a series of trends and evolutions that together set the tone for intergovernmental relations. In this respect, the federal approach is indeed “becoming more collaborative.” In its November 1999 report, the Auditor General of Canada documents a significant increase in various forms of collaborative arrangements initiated by the federal government. Some of these arrangements involve private partners and are not intergovernmental, but many concern some or all the provinces, often on important policy issues.6
But what does collaboration mean in this context? Is it equivalent to “cooperation between governments,” as Intergovernmental Affairs Minister Stéphane Dion suggests? Does it really deserve to be celebrated as “governance the Canadian way”?7 Does it provide a good and genuinely new avenue for non-constitutional reform?
At the outset, a note of caution is necessary regarding the looseness of the vocabulary on these questions. In current discussions, collaboration and cooperation are often used interchangeably. The notion of collaborative federalism has also been adopted in the recent past, to describe a different situation, namely the high-profile reconciliation and accommodation efforts of the Mulroney government in the middle of the 1980s.8 As for the cooperative federalism of the 1960s, it could probably be better described as a form of “federal unilateralism.”9 I will work, here, with Harvey Lazar’s definition, which associates collaborative federalism with “the idea that the two orders of government should relate to one another on a non-hierarchical basis.”
To collaborate and to cooperate mean roughly the same thing, namely to work together. If the work is done on a non-hierarchical basis, it implies that decisions are not determined through a chain of command. All partners may not have the same influence on the outcome, but agreements are produced through mutual adjustments and negotiations, rather than through hierarchy.
Collaboration, or cooperation, is always difficult. Even when there is a common interest, working together requires coordination and compromises. When the promotion of a common interest is automatically obtained and does not demand mutual adjustments, coordination or negotiations, there is no need to collaborate. “Harmony,” explains Robert Keohane, “is apolitical. No communication is necessary, and no influence need be exercised. Cooperation, by contrast, is highly political: somehow, patterns of behavior must be altered.”10 In other words, cooperation, or collaboration, presupposes conflict. It is the ability to overcome conflict, to work together despite differences in interests or in perceptions, that is the true test of a collaborative arrangement.
In this perspective, the achievements of collaborative federalism are decidedly mixed. Collaboration has emerged, but often in areas where little adjustments were necessary, where in fact collaboration did not mean much. With the more critical National Child Benefit and Social Union Framework Agreement, by contrast, collaboration was only obtained because the provinces yielded and because the absence of Quebec at the table was deemed unimportant.
In Saskatoon, in August 1998, the Quebec government had joined a modified inter-provincial consensus, to participate fully in the ongoing federal-provincial discussion on the social union. For the Quebec government, the move was significant. Until then, it had denounced the whole process as “another exercise in pan-Canadianism” that could only undermine Quebec’s traditional demands and legitimate the federal government’s social policy ambitions. The 1996 agreement on the National Child Benefit, which imposed on Quebec norms and mechanisms defined by the federal government and the provinces, led the Quebec government to define three conditions under which it could join the ongoing discussions on the social union. In Saskatoon, the Quebec government compromised on these three conditions: it left aside unsolved constitutional difficulties to join a bargaining process that did not make the opting out formula unconditional; it accepted much of the inter-provincial – and pan-Canadian – discourse on the social union; and it recognized implicitly a legitimate role for the federal government in social policy. In exchange, the other provinces agreed to include an opting-out formula in their common position, but only as a bargaining position.11
The importance of Quebec’s concessions has been underestimated by commentators, both in Quebec and in English-Canada. When the provinces turned around to accept, very rapidly, a framework that represented even less than their own, long-held, pre-Saskatoon position, most concluded that the Quebec government was responsible, because it never intended to reach an agreement. The fact that real concessions had already been made was not recognized. The fact, as well that the provinces did not even come close to their own demands did not seem important.
From the point of view of the federal government, the outcome made sense. The Social Union Framework Agreement that was adopted was basically the one federal civil servants had written, borrowing the inter-provincial vocabulary but conceding little that had not already been offered. The lack of support in Quebec, even from the federalist official opposition, which did not approve the Agreement, apparently did not matter.
For the provinces, the result seemed more problematic, since the Framework Agreement did not retain much from years of efforts to circumscribe the use of the federal spending power.12 In part, the outcome can be explained by the initial position of the provinces, which did not include an opting-out formula. Contrary to Quebec, many provinces wanted less to develop their own social programs than to be heard in the planning of pan-Canadian policies.13 Many also distrusted their ally, the Quebec government, more than their counterpart, the federal government. These divergences do not explain, however, the distance between the pre-Saskatoon inter-provincial position and the final agreement. A popular, and quite plausible, interpretation is that the provinces accepted the agreement in exchange for enhanced health financing and for a new equalization formula, long demanded by the largest and wealthiest provinces. There is no doubt that financial considerations played a role. Still, given the public pressure for better health funding and the difficulty, for the federal government, of supporting health services without going through the provinces, improved financing was likely in any case. A new equalization formula was also to be expected, if not right away, at least in the years to come. The different factors, however, added up, as they did many times before.14 Governments from the small provinces wanted a strong central government; all except Quebec accepted, indeed demanded, a pan-Canadian vision; Quebec was not trusted as an ally; and provincial politicians knew citizens favored Ottawa on these questions.
Whatever the case, it can still be concluded that there was a collaborative outcome in the case of the Social Union Framework Agreement since the Framework was negotiated and agreed upon by the different parties, except Quebec. Consider, however, how closely the process fits with Keohane’s understanding of hegemonic cooperation, “which relies on a dominant power making rules and providing incentives for others to conform with those rules.”15 Hegemonic cooperation, explains Keohane, “is not a contradiction in terms.” It simply integrates the fact that cooperation emerges out of conflicts, and not necessarily of conflicts among equals.16 The notion fits oddly with the idea of non-hierarchical collaboration, but it corresponds fairly well to the course of events in Canada. A negotiation among equals would center on the credibility of mutual commitments, as each partner limits its own autonomy to reduce uncertainty about the actions of others.17 Hegemonic cooperation, on the contrary, is likely to leave the less powerful parties uncertain about a dominant power that basically controls the rules of the game. Because it has failed to enunciate a clear overarching view or a well-defined set of principles, notes Harvey Lazar, “the federal government’s behaviour is difficult to predict, making it an uncertain and at times unreliable partner for the provinces.”18 This uncertainty, which in recent months has been manifest in various ill-fated improvisations, can hardly be seen as an indicator of non-hierarchical collaboration.
There is another way, however, in which collaborative federalism can be understood as non-hierarchical. In the language of the new theories of public management, the formal hierarchy of the Weberian bureaucracy should be, as much as possible, replaced by the non-hierarchical logic of the market. The conventional bureaucratic chain of command guarantees democratic accountability, neutrality and continuity, but it also creates rigidities and perverse incentives. The new public management proposes to focus on outcomes rather than on process, and to let autonomous agents provide public services.19 This is precisely the logic of collaborative federalism. The various collaborative arrangements proposed by the federal governments are non-hierarchical in the sense that they replace the traditional logic of public administration by that of alternative delivery systems. “Under these arrangements,” explains the 1999 report of the Auditor General, “the federal government involves external partners in the planning, design and achievement of federal objectives, replacing delivery by federal employees, contractors or agents. These partners are not accountable to ministers or Parliament.” They can be other governments, but also private firms or voluntary organizations.20 In this perspective, the absence of hierarchy only refers to the absence of a conventional chain of command. It does not mean all partners are equal or have an equal say. The opposite may, in fact, be true.
Indeed, the new public management clearly distinguishes the role of the central state from that of its agents, be they sub-unit governments or private actors. The central state, goes the standard slogan, should be “steering not rowing.” It should set the objectives and define the desired outcomes and let others do the implementation. Service providers will have some autonomy, but they will be held accountable with regular evaluations from above, based on pre-established performance indicators.21 Transposed to the Canadian federal context, this logic is odd but revealing. It associates democratically accountable provincial governments to private or voluntary sector service providers that are in a principal-agent relationship with the federal government. Like other agents, provinces row for the federal government, which steers and sees that they row in the right direction, at the appropriate rhythm and with sufficient energy.
To a large extent, the new public management remains an ideology, a rhetorical point of view more than an achievement. The implementation of the proposed reforms is difficult and fraught with contradictions and, at least in Canada, it is not very advanced.22 In intergovernmental relations, in particular, the federal government has not abandoned completely the idea of rowing as well as steering, and the provinces remain more than ordinary partners. Still, just as it does away with the conventional bureaucratic emphasis on legal constraints and formal procedures, the adoption of a new public management rhetoric tends to dismiss the logic of federalism. The division of powers matters less than concrete results, and policies can best be judged by their outcomes. This is why the sharing of information and the development of performance indicators becomes so important. The logic is akin to that of subsidiarity, a concept that was embraced for a while, but later dropped, by the federal government.23
The a-federal character of the new public management explains why observers diverge in their evaluation of the centralizing or decentralizing impact of collaborative federalism. Many see the overall pattern as obviously decentralizing, either toward the market or toward the provinces.24 Others present a more prudent evaluation, arguing various outcomes remain possible.25 Some, particularly in Quebec, interpret the Social Union Framework Agreement and other Canada-wide initiatives as indicative of “an increased preponderance of the federal government in social policy.”26
There is more to these different evaluations than the various standpoints of the observers. These divergences have to do, as well, with the peculiar impact of the new public management on the distribution of power between different orders of government. Paul Hoggett, who studied the introduction of public administration reforms in the United Kingdom, notes that the new models of operational decentralization were promoted by the same Conservative politicians that had “created one of the most centralized forms of government Britain had experienced this century.”27 Operational decentralization, Hoggett argues, “served to reinforce centralization processes.” Implementation has been devolved but “control over policy and the allocation of resources had become increasingly concentrated.” “Performance-based funding,” in particular, has proven to be a powerful tool of central control. According to Hoggett, the new, presumably decentralizing, organizational approaches were “harnessed to a political project which was designed to destroy virtually all alternative power bases within society which might challenge Conservative hegemony.”28 In countries like Germany and Denmark, where such alternative power bases were better established – whether in the public service or in federal, corporatist, or local arrangements – new public management proposals simply failed.29 In other words, one needs a strong centralizing government to establish new forms of non-hierarchical collaboration inspired by the new public management, and if these reforms succeed, central control tends to be enhanced.30 The success of decentralization, understood in this way, bodes well for centralization, understood in a more conventional fashion.
At this point, we could accept as a relevant description of current trends the idea that Canadian federalism is becoming more collaborative, insofar as more collaboration is indeed taking place. This collaboration, however, is hardly non-hierarchical. For one thing, when important interests are at stake, the pattern is more akin to hegemonic cooperation than to a negotiation among equals seeking to reduce uncertainty. In addition, the logic of the new public management that informs collaborative federalism clearly places the federal government at the top, in control of a vessel where provinces, like other agents, are expected to row in the same direction. It takes a strong central government to impose this “non-hierarchical” logic to the various social and political actors.
But if collaborative federalism cannot be understood as non-hierarchical, what is it exactly? A good way to make sense of this new form of intergovernmental relations is to understand it not as a specific form of relationships among governments (disentangled or not, hierarchical or not), but rather as a more or less institutionalized governance structure, largely defined by what J. Stefan Dupré called “intragovernmental considerations.” Cabinets, argued Dupré, are the “key engine(s) of the state,” and they “can operate in vastly different ways.”31 From the 1920s to the 1960s, the “departmentalized cabinet” left much autonomy to ministers and officials responsible for a given portfolio and it facilitated intergovernmental cooperation along functional lines. This was the era of cooperative federalism. In the mid-1960s, the rise of the “institutionalized cabinet” and of central agencies encouraged the formulation of more encompassing priorities and objectives and made intergovernmental relations more visible and competitive. First Ministers’ Conferences became the main instrument and the constitution the main issue.32 Executive federalism, as it would be called, prevailed definitively by 1980, when Prime Minister Pierre Elliott Trudeau declared the end of cooperative federalism and announced more competitive relationships. Informed by an overarching vision of the federation, Trudeau “tested the limits of federal power in field after field” and effectively transformed the constitution and the country.33 Later attempts at reconciliation, undertaken by the Conservatives in the 1980s and early 1990s, failed to reform a legacy that took hold solidly in Canada outside Quebec.
Meanwhile, the structure of power continued to evolve within the federal government (and probably in the provinces as well). The “institutionalized” Cabinet, argues Donald Savoie, “has joined Parliament as an institution being bypassed… in the late 1990s, effective power rests with the prime minister and a small group of carefully selected courtiers.”34 Decisions, large and small, are taken at the top, often without involving Cabinet or the departments concerned. Chrétien’s Verdun speech on national unity at the end of the 1995 referendum campaign, cuts in unemployment insurance, the Millenium Scholarship Fund, the Canada Foundation for Innovation, and a host of other measures significant for intergovernmental relations emanated from the center, without going through the Cabinet process.35 Even the introduction in 1995 of the Canada Health and Social Transfer, which involved $ 25 billion in transfers to the provinces and gave rise to the social union process, was decided without involving Cabinet.36
For intergovernmental relations, this concentration of power has at least three consequences. First, the whole process becomes controlled at the top, because line ministers are not trusted to manage intergovernmental, or other, issues.37 In some cases, this centralization has the advantage of accelerating or simplifying decision-making. Often, however, it leads the government to ignore important regional and functional considerations and to lose sight of issues that are not at the top of the political agenda. For instance, the millenium scholarship fund, the offer to support professional hockey teams, and the clarity bill contradicted other governmental objectives or were advanced despite significant opposition in Cabinet.
Second, this concentration of power makes intergovernmental relations more haphazard and unpredictable. In the absence of a clear overall vision or strategy, the central government tends to react to circumstances, events or provincial initiatives, forcing the provinces to scramble with the results. The centre, argues Savoie, governs by “bolts of electricity” and “micromanages” the files selected as important by the prime minister.38 As a result, the provinces have come to see Ottawa as a “most unreliable partner.”39 This impression certainly was not dispelled when Jean Chrétien quipped, in the days preceding the social union agreement, that “sometimes on Monday I feel like giving the provinces more money, and then on Tuesday not.”40 Few statements better capture the implications of court government for intergovernmental relations.
Third, the concentration of power around the prime minister creates a situation where those in charge tend to have “little patience for due process.”41 Unable to develop support for constitutional change in the provinces, Ottawa moved alone, with little success, to recognize Quebec as distinct and to introduce regional constitutional vetoes.42 Unable to compromise on the social union, the federal government pushed forward a Framework Agreement that will apply to Quebec even though it is not a signatory. Even the process that was created on February 4, 1999 seems disposable. Less than two weeks after the agreement, the federal budget introduced major changes in the Canada Health and Social Transfer that had not been preceded by the formal consultations specified in the Framework Agreement. The fact that the issue had been publicly debated, in a general way, was deemed sufficient to meet the requirements of the Agreement.43 The $ 1 billion management failure at Human Resources Development that was unveiled early in 2000 is probably the most spectacular example of this general tendency to disregard due process. “Administrative problems of this nature always exist,” simply said the prime minister.44
Collaborative federalism is hierarchical. In fact, Canadian federalism has never been more centralized. Power is concentrated not only in Ottawa, but also at the top, in the prime minister’s “court.” Collaboration occurs, but it is the collaboration of rowing agents who follow the indications of a steering principal whose behaviour is difficult to understand, let alone predict. Negotiations do take place, but they do not involve more or less equal partners seeking to reduce uncertainty. They are defined, instead, by the most powerful player, able to induce others to abide by rules that it can change at will. Collaboration, of course, also takes place without Quebec, the only government that would challenge federal hegemony in a fundamental way.
La différence est un caillou dans l’engrenage de la nation.
Gérard Bouchard, 199945
Quebec has never been so marginalized in the Canadian federation. In the past, the government of Quebec often maintained a distance in intergovernmental negotiations, but it always remained a critical player. Even in 1981-82, it was only with hesitation and a certain malaise that the provinces, specialists, and Canadian public opinion approved a federal constitutional approach that bypassed the government of Quebec. With the Social Union Framework Agreement, a deal without Quebec was reached in just a few days, with little afterthought. More generally, collaborative arrangements progress unabated, with only a footnote to indicate the absence of a central province. In December 1999, a federal clarity bill was introduced to empower the federal government to disallow the positive results of a Quebec referendum that would put forward any question mentioning “economic or political arrangements with Canada” or that would produce a majority that Ottawa would deem unclear, in light of any “matters or circumstances it considers to be relevant.”46 This bill, which challenges the powers of Quebec’s National Assembly and presents the Quebec electorate as unsophisticated, was broadly supported across Canada, including by the main opposition parties. If this last year is indicative, collaborative federalism does not seem to bode well for Quebeckers.
In English-Canada, some observers nevertheless see a silver lining in this evolution. Roger Gibbins, for one, appears particularly optimistic. The Social Union Framework Agreement, he writes, “clearly sets Quebec apart from the other provinces.” While the other provinces will engage in collaboration with the federal government and abide by its conditions, Quebec “will stand apart, its jurisdictional integrity uncompromised.” The Quebec government may still harmonize its programs with those of other provinces, “but it is under no obligation to do so” and can remain autonomous “with no financial risks.” Quebec’s “political clout in Ottawa and the remaining threat of separatism” gives the province “most favoured nation status in its bilateral, one might say binational, negotiations with the federal government.”47 The result, concludes Gibbins, is remarkable:
… the Prime Minister has achieved for Quebec what the majority of Quebec nationalists have sought for the past 30 years – a distinct position within the Canadian federal system in which Quebec is not a province like the others but rather has the de facto status of a separate national community, dealing one-on-one with the government of Canada.
… Quebec maintains its autonomy, is able to block constitutional change, and is able to use its political leverage in Ottawa to ensure that no financial costs are imposed. … The Prime Minister may have delivered on the ultimate paradox: an independent Quebec within a strong Canada.48
The lack of enthusiasm manifested by Quebeckers raises some doubts about Gibbins’ analysis. In Quebec, the most positive assessments of the February 4, 1999 agreement concluded that it basically “changed nothing.”49 At the same time, the Social Union Framework Agreement undeniably introduces some de facto asymmetry. Could it point in the right direction? Could it lead, incrementally, to a situation more representative of the dual character of the Canadian federation? Probably not.
First, in a federation as in any institutional arrangements, the formal and informal rules of the game matter, and the capacity to change these rules matter even more. All sides agree that the Social Union Framework Agreement is an important change in the rules of the game, since it circumscribes the use of the federal spending power in a novel way. The same is true for the National Child Benefit, which concerns the provinces as well as the federal government, and for the Clarity Bill, which introduces new rules for a future Quebec referendum. In all these cases, the federal and the provincial governments did not seek genuinely the approval of the Quebec government, or even of the official opposition in the National Assembly. This approval hardly seemed to matter. Regardless of the consequences, this capacity to alter the rules unilaterally poses a problem.
A free and respectful multinational society, argues James Tully, allows changes in the rules of mutual recognition, but does not permit unilateral changes in these rules. In Canada, Quebec is not free because constitutional amendments can be adopted without its consent, because it cannot in practice seek constitutional recognition, and because it would be bound, following a referendum, by constitutional rules it has not approved.50 This is not simply a symbolic or abstract constitutional question. As Tully notes, debates about recognition also define power relations in a society.51 On a series of questions, Canada now marches on as if Quebec did not exist or did not matter. The Social Union Framework Agreement is a case in point. As with the constitution, Quebec will be bound by an agreement it did not demand and did not approve. No matter how the Quebec government uses the situation to act autonomously, the outcome has more to do with domination than with freedom. In the process, the little trust that had been built between the provinces was also seriously undermined.52 Nothing prevents future changes that would be inimical to Quebec.
Second, Gibbins’ argument on de facto asymmetry underestimates the importance of the Social Union Framework Agreement in legitimizing the federal spending power. Even though the Agreement is only an administrative, time-limited document, it provides the first explicit recognition, by the provinces, of the legitimacy of federal expenditures and standards in areas under their jurisdiction. The fact that this recognition has no legal status is secondary because the issues surrounding the spending power are not primarily legal or constitutional. In most cases, the federal government simply cannot use its spending power without the collaboration of the provinces, since they play a critical role in the definition and implementation of social policy.53 This is true even for direct transfers to individuals, as can be seen with the Millennium Scholarship Fund or the National Child Benefit, which cannot be implemented without provincial collaboration. In such cases, relative power and negotiations matter more than legal constraints. In a discussion of the Meech Lake Accord, Keith Banting made similar observations and concluded that a consensus on the use of the spending power would strengthen this federal prerogative “far more than a favourable ruling of the Supreme Court ever could.”54 The Social Union Framework Agreement fell short of a consensus, but it came as close as ever to the conditions enunciated by Banting. The balance of power has shifted in favour of the federal government, and this certainly does not make Quebec more autonomous in social policy. Indeed, the National Child Benefit indicates how Quebec, even more so than the provinces that participate in the discussions, can become a policy-taker in areas within its own jurisdictions.
Third, Gibbins underestimates the contribution of the Social Union Framework Agreement to a process that is reshaping social policy in Canada. More will be said on this question in the next section, but one example is worth mentioning. With respect to mobility within Canada, the Framework Agreement goes much beyond the previous inter-provincial position and calls for the elimination of “any residency-based policies or practices which constrain access to post-secondary education, training, health and social services and social assistance unless they can be demonstrated to be reasonable and consistent with the principles of the Social Union Framework.”55 A new legal norm, more social, is created that goes beyond the logic of free trade and imposes new obligations and constraints on the provinces. Quebec is not a signatory, but this new framework is likely to affect it nevertheless, because it will impose the re-negotiation of agreements such as the agreement on Internal Trade, which Quebec has signed.56 In other words, a new pan-Canadian social framework is emerging and it is unlikely that Quebec can remain unaffected by its norms and constraints.
While it is true that collaborative federalism with a footnote creates a form of de facto asymmetry, it is not a form of asymmetry that responds to Quebec’s demands for recognition and autonomy. On the contrary, this new brand of federalism changes the rules of the game without the consent of the Quebec government, it reinforces the federal spending power, and it contributes to advance a new pan-Canadian vision of social policy that will affect Quebec, with or without its approval. The only autonomy that is enhanced for Quebec is the autonomy of the footnote, the negative autonomy of the non-participant. The Quebec government can be an irritant in the workings of Canadian federalism, a grain of sand in the mechanics of Canadian nationalism, but this is hardly an achievement worth celebrating.
The problem with collaborative federalism with a footnote is less a lack of asymmetry than a lack of federalism. Canada was created as a federation precisely because federalism allowed different peoples to live together and, at the same time, separately. More than provinces, the peoples involved were, initially, the peoples of Lower and Upper Canada, the French and the English Canadians. The need to accommodate the two remains the most critical dimension of federalism in Canada. When they opted for the politics of the footnote, the governments of this country chose an easy way out, but left unattended this central imperative, which made federalism necessary in Canada.
Asymmetry or reconciliation will not be obtained as unplanned bonuses for signing an agreement without Quebec. In the 1960s, Quebec obtained a de facto special status in social policy, but lost it a few years later when the federal government decided to erase anything that suggested that there were two distinct nations in Canada.57 Existential questions such as this one are unlikely to be settled “by stealth,” to use an expression familiar to social policy analysts. The issue that makes asymmetry necessary is not social policy autonomy per se, but rather recognition, and recognition cannot be granted unknowingly, by accident.
I believe we know what is required to address the main problems that we face at this moment.
Allan Rock, Minister of Health, January 27, 200058
The Social Union Framework Agreement and collaborative federalism are often presented as alternatives to the old, presumably dated, debates about the division of powers or about centralization and decentralization.59 From now on, the test of success would be less whether intergovernmental arrangements seem to work for governments, and more whether they produce satisfactory results for citizens. This reasoning, often associated with the notion of subsidiarity, is a very common form of wishful thinking. Policies, however, are never detached from processes. Collaborative federalism with a footnote does not simply “work for Canadians.” It makes some policy choices easier than others. It may well contribute, in particular, to develop Canadian social policies around persons and through income taxes rather than around places and through services. Whether or not this is the best, or only, possible outcome appears highly debatable.
In recent years, federal social policies and social policy instruments have changed significantly. In 1995, the creation of the Canada Health and Social Transfer pooled together, will little conditions except for health services, the transfers to the provinces devoted to health, social assistance and social services, and post-secondary education. In February 1999, a new, per capita, rule was designed for these pooled social transfers. Concretely, this new rule implied that the federal government had simply left the social assistance field, the least popular and least rewarding politically of its previous responsibilities. Nominally, some transfers are still associated with social assistance, but CHST transfers are now pooled and blind to social conditions and they can be used for practically any aim. Per capita, Alberta will now get as much social assistance money as, say, Quebec or Newfoundland. One could argue that equalization payments correct for this, but equalization is not a social policy; it is simply a means to provide the different provincial governments with comparable resources. In a context where employment insurance covers less and less unemployed persons, the federal government actually withdraws transfers that used to be devoted to social assistance in high unemployment provinces. Once it is on a per capita basis and blind to social conditions, it is hard to see a social rationale for the Canada Health and Social Transfer.
This movement away from social assistance transfers, along with the creation of the National Child Benefit, of the Millenium Scholarship Fund and of a host of boutique programmes, can be associated to an evolution that John Kincaid, an American federalism scholar, has called the movement from places to persons.60 The Canadian government, like its American counterpart, has become less interested in transfers to the provinces and more attracted by direct or fiscal transfers to groups or individuals. In health and post-secondary education, such a transition is more difficult because the federal government cannot run hospitals or universities. Still, the trend can be observed in Allan Rock’s various efforts to reaffirm federal leadership and visibility in health policy, as well as in the creation of various boutique programmes supporting research, innovation or higher education. In all these cases, the federal government attempts to circumvent the provinces, to reach directly persons and institutions.
In social assistance, the move is decisive. Provinces will receive smaller transfers, unrelated to their social situation, but individuals will receive checks with a large maple leaf on the front. The federal money will be for their children. If their general financial situation deteriorates, however, it will remain their province’s responsibilities. The provinces will also have to think twice before providing job training to these persons. If they have not qualified for employment insurance in recent years, an increasingly likely scenario, no federal money will go toward their training. The federal government, after all, must be accountable for the funds it collects through employment insurance contributions. Other than miscellaneous federal projects, they should not be used for purposes unrelated to unemployment. Canadians will feel reassured knowing that Human Resources Development keeps a watchful eye on free-spending provinces, which may be tempted to divest employment insurance money to train undeserving social assistance recipients.
It is too early to evaluate the full consequences of this general move from places to persons. Because the federal government can more easily send checks than provide social services and is more adept at promoting negative (removing barriers) than positive integration (creating institutions), the impact will be redistributive but not necessarily innovative.61 Various persons will see their situation improved or worsened, but better practices or institutions will not necessarily follow. Fiscal deductions for child care, for instance, may support community-based or private services; as they exist now, they discourage the universal model adopted in Quebec.
Canada will have a more integrated market governed by a more present central state, but it may have difficulties in fostering various models of community-based, integrated and enabling services.62 Indeed, the concentration of power identified above implies the prime minister and the minister of Finance now stand as the almost unchallenged social policy leaders. Court government will tend to circumvent the provinces, the main social actors and a good part of the policy community.63 Finance will promote income and tax-based policies aimed at individuals, and particularly at individual children, at the expense of community-based and service-oriented options. Other voices will be heard in and around Ottawa, regarding health care in particular, but they are unlikely to be as powerful in setting the agenda.
In a more general way, collaborative federalism with a footnote corresponds to a certain idea of the politics of the “Third Way.” Important ideas are currently developed by social-democrats under the Third Way umbrella. There is a tendency, however, to replace Margaret Thatcher’s TINA (“There is no alternative”) by a Third Way TIBOO (“There is but only one”). The new discourse on social investment, which is put forward at the OECD and in most capitals, tends to present choices as mostly technical and best defined by central governments. In the social-democratic tradition, however, there is a powerful current that emphasizes local autonomy, genuine empowerment, and the superiority of reforms defined from below.64 In Canada, collaborative federalism with a footnote does not point in this direction. One hope is that Quebec, which is more progressive than other provinces on these questions, will forge ahead and create something like a distinct model, which could be emulated. This is the case, in my opinion, with family policy. But, again, the social union does not make such an evolution easy.
Deviens-tu c’que t’as voulu?/Deviens-tu c’que t’avais vu?/Deviens-tu c’que t’aurais pu?/T’as-tu fait c’qu’y aurait fallu?
Daniel Boucher, Deviens-tu c’que t’as voulu?
The Social Union Framework Agreement of February 4, 1999 was particularly regrettable because it was a missed opportunity, a lost occasion to take advantage of important concessions that had been accepted by the different provinces. Why was this occasion missed? Why was is so important to sign a deal rapidly, without taking the time to build on these compromises? Apparently, a deal had to be signed because Jean Chrétien and his court had had enough. They wanted an agreement right away, and had the wherewithal to draft it and pay for it. As a result, what can be called, for lack of a better expression, collaborative federalism with a footnote, was entrenched, institutionalized.
This intergovernmental arrangement is hierarchical in at least three ways. First, it was defined in Ottawa and pushed forward in a negotiation that had more to do with hegemonic cooperation than with multilateralism. Second, it is inspired by the falsely non-hierarchical logic of the new public management, where the center steers and the provinces row. Third, it constitutes an intergovernmental regime that corresponds well to the logic of court government, a system of government that concentrates power at the top, that emphasizes short-term solutions, and that pays little attention to formal rules and due process.
Despite the footnote, collaborative federalism does not deliver meaningful asymmetry. The Social Union Framework Agreement demonstrated, once again, how the rules of the game can be changed without the consent of Quebec. It also reinforced the legitimacy of one of Ottawa’s main tools in social policy, the spending power. Finally, it further defined and institutionalized a pan-Canadian vision of social policy that leaves little room for autonomy or difference. This intergovernmental regime is, to some extent, asymmetric, but it does not constitute asymmetric federalism, because it does not respect the federal principle, as it was defined in Canadian history.
For social policy, only time will tell what the outcome will be, but the Department of Finance is more likely to have a say in delineating this outcome than provinces or communities. On these questions, Finance is likely to ignore what Swedish political scientist Bo Rothstein has called “the central message of Italian double entry bookkeeping,” namely that, in social policy, redistribution is defined not only by what individuals receive but also by what they contribute.65 In other words, the contemporary calculus of individual benefits, very present in federal discussions of family policy, tend to lose sight of the broader picture. Universal programs or services are helpful not only because they create equal citizenship rights, but also because they make citizens more willing to pay a fair share of taxes.
As a way to conclude, it is worth bringing back the new public management image of steering and rowing. As explained above, this image provides a fair representation of collaborative federalism. It constitutes, however, a very poor and uninspiring representation for a country like Canada, especially if we add that one partner does not sing the same rowing song and that the hand at the helm appears strong but capricious. A not so different, but much more interesting image is offered by Bill Reid’s sculpture, The spirit of Haida Gwaii, as evoked by James Tully. Reid explains his sculpture of a black canoe carrying very diverse partners in these words:
Here we are at last, a long way from Haida Gwaii, not too sure where we are or where we’re going, still squabbling and vying for position in the boat, but somehow managing to appear to be heading in some direction. At least the paddles are together, and the man in the middle seems to have some vision of what’s to come.66
In my opinion, this sophisticated and open perspective is much more relevant to Canadian federalism than the steering and rowing image of the new public management. It cannot be found, however, in the Social Union Framework Agreement.
Is such a vision now out of reach for Canada? During the discussion, at a December 1999 IRPP roundtable on Canadian federalism, Claude Ryan observed that in the last decades he had attended so many of these well-intended meetings on federalism. It takes a lot of patience, he said in substance, to keep reiterating, year after year and without success, the same basic arguments. This was a sad comment, from a respected Quebec federalist who has devoted all of his professional life to public affairs. Being less committed to the Canadian idea, and less patient I think, than Claude Ryan, I wondered at the time where I would be and what I would say at his age. I am not sure I want (or you want me) to answer this question.
An earlier version of this Working Paper was presented at the Saskatchewan Institute of Public Policy Forum on “Perspectives and Directions: The Social Union Framework Agreement,” held in Regina on February 2-4, 2000. I am grateful to Raymond B. Blake and Hugh Segal for their invitation, and to the Forum participants for their helpful comments. I thank, in particular, Keith Banting, David Cameron, Jane Jenson, Harvey Lazar, Denis Saint-Martin, and Richard Simeon for their specific comments and suggestions.