Making Federalism Work: Intergovernmental Coordination and Institutional Capacity

Forum of Federations Forum des fédérations MAKING FEDERALISM WORK: Intergovernmental Coordination and Institutional capacity INTRODUCTION Federalism in Canada Regional Diversity The Design of Central Institutions Provincial Powers Accounting for Success Forum of Federations 700-325 Dalhousie Street Ottawa, Ontario K1N 7G2 Canada Tel.: (613) 244-3360 Fax: (613) 244-3372 Email: Forum des fédérations 700-325, rue Dalhousie Ottawa (Ontario) K1N 7G2 Canada Tél.: (613) 244-3360 Téléc.: (613) 244-3372 Courrier électronique : MAKING FEDERALISM WORK: Intergovernmental Coordination and Institutional capacity Notes for a presentation Forum of Federations International seminar on Modernization and Fiscal Federalism: Alternatives to tax Wars: Sustainable Economic Development in Brazil Sao Paulo, Brazil June 21-22, 2001-06-14 By Richard Simeon Professor of Political Science and Law University of Toronto ( ) INTRODUCTION Despite the many and obvious differences in history, economic development, and institutions, my quick crash course on Brazilian federalism has convinced me that Canada and Brazil have much in common. • Both are highly decentralized relative to other federal systems. • Both are regionally diverse, with major disparities among regions, generating important regional tensions. • Both have recently undergone major fiscal crises with important implications for federal finance • Both have had to adjust to the implications – economic, political, even cultural – of globalization • Both have therefore faced the challenge of balancing regional and national interests and of managing the competition between among the states or provinces, and between them and the central government. For both countries the larger question is: how do we realize the benefits of federalism in terms of experiment, innovation, and responsiveness to diverse interests; while avoiding the costs of excessive and destructive competition, and the inability to coordinate on key economic and social issues? Or, as Richard Bird puts it, ‘How do we avoid the tendency of subnational governments to spend too much, to tax too little, and generally to behave in a fiscally irresponsible fashion?’ The premise of this conference is that Brazilian federalism does not score well in these respects. Brazilians talk about ‘fiscal wars;’ and of Brazil, as one writer puts it, as ‘a deviant case of market-destroying federalism; and of the need to ‘redefine our concept of federalism.’ The argument I want to make in this presentation is that an examination of the case of Canada demonstrates the potential for a similarly destructive pattern in our country too.. But, while many of the same issues and tensions that plague Brazil arise in Canada, they do so in a much more muted and manageable way. I begin by exploring some reasons why this potential exists, then examine the Canadian record, and conclude with some explanations, and some possible lessons for Brazil. My analysis closely follows and expands on some excellent recent work on Canadian and comparative fiscal federalism by Richard Bird and his colleagues, including Francois Vaillancourt. Federal system display a wide variety of mechanisms designed to achieve coordination and harmonization, and to minimize destructive conflict among governments. These may include: • Constitutional limits on one or both levels of government, such as constitutional guarantees of individual mobility, or of free movement of goods and services. • Federal government powers to limit or constrain provincial actions that are deemed to be harmful. The 1996 South African Constitution is perhaps the extreme case of this approach in modern federalist systems. • ‘Expert’ bodies with powers either to make recommendations, or to make rules. The Finance and Fiscal Commission, which advises on the ‘equitable share’ of revenues that should go to provincial and local governments is an example of the first; the Australian Commonwealth Grants Commission that must approve state borrowing is an example of the latter. • Negotiated codes of conduct and harmonization agreements. This is the dominant Canadian pattern. • ‘Self-restraint’ based on transparent and accountable political practice and on bureaucratic norms and values. • Market discipline. Different federations employ these mechanisms in different ways, depending on their constitutional design and the play of political forces. We also need to clarify what we mean by excessive or destructive competition among states and provinces. A basic premise of federalism is that it is designed to stimulate some forms of competition. Jurisdictions are expected to use whatever instruments they have at their disposal to respond to varying citizen preferences, and to promote their economic development. But another basic premise is federal ‘comity’ – the idea that states and provinces should not engage in behaviour that harms others, or that harms the national economy as a whole. So the question is: what is ‘healthy’ competition; what is ‘unhealthy competition?’ There are no clear answers. Some state/provincial actions seem clearly legitimate – including such measures as development of education systems to produce the most highly qualified work force; or provision of excellent infrastructure. Other actions are just as clearly a violation of federal comity – such as policies that impose externalities on other jurisdictions. But there is a large grey area in between. This is a question that needs further exploration. Federalism in Canada Why might the potential for unmanageable intergovernmental and interregional conflict in Canada be similar to Brazil’s? Several social, economic, and institutional features of the country point in that direction. Regional Diversity First, Canada is a highly regionalized country. In economic terms, this means that economic disparities between richer and poorer provinces, while smaller than those in Brazil, remain large and persistent. There are also strong and persistent differences in the economic structure, and therefore in the interests of different provinces, with manufacturing and finance based largely in the large central provinces of Ontario and Quebec, and resource based industries dominant in Western and Atlantic Canada. Historically, debates over trade policy, energy policy and the like have pitted Canada’s regions against each other, sustaining a strong pattern of regional grievances which continue to influence the current political climate and our responses to current challenges. These differences, along with differing patterns of settlement, have also fuelled strong provincial identities, centred on the provincial governments. This leads to pressures for greater provincial autonomy and the widespread belief among citizens in the western provinces in particular, that they are the ‘losers’ from the Confederation bargain, and that their interests are not given sufficient weight by the federal government. They have also generated a number of regional protest movements and parties. By far the most important of these differences is of course the presence in Canada of Quebec. In addition to being a province like the others in formal terms, Quebec is also the homeland of a distinct, largely French-speaking community, which has increasingly defined itself as a nation. Quebec, therefore, has always led the way in pushing for greater provincial powers, and for limiting the federal ability to encroach on provincial areas of jurisdiction. Since the 1960s, Quebec has been engaged in a ‘nation-building’ project. The major debate within Quebec has been whether the nation’s interests are best served by remaining within the Canadian federation, or by becoming a sovereign country. In a 1995 provincial referendum on the question, the population was virtually evenly divided. While the salience of nationalist aspirations has faded somewhat since then, the question is far from being resolved. The modern forces of globalization interact with these historical inheritances. Globalization in Canada plays out largely in terms of Canada’ s economic integration with the United States, a process greatly deepened by the effects of the North American Free Trade Agreement. More than 40 per cent of Canada’s GDP depends on external trade; and more than 85 per cent of its exports go to the United States. Many commentators, most notably Thomas Courchene, argue that globalization accentuates the centrifugal forces in Canadian federalism. He has coined the term ‘glocalization,’ suggesting that globalization erodes the capacity of the national government, and hence its importance In the minds of citizens, more than it does the provinces. Globalization also means that the East-West ties of trade, are weakened. Increasingly, Canadians trade less with each other than they do across the Canada-US border. For example, historically, Ontario’s manufacturing and financial industries were based on building trade within Canada behind a high tariff wall. Free trade fundamentally reoriented its economy on north-south lines. Canadian regions, therefore, are differentially integrated into the North American economy. This makes the idea of a single ‘national’ economic policy seem remote. It also makes it harder to maintain what Courchene calls the east-west ‘social policy railway,’ which has become such an important part of Canadian identity. Under these conditions, the predictions are that provincial interests will increasingly diverge, that competition between provinces for economic growth will increase, that the commitment to interregional sharing, to Canada as a ‘sharing community,’ will decline, and that the clout of the federal government will diminish. In such conditions, Courchene suggests that a province like Ontario has strong incentives to move itself to a ‘region-state,’ disentangled from the federal government, and far less integrated with the other Canadian provinces. Similarly, some commentators in energy-rich Alberta argue that the province should build a ‘wall’ isolating it from the potential depredations of Ottawa and central Canada. Such predictions may well be exaggerated. Some economists have shown that the intensity of economic linkages within Canada remains orders of magnitude greater than links with the US. There is little evidence of a ‘rush to the bottom’ in provincial social policies; nor is the commitment of governments and citizens to equalization weakening. Most Canadians retain strong identification with the national government and society – even including most of those who support sovereignty for Quebec. There is little evidence of public support for a ‘fortress’ Alberta or Ontario – the latter sends a phalanx of Members of Parliament to Ottawa wearing the colours of the federal Liberals, the chief defender of a centralized and united Canada. Indeed, it can plausibly be argued that in some respects globalization might be a centralizing force in Canadian federalism. While provinces are assertive in expressing their economic interests in the international arena, and call for close consultation in international trade negotiations, it remains that Ottawa is constitutionally responsible for international trade and foreign policy, and plays the central role in managing the interface between Canada and the global environment. Canada’s influence in world economic forums depends at least in part on the country’ ability to speak with a single rather than a divided voice. The Design of Central Institutions The institutional structure of Canadian federalism also seems to be almost perversely designed to accentuate intergovernmental and interregional conflict. First, the federal government itself is poorly positioned to represent and accommodate Canada’s regional diversity. In most federations, the Upper House or Senate is designed to represent subnational governments in the central arena. This can be done either through an elected Senate, with Senate representation counteracting the purely proportional lower house – an issue especially important for Canada, where parliamentary representation is dominated by the large numbers from Quebec and Ontario (which together account for 61.9 per cent of the population). Or it can be done through the direct participation of representatives of the provincial government in the national parliament, as in Germany, and now South Africa. The reality, however, is that the Canadian Senate does neither of these things. Its members are appointed solely by the federal executive, rendering it largely a patronage body with virtually no legitimacy, no links with provincial governments or their electorates, and hence no ability to mediate between national and regional interests. Moreover, Canada is an example of ‘parliamentary federalism,’ combining multilevel government with the British or Westminster model. This embodies single party executives, strict party discipline, and a concentration of power in the hands of the Prime Minister and cabinet. This means that individual MPs have little ability to act as representatives or advocates of local and regional interests (except in the relative secrecy of the party caucus). It also means that legislatures and their members play virtually no role in intergovernmental relations. In parliamentary federalism, the relation between federal provincial governments takes place in direct relations between senior executives, in Canada called ‘executive federalism’, or ‘federal-provincial diplomacy.’ The consequences of parliamentary federalism are reinforced by the regionalized character of the Canadian party system. In Canada, only one party, the long governing Liberals, can claim to be a fully national party, and even it has historic weaknesses, especially in western Canada. Today, the two major opposition parties are both regionally concentrated – the Bloc Québécois, supporting Quebec sovereignty in Quebec, and the Alliance Party of Canada, concentrated in the west, despite failed attempts to reorient itself as a national party. The regionalization of the party system is in turn reinforced by the effects of the electoral system, which exaggerates the seats won by narrow regionally based parties, and weakens parties who support is spread more nationally. The result is that major parts of the country feel frozen out of power in Ottawa: they are not represented in the governing party, where authority is concentrated; and the opposition parties they do support are virtually powerless in the Westminster system. It is not surprising, then, that citizens turn to their provincial governments as the ones that better represent them, and support their efforts to defend provincial interests in the arena of executive federalism. Thus Canada is – in sharp contrast to ‘integrated’ federations such as Germany — a ‘divided’ federalism with few effective political links knitting the two orders of government together. Indeed the pattern continues in other ways. In many cases distinct provincial parties hold local office; in others there are few links between federal and provincial parties even of the same name. Typically different parties hold office at each level. Most politicians make their career at one level or another: there is little movement between them. Federal and provincial elections are held at different times, and thus involve quite different issues and concerns. While it would be a gross exaggeration to argue that national political leaders fail to consider regional interests, the Canadian federal design greatly inhibits their ability to do so, especially as the disparity in numbers means that westerners tend to see federal ‘sensitivity’ to the regions as little more than a preference for the interests of the large provinces with the most parliamentary seats. The Brazilian case might lead me to temper this analysis a bit. There can be too much regional influence at the centre, turning it into an area for log-rolling local barons, and eroding its capacity to focus on national issues, as appears to be the case in Brazil, and to a lesser extent in the United States. Canada avoids the danger of a central government penetrated and colonized by parochial interests. The provinces can put pressure on the federal government. But they have no influence over individual members of parliament, and no way to undermine the authority of the central executive., Provincial Powers Partly as a consequence of these characteristics of the national system, provincial governments are independent, autonomous, assertive – and powerful. The original constitution of 1867 seemed to give predominant powers to the centre. Not only did it receive the most important governmental powers of the time – trade and commerce, money and banking, etc., but also it had wide powers to override provincial actions, for example through a ‘disallowance ‘ power; and through a general power to make laws for the ‘peace, order and good government’ of Canada. But judicial interpretation and evolving practice have eroded most of these powers. For example, the peace, order power has been restricted to an emergency power; unlike the American experience, the trade and commerce power was interpreted to retain considerable provincial influence over their local economies; the federal power to make treaties is constrained by its inability to make commitments in areas of provincial jurisdiction; the critical federal responsibility for tariffs has virtually disappeared in an era of global free trade; and the disallowance power is generally agreed to be a constitutional dead letter. The constitution also gave extensive powers to the provinces, and their importance has increased over time. For example, provincial ownership of ‘crown lands’ is the basis for their ownership and control of natural resources – forests, hydro power; oil and gas. Their jurisdiction over health, education and welfare, minor areas of public spending back in 1867, now means that they have primary jurisdiction over the major areas of expenditure in modern government. A general power over ‘property and civil rights’ sustains extensive provincial regulation of the economy and the environment, including labour markets, major aspects of financial markets, environmental practices, and the like. Provinces also have jurisdiction over local government, and have used this authority to exercise close control over all aspects of municipal life – their boundaries, powers, revenues, and so on. Provincial governments have also successfully blocked the development of close linkages between the central and local government spheres. The result is that while Canada is one of the world’s most decentralized federations in terms of relations between centre and provinces, it is one of the most centralized with respect to the autonomy of local government. This is one of the major differences between Canadian and Brazilian federalism. Many contemporary observers in Canada see major cities as the chief actors in a global political system and look to a revitalization of democratic politics at the local level. But it is unlikely to happen in the foreseeable future. The constitution also gives the provinces the power to raise revenues through any form of direct tax – giving them the opportunity to impose a wide variety of taxes — on income, sales, real property and the like. It places no constraints on the ability of provinces to borrow either in domestic or foreign markets. Provinces, at least until recently, have made considerable use of their ability to establish government-owned enterprises. The constitutional provisions with respect to the economic union also place relatively few constraints on the provinces. As I noted, and unlike the US, the trade and commerce power does not give a sweeping mandate for the federal government to regulate provincial behaviour. Section 121, the ‘economic union’ clause of the constitution, has been interpreted narrowly, leaving provinces free to impose a large number of non-tariff barriers. Many attempts to strengthen the economic union and to limit provincial ability to discriminate in favour of local suppliers have been attempted. Some proposals were aimed at strengthening the federal trade and commerce power, making Ottawa the guardian of the economic union; others sought to strengthen the judicially enforceable section 121. All failed to secure agreement in constitutional negotiations. Instead, Ottawa and the provinces came to an informal agreement, the Agreement on Internal Trade, designed to strengthen the internal common market, but it leaves many of the provincial preferences intact. The conclusion, then, is that neither the constitution, nor the power of the federal government, significantly constrains potentially destructive provincial competition. Provinces have also increased their fiscal autonomy over time. In 1970, total federal revenues equalled 16.7 per cent of GDP, provinces 17.6 per cent, and municipalities 9.2 per cent. In 1999, federal revenues had risen marginally to 18.9 per cent of GDP; provincial revenues had risen to 22.7 per cent, and municipal revenues were 8.6 per cent. The federal government accounts for 45 per cent of all revenues, and 37 per cent of government spending. By contrast, the United States federal government collects 61 per cent of all revenues; and the Brazilian government 36.5 per cent. There has been a progressive decentralization of federal control over income taxes. During the crisis of world war two, provinces delegated their power to tax income to Ottawa, in return for cash payments – called the ‘tax rental system. By the 1960s, provincial taxation resumed as a result of Ottawa making room through ‘tax abatements.’ Ottawa initially remained the collection agent, and fixed the permissible provincial rate. Then provinces were freed to set their own rates, but still using the federal base. Then they were permitted to impose various tax credit arrangements. And, from 1999 they have had even greater flexibility to use the tax system for their own purposes by shifting from the ‘tax on tax,’ to a ‘tax on income.’ Throughout Ottawa has continued to collect taxes on behalf of all governments except Quebec through a set of tax collection agreements (TCA’s), but its control of provincial tax policy and its ability to use the agreements to ensure tax harmony has been successively weakened, to the extent that some now wonder about the continued viability of the tax collection arrangements. One consequence is wide variation in provincial income taxes – from 69 per cent of the federal personal income tax in poorer Newfoundland to 39.5 per cent in Alberta. Similarly, large intergovernmental transfers have long been a major feature of Canadian federalism. Federal transfers to the provinces make up 13 percent of federal spending (compared with 30 per cent in the US), and they make up anywhere between 7 and 30 per cent of provincial revenues. In response to provincial pressures, Ottawa’s ability to use transfers to shape provincial policy has greatly declined. The largest single transfer is the Equalization program, designed to ensure, as a constitutional amendment of 1982 states, that all provinces are able to provide ‘comparable levels of public services at comparable levels of taxation.’ This transfer is entirely unconditional; moreover, the total spending that Ottawa is responsible for under the program has been driven in large part by provincial revenue decisions that Ottawa does not control. While fiscal restraint and the increasing divergence among provinces has placed considerable pressure on the equalization system, it is one of the great successes of Canadian federalism. It ensures that even poorer provinces have the revenue base to benefit from the autonomy that federalism provides. Using 100 as represent the average, before equalization, the revenue raising capacity of provinces ranges between 62 and 161, a ratio of more than two to one. After equalization all provinces score a minimum of 96, and only Alberta with its huge oil and gas revenues differs significantly from the average. The equalization system also places a restraint on competitive actions by the poorer governments (7 of the 10 provinces). Overly aggressive programs could easily undermine support for equalization in the richer provinces that bear the cost. Following World War Two, earmarked transfers – ‘conditional grants’ — became another important feature of the system, financing the federal share of health, education, welfare and post-secondary education. They were enabled by the federal ‘spending power,‘ its ability to spend money in any area including those under provincial jurisdiction. While not specifically included the in list of federal powers, this power is implicit in other parts of the constitution. This is a powerful instrument for federal influence over provincial activities, but the conditions the federal government attach to its funds have steadily declined, again in response to provincial pressure. In 1996, a wide variety of programs were consolidated into a single block grant, the Canada Health and Social Transfer (CHST). The only major conditions that remained were broad requirements respecting universality and public provision in the health care field, under the Canada Health Act. One estimate suggests that only 4.3 per cent of total federal transfers to the provinces is conditional in any meaningful sense. Ottawa gets very little policy influence for its federal dollars. Thus it has neither carrots nor sticks to shape provincial behaviour. Here is the picture I have drawn so far: a deeply divided federal society, with globalization potentially adding to the centrifugal pressures; a federal government poorly equipped to represent this diversity, and with few instruments at its disposal to constrain provincial actions; a set of strong, autonomous provincial governments, jealous of their powers, seeing themselves as fully equal partners with the federal government in Canadian governance. These are provinces whose share of total government spending, revenue, and public employment has grown sharply since the height of postwar centralization in the mid 1950s, and is now greater than any other federation in the world except Switzerland. All this, I suggest might plausibly be seen as a recipe for uncontrolled intergovernmental competition and conflict as severe as that in Brazil. Indeed, there are many grounds for worry about and criticism of the operation of the Canadian federal system. Critics continue to worry about the ‘difficulties of divided jurisdiction,’ especially in areas of overlapping responsibility such as environment and social policy, and call for various measures either to ‘rebalance’ the division of powers in light of current needs, or to improve the mechanisms for intergovernmental coordination. There is some evidence of competitive interprovincial bidding for investment and jobs. Quebec has gone further than most other provinces in developing an explicit, state-led industrial policy to mobilize Quebec resources, in a form of economic nation-building, described as ‘Quebec Inc.’ But all other provinces also see it as their role to promote their local economic development. Many worry that the harmonization in the tax system may be unravelling with the pressure on the tax collection agreements. Provinces worry about a vertical imbalance, as costs in areas of provincial responsibility such as health care continue to escalate. There are strong pressures to reform the equalization system, and the disparity in provincial resource revenues has the further potential of exacerbating conflict. In addition, many flaws have been identified in the government to government form of intergovernmental relations known as ‘executive federalism.’ Some lament the costs and delays of reaching agreements that may turn out to be little more than a ‘lowest common denominator’ – using a term we have borrowed from Europe – the’ joint decision trap.’ Others worry about the ‘democratic deficit’ – the lack of transparency and accountability in negotiations carried out largely behind closed doors. And yet others worry that substantive policy concerns become sacrificed to the political goals of rival government leaders, as they seek to win credit for popular measures and shift the blame for unpopular ones. There are in this system, too many intergovernmental ‘turf wars.’ Moreover, like all other countries, Canada has undergone a number of wrenching adjustments to external and domestic pressures, each of which has been considered a crisis at the time. In the aftermath of the depression and the Second World War, the question was whether federalism could adapt to the new economic roles of the state, and in particular whether it would stand in the way of building the welfare state. The conventional wisdom at the time was that only a powerful centre could take on the new social and economic roles of the states. Influential critics argued that federalism was ‘obsolete.’ In the 1960s, a ‘Quiet revolution’ in Quebec and the rise of Quebec nationalism led to worries about the very survival of the federation. In the 1970s, the end of the postwar boom coupled with the energy shocks of the time led to unprecedented conflict between central and western Canada, and between Ottawa and the energy-producing provinces over the control of development, prices and revenues. In the 1980’s came a set of challenges as great as those of the depression and war: the rise of debts and deficits, provoking massive shifts and downloading of costs and responsibilities from Ottawa to the provinces and provinces to local governments; the erosion of the welfare state; the rise of neo-liberal models dedicated to smaller government; and the new pressures brought on by globalization and free trade, pressures were perhaps even greater in Brazil. At the same time Canada was also adjusting to domestic change – to the increasingly multi-ethnic character of the country, to the changes in the role of women in society, and to the mobilization of Canada’s aboriginal peoples seeking recompense for past injustices and self-government for aboriginal communities. In all these cases, the institutions and practices of Canadian federalism proved resilient; the model was strained, but did not break. • The postwar welfare state was constructed as a federal-provincial enterprise, with social security and unemployment insurance moved by constitutional amendment to the federal level, and with federal dollars helping to finance the growth of provincial programs. • With Quebec, a combination of policies — on the one hand to increase provincial autonomy and resources, including a significant degree of special status, or asymmetry; and on the other hand to strengthen French-Canadian representation in national institutions — did not put an end to the sovereignist idea, but did direct the on-going debate into regular and peaceful political channels. • The energy crisis came to an end largely as a result of the drop in energy prices in the early 1980s, and of the election of a Conservative government in which westerners were strongly represented. It quickly dismantled the hated federal National Energy Program. • All governments were forced to respond to the later fiscal crisis. All reoriented their spending and taxing regimes in roughly parallel ways, with a few provinces taking the lead. As the crisis diminished, some of the pain of federal cuts in transfers was eased by restored funding. Debate remained intense in social policy, as provinces sought freedom from the conditions of the Canada Health Act. But all governments except Quebec came to a Social Union Framework Agreement in 1998, which set out broad guidelines for federal-provincial cooperation, and placed some limits on Ottawa’s power unilaterally to alter funding, or introduce new programs. Despite the inability to agree on a new constitutional framework for managing the economic union, the negotiated AIT has proven a reasonable second best. Competitive interprovincial bidding for investment remains limited. So there have been lots of battles in Canadian federalism, but few extended outright wars. However messy, acrimonious and drawn out they were, broad accommodations have been reached. The system has proved capable of adaptation to external circumstances and shifting views of the role of government. Accounting for Success What accounts for this (admittedly relative) success? In the absence of constitutional rules, or federal dominance, a varied array of factors seems to explain the successful functioning of federal systems. • First is an effectively operating system of intergovernmental relations than can operate to facilitate cooperative solutions to problems, and encourage self, rather than imposed, discipline. There are a great many flaws in the existing system of Canadian executive federalism, as I have mentioned. But it has major strengths as well. Cooperation has been achieved through intergovernmental negotiation on a number of major issues – internal trade, social policy, and environmental regulation, for example – where attempts at formal constitutional amendment have failed. Canadian governments, especially the provinces, are now committed — rhetorically if not always in practice – to a collaborative form of decision-making, emphasizing full reporting, best practices, shared provision of services and the like. Federal-provincial ministerial councils, which meet regularly, often under joint chairmanship, have been formed in a wide variety of policy areas. Some have performed very well, others less so. Such meetings can facilitate a common understanding of issues, and help build trust relationships among executives of different governments. With their very long plane rides to far flung countries, ‘Team Canada’ visits by the First Ministers seeking investment and markets, may be a particularly useful device for informal communication among First Ministers.. Of special note in Canada is the growing role of provincial-territorial, as distinct from federal-provincial, relationships. These take the form of an annual Premier’s conference, and a rotation among provincial premiers as their primary spokesperson each year. The Conference often delegates work on particular problems to sub-committees of ministers and officials. Regional groupings of premiers, in the Atlantic and Western provinces, have also been constituted; and, interestingly, both have formed regular links with nearby U.S. state leaders. The cynical view of the PT arena is that it is little more than an opportunity for provinces to ‘gang up’ on Ottawa. In part this is true. But in building trust and mutual understanding among provincial leaders, it can help develop common solutions to common problems, and to act as an informal restraint on overly competitive behaviour. Moreover a strong tradition of mutual deference to each others’ concerns has developed. Since the meetings bring together rich and poor, small and large, provinces, differences of interest that could easily arise tend to be mitigated. Moreover, through negotiated informal arrangements the system can allow a considerable degree of variation in provincial powers producing a de facto asymmetrical federalism. There are many examples of this in Canadian practice, even though all attempts to enshrine asymmetry in the constitution have been resoundingly defeated. • Without some underlying consensus, however, intergovernmental conferences can as easily become areas for deadlock as for compromise. This suggests another set of factors required to promote harmony and avoid destructive competition. These are largely cultural. Federalism depends not only on strong regional identities, but also on a strong sense of ‘vouloir vivre ensemble,’ of shared identity, of solidarity and a sense of mutual obligation to residents of other regions and provinces. In Canada, as I have noted, strong provincial identities coexist with strong national identities. Support for equalization remains very high, even in the wealthier provinces that pay the cost. But ideas are also substantive. Governments interact within the context of a larger climate of ideas. In the postwar period, citizens and governments across the country all agreed on the basic project of building the welfare state. Consequently, all (with the partial exception of Quebec) were willing to accept federal leadership at the time, and to practice a cooperative form of federalism. Similarly common agreement on the need to rein in debts and deficits and to limit the growth of government greatly constrained intergovernmental conflict more recently, even as huge cuts in federal transfers to the provinces were being made. By then as well, almost every government had come to reject an industrial policy based on targeted subsidies and tax expenditures as self-defeating. And widespread agreement on the broad goal of North American Free Trade muted provincial concerns about any diminution of their powers. The lesson is that federalism is most conflictual when governmental turf interests are aligned with different regional interests, and with major ideological differences. A common political culture and view of government across the country minimizes the frequency of such alignments. • A third set of factors is even further removed from the design of the federal system itself. One analysis of Brazil observes that in Brazil, ‘federalization preceded democratization.’ An open, responsive political process, fully accountable to citizens, is likely to be largely immune from the patronage-based clientelistic politics that seems to be at the root of many of Brazil’s fiscal wars. Canada is further along the democracy path (though not by any means there yet!) than Brazil. Similarly, a strong, competent, professional bureaucracy is likely to help restrain conflict, especially if, as in Canada, bureaucrats at both levels tend to share similar professional and programmatic values. • Finally, and perhaps most important, is the discipline imposed by the market itself. In a market as open and fully developed in Canada, irresponsible provincial, or federal, actions are likely to attract swift responses from investors and bond-rating agencies. This, Richard Bird suggests, is a powerful alternative to hierarchical rules, or even negotiated codes of conduct, in decentralized federal systems. Thus despite the potential for conflict I described at the start, I agree with Bird and Francois Vaillancourt when they conclude: ‘If one’s definition of an effective state is a state that can effectively deliver required services in a flexible way to a diverse population, Canada has thus done fairly well,’ even if its intergovernmental relations is often messy, noisy, unruly and have ‘consumed enormous political energy.’ Whether it is safe to project such a favourable analysis into the future remains unknown. There is a long list of items on the current intergovernmental agenda. Decentralizing forces remain strong, and some ask whether the ‘centre can hold.’ The impact of globalization has yet to fully play out. But it seems unlikely that future crises will be more daunting than the challenges Canadian federalism has been able to overcome in the past. Finally, what lessons might this analysis hold for Brazil? Perhaps not many, because I have suggested that success involves many factors not part of the federal design itself, such as a strong sense of common citizenship and commitment to sharing; and an effective democratic and bureaucratic state. We have had well over 100 years to fine tune our federalism; you are digesting major recent changes. But Brazil might be able to learn from Canadian experience in some areas. They could include the enormously successful Canadian equalization system (despite its current strains); the development and operation of the Agreement on Internal Trade, and the development of effective intergovernmental machinery, perhaps especially building on our practice of sectoral ministerial councils and inter-provincial linkages. Other countries suggest different means of mitigating the conflicts inherent in fiscal federalism – the Commonwealth Grants Commission in Australia, and the Finance and Fiscal Commission in South Africa come to mind. The Canadian case also suggests the desirability of adapting the federation to new needs through informal agreement, rather than by constitutional amendment, which is both very difficult to achieve, and if achieved, locks the system in. To conclude: with respect to fundamental issues of democracy, good governance and the resolution of conflict in divided societies, federalism is Janus-headed: it has both strengths and dangers. The trick is to find ways to find ways to build on the advantages and minimize the disadvantages.