Australia

Australia created a stand-alone equalisation programme in 1933. That year, the Commonwealth Grants Commission (CGC) was launched as an agency operating at arm’s length from the federal government (CGC 2008, 1). Following its creation in the mid-1930s, for almost 50 years, the CGC advised the Commonwealth Government on the allocation of special grants to at least one of the four lesser-off states (Queensland, Tasmania, South Australia and Western Australia). This logic of “special grants” was aimed at reducing territorial inequalities while helping poorer states to offer services similar to those that existed in wealthier states without having to impose substantially higher tax rates on their citizens (Else-Mitchell Reference Else-Mitchell1979, 162).

In the early 1980s, the Commonwealth and the six state governments decided that “a proportion of personal income tax revenue would be shared among the states in a way that equalized their financial capacities” (CGC 2008, 32). Since then, the CGC has been assessing the financial situation of the six Australian states instead of focusing exclusively on claimant (poorer) states, as was the case between 1933 and 1982.Footnote ² In 2000, the Commonwealth Government instituted a Goods and Services Tax (GST), of which proceeds are distributed to states (Morris Reference Morris2002).Footnote ³ While some of the GST money is allocated to the states on a per capita basis, the rest serves for equalisation according to fiscal capacity and expenditure needs. The guiding principle at the core of the Australian equalisation model is that, “if each [state] made the same effort to raise revenue from its own source and operated at the same level of efficiency, each would have the fiscal capacity to provide services at the same standards” (CGC 2008, 34). Although wealthier states like New South Wales and Victoria have periodically passed negative judgement on equalisation payments, the CGC has long been a respected institution (Else-Mitchell Reference Else-Mitchell1979) whose legitimacy serves to shield the Commonwealth government from criticisms over the evaluation of the fiscal capacity and, especially, the expenditure needs of the six states. Overall, in the fiscally centralised Australian federation (Nahan Reference Nahan2005), equalisation is viewed as being based upon a legitimate principle of redistribution, consistent with citizenship and the country’s egalitarian ethos.

Canada

Despite the fact that equalisation only emerged as a distinct federal programme in 1957, horizontal fiscal redistribution is as old as the 1867 British North America Act, which featured statutory subsidies that “contained an element of equalization in that they were per capita grants up to a maximum population” (Courchene Reference Courchene1984, 65). Yet, the idea of a formal equalisation programme emerged only in the late 1930s (Milne Reference Milne1998, 181). During the 1940s and the first half of the 1950s, the tax rental agreements between Ottawa and the provinces “took into account differences in fiscal capacity through per capita payments (implicit equalization) and provided minimum base grants to poorer provinces” (Marchildon Reference Marchildon2005, 422; on the tax rental agreements see Courchene Reference Courchene1984, 27–35). Yet, a series of developments soon triggered the establishment of a formal equalisation programme. Among these, the most crucial was probably the 1954 decision by Quebec’s Duplessis administration to re-establish the province’s income tax. This decision, alongside the publication of Quebec’s fiercely autonomist Tremblay Report, made it clear that “any trade off of equalization for federal occupation of the central tax sources (…) was impossible” (Milne Reference Milne1998, 190). Facing growing political pressures to expand territorial redistribution, fight regional inequalities and, especially, maintain fiscal and policy links with Quebec despite its exit from the tax rentals system (Bryden Reference Bryden2009, 81), the federal government created a national equalisation programme in 1957. There was a need for this type of territorial redistribution programme in many provinces, but it was the political question of Quebec that drove the creation of the programme. Over time, equalisation not only served the interests of Quebec but also those of the less well-off Atlantic provinces, as well as Manitoba and Saskatchewan.

This federal programme differed from the “Australian model” in at least three distinct ways (Lecours and Béland Reference Lecours and Béland2010). First, the federal government focused exclusively on fiscal capacity, as the assessment of expenditure needs appeared as a direct threat to provincial autonomy (Milne Reference Milne1998, 191). Second, the federal government, rather than an independent body such as Australia’s CGC, directly governed the equalisation programme (MacNevin Reference MacNevin2004, 188). Since 1957, the federal equalisation programme has become increasingly complex and few citizens understand how it actually works (Marchildon Reference Marchildon2005). Although the programme has periodically been a controversial political issue (Lecours and Béland Reference Lecours and Béland2010), equalisation as an institutional principle is embedded in the 1982 Canadian Constitution and is therefore there to stay. In Canada, equalisation represents about 27 per cent of federal–provincial transfers (Department of Finance Canada 2012), while in Australia, approximately four per cent of Commonwealth–state transfers are used for equalisation (CGC 2012, 25). Federal transfers are more important in Australia than in Canada, with around $100 billion transferred in Australia compared to roughly $57 billion in Canada. This is a reflection of the greater fiscal centralisation of Australia in comparison to the Canadian federation.

Revenue Sharing in the United States (1972–1986)

In the United States, massive growth in federal grants-in-aid to the states was witnessed in the post-war era and, especially, in the 1960s. Despite the election of Republican President Richard Nixon in 1968, “federal aid spending nearly doubled in nominal terms between 1968 and 1972, growing from $18.6 billion to $34.4 billion, and from 10.4 percent of all federal outlays to 14.9 per cent. It grew 22.4 percent between 1971 and 1972 alone” (Wallin Reference Wallin1998, 31). Most of this money took the form of categorical grants, which imposed strong policy constraints on receiving states. In such a context, states and local politicians began to pressure Washington, D.C. to channel money to them with few or no strings attached. From this perspective, in the late 1960s and early 1970s, the main political rationale for what became known as General Revenue Sharing (GRS) was not to reduce fiscal inequalities among constituent units, as was the case with equalisation programmes created earlier in Australia and Canada. Instead, it was to tackle vertical fiscal unbalance while increasing the policy autonomy of state and local governments by allocating to them a certain percentage of federal income tax revenues (Reagan and Sanzone Reference Reagan and Sanzone1981, 82). These remarks explain why President Nixon pushed for GRS during his first term, as he hoped such a system would increase state autonomy and, indirectly, erode the centralising policy logic associated with Lyndon Johnson’s Great Society. Despite the opposition of Democrats in Congress, such as the House Ways and Means Committee Chair Wilbur Mills (Arkansas), growing pressure over time from the local and state lobby, as well as the efforts of President Nixon to push the issue to the top of the federal agenda, helped weaken the opposition to GRS. It became law in 1972, after presidential-hopeful Mills embraced revenue sharing for electoral reasons (Conlan Reference Conlan1998, 67–70; see also Dommel Reference Dommel1974).

As implemented in 1973, GRS operated according to a complex formula that took “into account such variables as population, relative income, tax effort, and urbanized population. One third goes to state capitals and two thirds to local-general purpose governments” (Reagan and Sanzone Reference Reagan and Sanzone1981, 104). The emphasis on local government is a key difference between GRC and the equalisation programmes created in the 1930s and 1950s in Australia and Canada, respectively. More importantly, although GRC featured equalising provisions, horizontal fiscal redistribution was not its main goal, especially when it became federal policy during the Nixon years. Even in the mid-1960s, when revenue sharing first emerged as a mainstream policy issue, its main proponents only mentioned equalisation as one of the several policy goals of revenue sharing. For instance, economist Walter W. Heller (Reference Heller1968, 5), who did much to promote the idea of revenue sharing during the 1960s, stated that the reduction of “economic inequalities and fiscal disparities among the states” was only one of the six main goals of revenue sharing. In other words, revenue sharing was not described exclusively, or even primarily, as an equalisation scheme aimed at reducing horizontal fiscal imbalances among states. From this angle, the fight against state-level territorial fiscal inequalities was not the main rationale behind GRS. This was especially true during the Nixon years, when the enhancement of state and local fiscal autonomy and capacity proved the dominant public rationale for revenue sharing.

This public rationale made GRS (a temporary scheme Congress needed to reauthorise on a regular basis) increasingly vulnerable in a changing economic and political environment, as the fiscal situation of the federal government began to deteriorate in the post-1973 era. Interestingly, in 1980, President Carter justified the decision to terminate GRS payments to states by referring to both the deterioration of the federal fiscal situation and the comparatively improved financial posture of state governments (Wallin Reference Wallin1998, 103–104). Later in the 1980s, after Ronald Reagan entered the White House, GRS for local governments came under attack in a political context characterised by the intensification of fiscal austerity at the federal level and by attempts to curtain inter-governmental transfers through block grants and other policy means. In fact, during the Reagan years, “federal aid as a percentage of total state-local outlays (…) dropped from 25 per cent in 1981 to approximately 14 per cent for fiscal year 1987” (Williamson Reference Williamson1990, 184). Considering the growth in federal deficits, what remained of GRS increasingly appeared as “an unaffordable luxury” for a government that was nearly $2 trillion in debt (January 1986 Chicago Tribune editorial cited in Williamson Reference Williamson1990, 184). By mid-1986, the claim that the federal government could no longer afford revenue sharing had become common wisdom that helped justify the termination of GRS that year (Wallin Reference Wallin1998, 117).

National Unity

The creation of a federal state, whether it follows the coming together of previously independent units or the decentralisation of a unitary state (Stepan Reference Stepan2001), involves a compromise between independence and interdependence. Constituent units agree to be part of a larger whole on the condition that they have some institutional autonomy. Managing the self-rule and the shared-rule dimensions of federalism can constitute a difficult balancing act, sometimes to the point where one or more constituent unit governments seek secession. Theories of federalism and the management of diversity leave room for both integrative and disintegrative effects of federalism (McGarry and O’Leary Reference McGarry and O’Leary2007). On one hand, federalism can be viewed as providing political security and the free expression of collective identities through constitutionally-protected autonomous decision making, thereby strengthening the feeling of belonging to the state. On the other, this autonomous decision making can be seen as detaching communities from the state and therefore leading to a “slippery slope” to secession. Indeed, many federations experience a challenge to their unity at some point in their history. In some instances, these challenges have come and gone, as in Australia and the United States. In others, they are perennial; federal states such as Belgium and Canada, for instance, have to continuously nurture their unity.

Researchers have identified various strategies used by federal states to prevent secession (Coakley Reference Coakley2003). Most of these strategies centre on the self-rule aspect of federalism; they involve granting more extensive autonomy to the constituent unit(s) in which there is a secessionist movement through administrative arrangements or formal constitutional change and/or recognising, politically or constitutionally, their “distinctiveness”. Much less emphasis has been placed on how federal states can act upon their shared-rule dimension to prevent secession – that is, how they can seek to bolster the attachment and commitment to the central state and the nation it projects. From this perspective, scholars have mostly pointed out how constituent unit governments can be given representation in central institutions (e.g. in upper chambers where they can shape federal policies) or how federal government can create national programmes – everything from infrastructure to health care and social security (Lecours and McEwen Reference Lecours and McEwen2008).

While there is no doubt that mechanisms of institutional representation within central institutions and the development of national service programmes serve “national integration” purposes, little has been said about how federal states dealing with nationalist and secessionist movements use territorial fiscal redistribution to foster country-wide unity. Yet, there are good reasons to think that equalisation programmes have been used to weaken support for secession in federations where such support exists in constituent units whose wealth is below the country’s average. Equalisation can work towards unity in at least two ways. First, equalisation payments can make federalism seem like a good fiscal proposition. Hence, even those who may feel their constituent unit is politically marginalised within the federation, or that the federal government promotes and projects an identity that is not theirs, might accept the status quo rather than support independence. Differently put, equalisation raises the costs of secession in constituent units that are net receivers of this type of programme. Second, in federal states, equalisation may do more to strengthen national unity than “buy off” political support from poorer constituent units. It is reasonable to think that, in the longer term, equalisation programmes can work to create a community of redistribution that may generate feelings of country-wide togetherness and solidarity in recipient constituent units, thereby making support for secession less likely.Footnote ⁴ Indeed, from this perspective, equalisation gives substance to abstract notions of territorial solidarity through concrete territorial fiscal transfers; it can thus represent a nation-building device (Béland and Lecours, forthcoming; Théret Reference Théret1999).

The cases of Australia and Canada show that this type of thinking about equalisation contributed to the creation of these programmes. In Australia, the practice of financial assistance to poorer states that began immediately after 1901 was aimed at reducing discontent in Western Australia, Tasmania and, to a lesser extent, South Australia (CGC 1995, 4–12). Western Australia was a reluctant participant to the federation; not only was it at the territorial periphery of the new country and “felt itself particularly neglected by the government on the other side of the continent” (Macintyre Reference Macintyre2009, 182), but the transfer of customs and excise duties to a Commonwealth government would hurt its economy particularly badly. The framers of the Constitution took this into consideration by allowing Western Australia to raise the customs duty on inter-state exports for the first five years after common tariffs had been introduced but, by 1909, the Commonwealth was already providing special payments to the state and, in 1924, created a Royal Commission on the Finances of Western Australia as Affected by Federation (CGC 1995, 8). These measures were insufficient to satisfy Western Australia and, by the 1920s, dissatisfaction with the economic and fiscal consequences of the union for the state had produced a strong secessionist movement. At a broader level, in the early 1930s, the three poorest states (Tasmania, Western Australia and South Australia) pressured the Commonwealth government to create an independent commission to administer and formalise financial assistance to the states, even presenting a joint request to Prime Minister J. A. Lyons in 1933 (May Reference May1971, 30). In early 1933, a referendum campaign for secession in Western Australia was heating up, and Prime Minister Lyons went to the state to promise the creation of a special commission to deal with the allocation of special state grants. Although this promise was not enough to convince a majority of Western Australians to vote against secession, it shows how horizontal fiscal equalisation can be thought of and used by national politicians as a tool for fighting off secessionist movements. Moreover, the Australian case suggests there are good reasons for politicians to think that equalisation can work to stave off secessionist forces. While Great Britain’s reluctance to recognise the secession of Western Australia explains the continued territorial integrity of the Commonwealth of Australia in the short term, the disappearance over time of all secessionist, and even nationalist, feelings in the state can be attributed in large part to the systematisation of equalisation beginning in 1933 (Craven Reference Craven1986).

In Canada, a large part of the rationale for establishing an equalisation programme in 1957 was linked to national unity. The creation of the Canadian federation in 1867 was the product of a compromise between French-Canadian elites who sought substantial autonomy for their community and English–Canadian leaders focused on nation-building, spearheaded by a strong central government. From a political standpoint, this type of pan-Canadian nation-building proved challenging. In the first 90 years of the country’s history, several major issues (namely, the execution of Métis leader Louis Riel and conscription in both world wars) divided French– and English–Canadians, and Québec governments, especially when formed by the conservative Union nationale (1936–1939, 1944–1960), successfully fought against the implementation of national social programmes in the provinces. In the face of a strongly isolationist French–Canadian nationalism, creating a national programme of horizontal redistribution that would benefit the comparatively poorer province of Québec presented potential for generating solidarity between French– and English–Canadians. In fact, as stated above, it is when Québec opted out of the tax rental agreements first developed during World War II that Ottawa decided to create a federal equalisation programme, largely to keep the province included in federal–provincial horizontal redistribution beyond the one central to these agreements (Bryden Reference Bryden2009). While it is unclear to what extent equalisation worked to end the isolation of Québec, the programme became central to the “federalist” position against the Québec secessionist movement that developed in the 1960s. Since then, “selling” Canada in Québec has typically involved, among other things, making a claim about material gain (or, at least, what Quebec may lose if it would separate). In other words, it has meant developing the idea that it is profitable for the province to remain part of Canada. Thus, in that country, equalisation has largely served as a fiscal and territorial device to strengthen national unity, if only instrumentally. This is true partly because the equalisation programme can be used by Ottawa to attempt to shape Québec politics since the programme is entirely managed by the federal government. For example, in the run-up to the 2007 Québec elections where the incumbent government of the federalist Québec Liberal Party (QLP) was struggling against two more nationalist parties, including the secessionist Parti québécois, the federal government delivered “good news” on the equalisation front in a blatant attempt to help the QLP against its provincial opponents (Béland and Lecours, forthcoming).

Overall, then, in twentieth-century Australia and Canada, national unity was an important motivation for the development of a stand-alone equalisation programme. Throughout that century, no major concern existed over the territorial integrity of the United States, which meant that, as opposed to the situation prevailing in Australia and Canada, there was no national unity-related political incentive to create a stand-alone federal equalisation programme. There was, of course, a serious secessionist challenge in the mid-nineteenth century. Early American federalism was strongly rooted in a state independence tradition that generated a discourse about states’ rights, where the Union was conceptualised as a compact between separate entities that retained the exit option if their autonomy was threatened (Anderson Reference Anderson2004). Timing is crucial in politics (Pierson Reference Pierson2004), and mid-nineteenth-century solutions to secessionist threats did not generally involve the creation of national programmes of fiscal redistribution; these measures were much more coherent with the interventionist ideas of the 1930s and the post-World War II era. In any case, the creation of such a programme would not have solved the debate over slavery, which was so central in mid-nineteenth century America politics.

After the Civil War nixed the notion of a state’s right to secession (eventually declared illegitimate by the 1869 Texas v. White Supreme Court decision) and launched a process of nation-(re)building around a new and seemingly definitive understanding of federalism, direct threats to the territorial integrity of the United States disappeared. Racial politics, which became so central after World War II, did have a territorial component but, most of the times, claims were put forth without seriously challenging the territorial basis, or indeed the idea, of the American nation. As the United States moved into and through the twentieth century, territorial unity, perhaps with the partial exceptions of the remote states of Alaska and Hawaii, could virtually be taken for granted. Even though there were major wealth discrepancies between states, there was no existential danger in letting these be and, therefore, no unity-related need to develop an equalisation programme. Instead, during the Civil Rights movement and the War on Poverty, the federal government increased its role in American society in order to fight racial discrimination and fight poverty. Although this struggle had a territorial dimension, it was never about preserving the territorial integrity of the United States, a question that had been solved in 1865.

Territorial Redistribution and the Idea of Social Citizenship

Beyond the issue of territorial unity in the strict sense of the term, ideas about equality and social citizenship impact debates on horizontal fiscal disparities and redistribution in advanced industrial countries.Footnote ⁵ This claim is grounded in the ideational literature, which starts from the perspective that dominant ideas, principles and cultural categories can justify or, conversely, delegitimise specific proposals, a situation that is likely to shape the policy process (Béland and Hacker Reference Béland and Hacker2004; Campbell Reference Campbell2004; Parsons Reference Parsons2007; Steensland Reference Steensland2008; Padamsee Reference Padamsee2009). In the case of the politics of equalisation, we argue that, in the United States, a weaker sense of social citizenship and a reluctance to favour equality of outcomes over equal opportunity weaken the apparent legitimacy of the idea of horizontal fiscal redistribution. In contrast, in both Australia and Canada, the ideas of territorial equity and equal citizenship have long legitimised the existence of a highly redistributive federal equalisation programme. For instance, the 1982 Canadian Constitution implicitly links the federal equalisation programme created in 1957 to the possibility for citizens to access similar services, regardless of the province in which they live: “Parliament and the government of Canada are committed to the principle of making equalization payments to ensure that provincial governments have sufficient revenues to provide reasonably comparable levels of public services at reasonably comparable levels of taxation” (Constitutional Act 1982: Subsection 36[2]). It is hard not to link equalisation, as defined in the 1982 Constitution, with the universalist quest for social citizenship, which is integral to debates about the future of the welfare state and even of Canada as a country (Rice and Prince Reference Rice and Prince2000; Brodie Reference Brodie2002). For example, in Canada, universal access to health care has become the most prominent aspect of the relationship between the idea of social citizenship and national identity (Brodie 2002; Béland and Lecours Reference Béland and Lecours2008). This remark should not hide the fact that, as stated above, equalisation is another aspect of the Canadian national identity–public policy nexus. Thus, in Canada, the idea of equal citizenship, horizontal fiscal redistribution and the attempt to preserve national unity are directly related.

In Australia, where the idea of social citizenship is a major tenet of the welfare state (Shaver Reference Shaver2002), the Commonwealth equalisation programme is associated with an egalitarian creed even bolder than the one stated in the 1982 Canadian Constitution. While equalisation in Canada aims at enabling the provinces to deliver reasonably comparable levels of public services at reasonably comparable levels of taxation, the definition of equalisation formulated by the CGC for its 1999 report points to a greater level of egalitarianism: “State governments should receive funding from the Commonwealth such that, if each made the same effort to raise revenue from its own sources and operated at the same level of efficiency, each would have the capacity to provide services at the same standard” (CGC 2008, 27). This egalitarian logic points to the existence of a strong sense of shared economic and social citizenship in Australia, a reality that is hardly new in a country where the political culture is grounded in an egalitarian ethos. In Australia, territorial redistribution associated with the equalisation programme has long been tied to the need to foster equality between citizens living in the various regions of the Commonwealth. From this angle, a strong sense of social citizenship is behind the existence of comprehensive horizontal fiscal redistribution and, more specifically, the federal equalisation created in 1933 to explicitly fight regional disparities and help poorer states deliver quality public services.

In the United States, in contrast to the situation prevailing in Canada and Australia, there is no strong sense of social citizenship (Fraser and Gordon Reference Fraser and Gordon1992; Myles Reference Myles1997). The weakness of the idea of social citizenship in the United States stems directly from the strict opposition between “charity” (redistribution through social assistance) and “contract” (social insurance) that has characterised American ideas about social policy since the adoption of the Social Security Act in 1935 (Fraser and Gordon Reference Fraser and Gordon1992). From this perspective, the only strongly legitimate entitlements in the United States are the “earned rights”, derived from social insurance contributions. To this we can add that the ideational and cultural dichotomy between the deserving and undeserving poor prevalent within the field of welfare reform (Steensland Reference Steensland2008) further weakens the sense that basic, universal social rights are an integral part of the social contract and, by extension, citizenship, as is clearly the case in Australia and Canada. At a deeper level, in the United States, this reality reinforces the traditional resistance towards explicit forms of redistribution, especially those promoting the equality of outcomes among individuals or states. This is true because, in the United States, the idea of citizenship is directly tied to equal opportunity rather than to equality of outcomes, which typically legitimises redistribution. Importantly, this focus on equal opportunity clashes with the existence of enduring forms of racial inequality in the United States (Wallace Reference Wallace2003). The impact of racial assumptions and hierarchies on American social policy has been well documented, and it is clear that race has done much to weaken the sense of social solidarity and citizenship in the United States. In the past, for instance, race pushed white political elites in Southern states to reject equal access to social services and benefits within their jurisdictions (Quadagno Reference Quadagno1988; Quadagno Reference Quadagno1994; Lieberman Reference Lieberman1998). Today in the United States, race remains a key factor in the politics of public policy (Gilens Reference Gilens2000; Soss et al. Reference Soss, Fording and Schram2011). As a societal factor, race weakens the idea and sense of social citizenship. As suggested above, this weak sense of social citizenship makes it harder to justify the very idea of a federal equalisation programme in the United States.

As we claim, taken together, strong national unity and, at the ideational level, a weak sense of social citizenship are two societal factors that help explain why there has not been a strong push to create a federal equalisation programme in the United States. While in countries such as Australia and Canada the quest for national unity and social citizenship legitimised the idea of equalisation, this has not been the case in the United States, because that national unity has not been a territorial issue since 1865, and the idea of social citizenship is too weak to explicitly justify the quest for systematic territorial redistribution between states. Yet, as we argue, even if the presence of at least one of these factors helped trigger a major debate on the need to enact a comprehensive federal equalisation programme, institutional forces would have greatly complicated the push for equalisation in the United States.

Political Institutions

The sources of change in federal systems are many. Federal systems are often created to manage societies that are linguistically or ethnically heterogeneous. As such, federal institutions can be especially sensitive to societal transformations involving shifts in the composition of groups and in their relations (Erk Reference Erk2007). Belgian federalism, for instance, has been shaped by Flemish pressures for greater decentralisation since federal structures were created in the 1970s (Deschouwer Reference Deschouwer2009). Thus, federal institutions often respond to societal pressures. Courts have also been central in sending federal systems in specific developmental pathways through their interpretation of constitutional texts. Australia’s High Court, for example, has rendered many important decisions that contributed to the centralisation of federalism in that country (Fenna Reference Fenna2007). Change can also be triggered by outside events. The relative power of federal and constituent unit governments can be seriously affected by powerful “external shocks” (Pierson Reference Pierson2000), such as wars and economic crises.

Thinking about the sources of change is only somewhat helpful for understanding a lack of change. Of course, stability in society, court judgements that preserve the status quo and the absence of external shocks can reduce the likelihood and scope of change. Yet, even in the face of a changing domestic and international environment, many aspects of a federal system can endure. For example, the rigidity of some constitutional frameworks can be a direct cause for the lack of path-departing change in federal structures; formulas requiring unanimous consent of the constituent unit governments, in addition to federal consent, make it difficult to change a constitution. However, not all change in federal systems requires constitutional change. While formal changes to the division of power or to representative institutions typically do, the creation of new federal programmes, such as equalisation schemes, most often do not. Thus, in itself, constitutional rigidity cannot explain why there is no equalisation programme in the United States.

It is likely that the extent to which constituent units have a say in federal policy making through their representation in an upper house, and even their demographic weight in a lower house, can impact the likelihood that a national equalisation programme be created. Indeed, this type of programme typically makes payments only to some constituent units (poorer constituent units receive money while wealthier ones do not), and these payments are differentiated (some recipient constituent units get more money than others). As such, the capacity for constituent unit governments to effectively oppose territorial redistribution can loom large in the creation of an equalisation programme.

Canada’s federal model, in which a network of inter-governmental relations is centred on federal and constituent unit executives in lieu of effective constituent unit representation in an upper house (Smiley Reference Smiley1987; Théret Reference Théret1999), facilitated the adoption of an equalisation programme in 1957. The inter-governmental forums where discussions on equalisation took place favoured the building of a consensus on the new programme, as provincial governments had no formal institutional means to block the creation of a purely federal programme. The 1950s were the heyday of federal power in Canada, as Keynesianism and federal nation-building legitimised centralising reforms, while Ottawa used federal–provincial meetings to rally the provinces to its national initiatives. The “veto power” (Immergut Reference Immergut1992) of provincial governments regarding federal initiatives was limited by the fact that the country’s upper chamber did not provide the provinces with any kind of effective representation, let alone a potential veto on the creation of new federal programmes. Although the founders of the Canadian federation spent much time debating the nature of the Senate in the 1860s, with Lower Canada (which later became Québec) as well as New Brunswick, Nova Scotia and Prince Edward Island pressing for some form of territorial representation (Mackay Reference Mackay1963, 38–41), the federal Senate never became a house for provincial representation. This is in large part because Canadian Senators were never elected or appointed by provincial governments but were simply nominated by the federal government, which typically looked for party loyalists rather than strong provincial voices. There is a territorial dimension to the Senate insofar as Québec and Ontario, as well as the Western and the Atlantic provinces (collectively), are each assigned 24 Senators. However, in the absence of democratic legitimacy, their ties to the provinces remain weak. As the Senate was never reformed to bolster this legitimacy (Smith Reference Smith2009), provincial governments had no institutional mechanism for stopping the creation of an equalisation scheme in the 1950s, even if many of them had opposed such a programme. As for the House of Commons, strong party discipline meant that Members of Parliament representing potentially non-recipient provinces really had no opportunity to voice their opposition.

The structure of Australian federalism is similarly unfavourable to constituent units blocking the creation of a national scheme of territorial redistribution. In structure, contemporary Australian federalism features both multilateral networks of inter-governmental relations and a Senate with substantial legislative powers where equal number of members are elected in each state (Hueglin and Fenna Reference Hueglin and Fenna2006, 222). Yet, the nature of the Australian Senate seriously limits the capacity of states to play a role in federal law making. The Senate was one of the most contentious issues in the negotiations surrounding the federation of the colonies. In the 1891 draft constitution, “the Senate was to be a States’ house comprising senators appointed by the States” (Galligan Reference Galligan1995, 79), but, in the final version of the constitution, senators were directly elected by the population. Although some observers of the time argued that the Australian Senate could still be seen as a states’ house, many, such as delegate Macrossan from Queensland, foresaw exactly how it would function. “The influence of party will remain much the same as it is now”, he said, “and instead of members of the senate voting […] as states, they will vote as members of parties to which they belong” (Federation Debates 1891, 434, cited in Galligan Reference Galligan1995, 82). In fact, the Australian Senate never functioned as a states’ house, as party politics and discipline always trumped the territorial dimension. Although it is often referred to as one of the most powerful federal upper chambers in the world as a result of its considerable constitutionally specified legislative powers, its connection to federalism is weak (Swenden Reference Swenden2004, 74). The notion of states’ rights, so strong in the United States before and even after the Civil War, never had much currency in Australia (Galligan Reference Galligan1995, 80–85); there was, therefore, no institutional expression of such a doctrine. Consequently, even if wealthier states such as New South Wales and Victoria may have had misgivings about formalising equalisation arrangements through the establishment of the CGC in 1933, the institutional arrangements of federalism gave them very little hope that they could effectively work against this major reform. The Australian House of Representatives, much like the Canadian House of Commons, featured disciplined parties that left little room for individual members to stray from the party line.

In the United States, the weakness of party discipline, combined with a federal structure in which constituent units are represented in an upper house through senators who have a direct say in federal policy making (Smiley Reference Smiley1987), meant that the development of explicit, large-scale territorial redistribution would face a major hurdle in the Senate. Bicameralism is strong in the United States (Baker Reference Baker1988); a bill does not become law unless both the House of Representatives and the Senate supports it through a majority vote. Moreover, senators are directly elected, a procedure (constitutionalised through the Seventeenth Amendment of 1913) that provides them with robust democratic legitimacy, and they are largely free of party discipline. Indeed, there is a “strong senatorial tradition toward individualism” (Campbell and Rae Reference Campbell and Rae2001, xi) that leaves senators with great autonomy to defend and promote their state’s interests (as they see them) within federal institutions, as there are virtually no tight partisan constraints upon their decision making.Footnote ⁶ Therefore, in general terms, while the Canadian Senate would not have been able to prevent the creation of equalisation even if there had been substantial provincial opposition to it because of its “democratic deficit”, and the Australian Senate would have been a similarly ineffective weapon in the hands of opposing states resulting from the grip of party politics on its members, the United States Senate would represent a major stumbling block for political forces looking to create a federal equalisation programme.

But why would the United States Senate not be receptive to a stand-alone equalisation programme? There are several reasons.Footnote ⁷ First, in the absence of party discipline, it would be very difficult for senators of states unlikely to receive anything from an equalisation programme to support the creation of such a scheme. Stark (Reference Stark2009) has estimated that, if a Canadian-style equalisation system were implemented in the United States, 21 states would not receive any payment. Moreover, we do not know if senators from borderline recipient states in Stark’s simulation (Oregon, South Dakota, Montana, Maine, Wisconsin and Iowa) would feel it would be in the interest of their state to support an equalisation programme. They might, on the basis of uncertainty, oppose equalisation for purely material reasons (fearing their state might not receive anything). In addition, smaller states, which disproportionately benefit from Senate apportionment (Lee and Oppenheimer Reference Lee and Oppenheimer1999) and have more political clout on a per capita basis than larger states, might resist the establishment of a stand-alone equalisation scheme that would allocate resources, at least partly, on a per capita basis. Still, even if we assume that 58 senators found that their state would gain financially from equalisation (a number technically large enough to pass legislation, although not large enough to surmount a filibuster), there are good reasons to think that a number of them would still oppose the creation of such a programme. It is also reasonable to think that a large number of representatives would similarly oppose a stand-alone equalisation programme.

In fact, from an institutional and political standpoint, equalisation is antithetical to the way members of Congress go about their business, especially pork-barrelling (Ferejohn Reference Ferejohn1974; Stein and Bickers Reference Stein and Bickers1997; Frisch Reference Frisch1998; Evans Reference Evans2004; Bertelli and Grose Reference Bertelli and Grose2009; Berry et al. Reference Berry, Burden and Howell2010). It is clear that pork-barrelling would be much more difficult around one big grant (for instance, a stand-alone equalisation grant) than it is around multiple smaller grants. Perhaps even more crucial, formula-based equalisation policy goes against the logic of “credit claiming” (Mayhew Reference Mayhew1974; Weaver Reference Weaver1986) associated with pork-barrel politics. Indeed, members of Congress typically like arrangements that allow them to “earmark” monies for specific projects because they benefit politically from them, while they would not directly benefit from formula-based general-fund financing (Frisch Reference Frisch1998, 18; Nordyke Reference Nordyke2008, 21). With earmarks, members of Congress can vote for a specific outlay, whereas with general-fund financing, they can only support a bundle of expenditures (Buchanan 1999 [Reference Buchanan1967], Ch. 6). Earmarks are particularly useful for members of Congress wishing to reward important contributors to their campaigns (Rocca and Gordon Reference Rocca and Gordon2013). Recent research also suggests that members of Congress who are active “earmarkers” have less primary election competition than those who are not (Lazarus et al. Reference Lazarus, Glas and Barbieri2012). Moreover, the constellation of lobbyists and interest groups that operate within the current federal political system is also a status quo force, because advocating (and gaining particular consideration) for specific projects is easier when dealing with politically sensitive members of Congress in need of campaign financing (Rocca and Gordon Reference Rocca and Gordon2013). For members of Congress, keeping politics at the centre of federal funding is attractive because they can potentially become winners within this system. For example, recent research shows that “members of Congress who belong to the president’s party are advantaged in the budgetary process” (Berry et al. Reference Berry, Burden and Howell2010, 783). In sum, because of its formula-based logic, a stand-alone equalisation programme would clearly reduce the credit-claiming ability of members of Congress stemming from pork-barrel politics, which helps to account for why they seem to have little appetite for this type of formula-based programme.

Finally, from an ideological perspective, a striking feature of Stark’s (Reference Stark2009) modelling is that most of the states that would be beneficiaries of equalisation are the so-called “red states”, in which politics is against government spending and redistribution. From this angle, creating a programme that could cost $68 billion (Stark’s estimate for a Canadian-style equalisation programme in the United States) might be opposed by many “red state” senators as well as by representatives from wealthier and more populated states such as California, New York and Illinois.