Introduction

A rapidly expanding economy, along with strong population growth and the expansion of the state throughout the 20th century, strengthened ethno- linguistic and regional distinctiveness. It also led to increasingly sophisticated interaction among the provinces and between them and the federal govern­ment (Baldwin, Brown and Maynard 2005; Lazar, Telford and Watts 2003; Smiley 1974). With the constitution beyond reach, managing change required a nearly constant series of intergovernmental meetings and agreements that effectively turned the constitution on its head (Scott 1977).^[31] Executive federal­ism, the meeting of first ministers from the provinces, territories, and federal government, became the pre-eminent mechanism for bootstrapping major institutional and constitutional change across the 20th century (Bakvis, Baier and Brown 2009, p.

The absence of a domestically controlled constitution presented Canadians with a double bind. They were prevented from directly changing their consti­tution to respond to local circumstances and lacked a legitimate and effective mechanism for resolving this situation. One would have to be created. To be acceptable and legitimate in a self-governing country, any institutional inno­vation that enabled a subsequent process of constitutional change would need to be identifiably Canadian and hold out the hope of finding widespread— perhaps unanimous—support across diverse provinces. This would help con­vince voters in Canada and politicians in Britain to support the use of the for­mal British amendment process to ‘patriate' the constitution with additions including a domestic amending formula.^[32]

Despite the arrival of executive federalism, it took 60 years and 14 separate federal-provincial meetings from 1927 to solve this central dilemma. As evi­dence of the long road taken, one of the amending processes in the 1982 consti­tution was first suggested in the report of a 1936 federal-provincial committee on constitutional change (Hurley 1996). Each failure on the road to success cre­ated both varied and often increased friction, with the final version imposed against the will of the Quebec government that to this day has not signed the patriated 1982 constitution. This hiccup in the self-generated process for creat­ing an amending formula has had little practical effect on the legitimacy of the new constitution. Given the continued difficulty of changing the constitution (it was gifted a series of complex amending formulas when it was patriated in 1982), the intensity of demands from Quebec, and the great diversity of prefer­ences among the provinces, executive federalism remains the central mecha­nism for shifting of power around the federation.^[33]

The centrality of executive federalism rests on two key features of Canadian politics that provide it with sufficient legitimacy and power to be effective.

Riker's emphasis on the centrality of bargaining and party systems in federal states is a useful starting point for an initial exploration of the possible inter­action of executive federalism and partisan dynamics in Canada. The goal is to explore his claim that the structure of party systems captures the character of bargaining—in essence, the distribution of power—among the constituent units of a federation. I hypothesize that the expansion of executive federalism as a means for reconstituting the character of Canadian federalism should be accompanied by evidence of changing electoral and legislative party-system dynamics as leaders strive to maximize their bargaining position (Riker 1964).

Riker's approach to the distribution of bargaining power in federations emphasizes, first, the importance of congruence in the ideology and organi­zation of the parties from which the bargaining governments are formed. That is, ‘The federal relationship is centralized according to the degree to which the parties organized to operate the central government control the parties orga­nized to operate the constituent governments' (Riker 1964, p. 129).

Second, the internal discipline of a party as a measure of the strength and reliability of its bargaining position in office. The mainly unicameral Canadian Westminster-style parliaments, in combination with the electoral system (here, single-member plurality), deliver nearly the entirety of government power to highly disciplined single-party governments.

In the Canadian case, the aim is to seek evidence that first ministers work to maximize their bargaining independence—the terrain over which they feel free to make agreements—and their bargaining authority—their capacity to deliver on their promises by guaranteeing the passage of enabling legislation. Independence is measured in terms of both intra-party organizational link­ages that might constrain first ministers, and provincial or regional support for the federal government at elections as a proxy for relative support of the local versus national protagonists. Authority is measured in terms of executive dominance of the legislature, operationalized as the numerical strength of the governing party, itself partly a function of the translation of votes into seats by the electoral system. Operationalization of these indicators is achieved using electoral data extracted from the Canadian Elections Database that contains all federal, provincial, and territorial elections since 1867 to the district level (Sayers 2017).

A first consideration, then, is limitations on a first minister's power that might flow from organizational linkages between parties in different juris­dictions, particularly across the federal-provincial divide. Maintaining such links might limit the policy discretion of a first minister if he or she is forced to balance competing views on a major issue in order to sustain organiza­tional coherence. This is then followed by consideration of whether a first minister's role as the sole representative of local interests is expressed in regional voting patterns. Weakness in regional electoral support for the fed­eral government would be consistent with voters viewing local first ministers as defenders of their interests against an ‘opposing' federal government. The final element explores the degree to which voters authorize first ministers to make binding agreements by giving them dominant seat shares of provincial legislatures.

This initial exploration suggests there may be value in pursuing this line of research.

Evidence of legislative correlates with respect to regional first minister authority (seat dominance) is less clear. Government dominance appears to track the dynamics of intergovernmental relations in some provinces but not in others. This may reflect a need to reconsider how best to both conceive of and measure authority. Future work should refine these approaches in order to better understand how electoral choice and government longevity interact with the authority of first ministers in intergovernmental bargaining.

2