The Space between Worlds: Federalism, Public Issues and Election Issues

Abstract

This paper reconciles two strands of literature on federalism and mass political behaviour behaviour, but the reconciliation is not a happy one. One strand of literature pointed out optimistically that Canadians keep their two political worlds—provincial and federal—more separate than citizens in any other federation. Another strand points to the obvious difficulty citizens would have in separating the contribution of the two levels of government to policy results on the ground. Are these arguments fundamentally at odds? These two strands of literature can be reconciled by showing how federalism, particularly the Canadian version, forces parties, voters, and the media to make elections about issues that feature less intergovernmentalism. Unfortunately, these issues are usually of less concern to citizens than the highly intergovernmental policy areas systematically ignored in both federal and provincial elections. In short: Canadian elections are about the wrong things and federalism is to blame.

Keywords:

• Federalism
• elections

Acknowledgements

The author would like to thank the anonymous reviewers, Richard Simeon, Stuart Soroka and Ailsa Henderson for comments on previous drafts. Top-notch research assistance came from Hilary Pearse and Katie Boothe. The Social Sciences and Humanities Reseach Council of Canada supported the research with a grant under the Federalism and Federations program. Any errors of fact or interpretation are the author's alone.

Notes

Author's own unpublished analysis of the Saskatchewan Election Study, 2003.

It is tempting to say that Quebec is the exception that proves the rule.

Of course, there are provincial elections where the government's approach to federal–provincial relations is a major issue. But this is a properly provincial issue; it is not the case that the provincial election is influenced by attitudes to the federal government independently.

BC Referendum Studies, 2005 and 2009.

I emphasize ‘positive’ here in contrast to an empirical theory of voting in multi-level governance, such as the “second-order elections” thesis of Reif and Schmitt (1980).

Much of the theory that follows, however, is equally applicable to an ex ante choice between competing alternatives.

Looking back to the 1993 election, unemployment was dominant, with 42% of valid open-ended responses referring to a need for more jobs. Taxes and general economic concerns ran a distant second and third, with health care mentioned by only 5% of respondents. By 1997 there was some movement toward debt/deficit concerns, national unity and health care, but unemployment was still out front with 33% mentioning it as most important (CES).

The sponsorship scandal is the obvious exception.

When 453 British Columbians and Albertans were asked in 2001 about the state of health care and were then asked “Which government do you think is mostly responsible for this situation? Is it the provincial government, the federal government, or both the federal and provincial governments equally?”, their responses were: 23% provincial, 9% federal, 51% both governments, and 14% other causes or don't know.

Obviously, the federal influence on the character and quality of social programmes is profound; but it is indirect and involves notoriously hard-to-follow money.

In fact, when asked in September 2004 to choose “Which level of government, federal or provincial, would you say is most responsible for the deterioration [in the quality of health care]?”, the federal government was the choice of 46%, compared with 37% saying the provinces. Until 2002, the provinces were the plurality choice (the Strategic Counsel, cited in Soroka, 2007).