1 Clarke, Harold D. and Kornberg, Allan, “Support for the Canadian Progressive Conservative Party since 1988: The Impact of Economic Evaluations and Economic Issues,” this JOURNAL 25 (1992), 53.
2 McMenemy, John, “Parliamentary Parties,” in Winn, C. and McMenemy, J., eds.. Political Parties in Canada Toronto:McGraw-Hill Ryerson, 1976), 14.
3 Trudeau, P. E., Federalism and the French Canadians Toronto:Macmillan, 1968), 119.
4 Cairns, Alan C., “The Electoral System and the Party System in Canada, 1921–1965,” this JOURNAL 1 (1968), 67.
5 Lijphart, Arend, Democracy in Plural Societies: A Comparative Exploration New Haven:Yale University Press, 1977), 125;Lijphart, Arend, Electoral Systems and Party Systems: A Study of Twenty-seven Democracies Oxford:Oxford University Press, 1994), 140; and Lijphart, Arend, Rogowski, Ronald and Weaver, R. Kent, “Separation of Powers and Cleavage Management,” in Weaver, R. Kent and Rockman, Bert, eds., Do Institutions Matter? Government Capabilities in the United States and Abroad Washington:Brookings Institute, 1993), 302–344.
6 Cairns, “The Electoral System and the Party System,” 73.
7 Laponce, Jean, The Protection of Minorities Berkeley:University of California Press, 1960), 139.
8 Clarke and Kornberg, “Support for the Canadian Progressive Conservative Party,” 53.
9 Nadeau, Richard and Blais, André, “Explaining Election Outcomes in Canada: Economy and Politics,” this JOURNAL 26 (1993), 787; and Cohen, Ronald I., Quebec Votes Montreal:Saje, 1965), 18.
10 Johnston, Richard, Blais, André, Brady, Henry E. and Crfite, Jean, Letting the People Decide: Dynamics of a Canadian Election Montreal:McGill-Queen's University Press, 1992), 30.
12 On geographic concentration see Johnston, R. and Ballantyne, J., “Geography and the Electoral System,” this JOURNAL 10 (1977), 857–866; on the number of parties and candidates see Spafford, Duff, “The Electoral System of Canada,” American Political Science Review 64 (1970), 168–176.
13 Robert Jackman, in decomposing the variance in vote swings into national, provincial and local (riding) sources for the period 1925–1965, found that regional (that is, provincial) effects were much more pronounced in Canada than in either the US or Britain (see “Political Parties, Voting, and National Integration: The Canadian Case,” Comparative Politics 4 , 511–36). J. Ferejohn and B. Gaines obtained similar results for the period 1908–1984 (“The Personal Vote in Canada,” in H. Bakvis, ed. Representation, Integration and Political Partiesin Canada, Research Studies of the Royal Commission on Electoral Reform and Party Financing, Vol. 14 [Toronto: Dundurn, 1991], 283–85). On the consider able degree to which electoral alignments in Canada are patterned on a provincial basis, see Donald E. Blake, “1896 and All That: Critical Elections in Canada,” this JOURNAL 12 (1979), 259–79.
14 The 80 per cent threshold is the same as used by Reid, Escott M. in his scheme for classifying ridings as francophone (“Canadian Political Parties: A Study of the Economic and Racial Bases of Conservatism and Liberalism in 1930,” Contributions to Canadian Economics 6 , 7–39). The election and census data on Quebec constituencies for 1908 through 1984 are based on machine readable data files provided by Donald E. Blake of the University of British Columbia. The census category “Ethnic origin” was used to define francophone for 1908 through 1965; “French home language” from 1968 onward. Census data on language as reported in Eagles, D. Munroe, Bickerton, James and Gagnon, Alain, The Almanac of Canadian Politics Peterborough:Broadview, 1991) were used to classify francophone Quebec constituencies for 1988 and 1993. On other provinces three data sources were used: Beck, J. Murray, The Pendulum of Power: Canada's Federal Elections Scarborough:Prentice-Hall, 1968), for the period 1878–1968; Feigert, Frank B., Canada Votes, 1935–1988 Durham:Duke University Press, 1989), for 1972–1988; and Elections Canada, preliminary results for the 1993 federal election as of March 1994. Valid votes cast rather than registered voters are used for calculating electoral cohesion since data for all provinces on registered voters is incomplete. This does mean, however, that block voting in Quebec may be exaggerated slightly. If one uses registered voters, the difference in block voting between Ontario and Quebec tends to shrink for certain elections for which data are available, especially 1974 and 1980. See Drouilly, Pierre, Le paradox Canadien: Le Quebec et les elections federates Montreal:Parti Pris, 1978), 163–168, 224–25.
15 This period was characterized by “ministerial” politics and the absence of a meaningful two-party system, caused in part through deferred elections and rapid population growth (see Blake, “1896 and All That,” 268). As well, there is confusion as to the partisan affiliation of candidates and considerable discrepancies in reported election results. We have relied on those reported in Beck, The Pendulum of Power.
16 In many ways the first decade in Newfoundland resembles British Columbia in the nineteenth century as a case of what Blake (”1896 and All That”) calls “emergent alignment.”
17 To the extent that tests of significance have any meaning with data of this sort, the LSD procedure for multiple pairs applied in conjunction with a one-way analysis of variance (equivalent to doing multiple t-tests) yields a significant difference between Ontario and Quebec of less than .01; the overall F ratio (6.73) was significant at .00001.
18 Reid, “Canadian Political Parties,” 20, 23.
19 Laponce, Jean, Languages and Their Territories Toronto:University of Toronto Press, 1987); and Dion, Stdphane, “Explaining Quebec Nationalism,” in Weaver, R. Kent, ed., The Collapse of Canada? Washington:Brookings Institute, 1992), 77–121.
20 The results here on the francophone leadership factor appear to be broadly consistent with those of Nadeau and Blais, who report that “a major party that selects a leader from Quebec gets a five to six point boost in its share of the vote” (“Explaining Election Outcomes in Canada,” 782). However, keep in mind their analysis applies to the country as a whole, that is, using national level outcomes, and their data are not broken down by region or province, or by language group. Also our dependent variable is the largest party in the province rather than the incumbent party nationally.
21 A number of possibilities were attempted, including the use of different time periods as “dummy” variables. The variable CONTEXT was put in quadratic form in order to capture the possibility identified by Reid (“Canadian Political Parties”) that francophones in “mixed” ridings might feel more threatened and hence be somewhat more likely to vote for the largest party. CONTEXT, and the model as a whole, did provide marginally more explanatory power when used in place of, or in addition to, francophone proportion alone.
22 Cairns, “The Electoral System and the Party System,” 73.
23 Johnston et al., Letting the People Decide, 230; see also 199–200 and 293, n. 7.
24 Sniderman, Paul M., Forbes, H. D. and Melzer, Ian, “Party Loyalty and Electoral Volatility: A Study of the Canadian Party System,” this JOURNAL 7 (1974), 282–283; and Johnston, et al., Letting the People Decide, 84–85, 90.
25 Blake, Donald E., “Party Competition and Electoral Volatility: Canada in Comparative Perspective,” in Bakvis, , ed., Representation, Integration and Political Parties in Canada, 256, 264.
26 Cairns, “The Electoral System and the Party System in Canada,” 73.
28 Cairns looks only at those elections in which the province gave the majority of seats to the party which formed the government (ibid.). This leads to the exclusion of 3 of the 14 elections in the 1921–1965 period he considers; and to the exclusion of five elections when his reversal technique is applied to Ontario.
29 Certain assumptions are made, namely, that in minority situations the party with the largest vote share forms the government and that the duration of governments remains the same. See Macpherson, Laura G., “Quebec Block Voting” (honours essay, Dalhousie University, 1991).
30 While the underlying relationship is nonlinear, in practice the range of data points is limited so that a straight line captures the relationship reasonably well (see Spafford, “The Electoral System of Canada,” 173; and Tufte, Edward R., “The Relationship between Seats and Votes in Two-Party Systems,” American Political Science Review 67 , 543). We did rewrite the equations in both quadratic and cubic formats, but these efforts had only a marginal impact on the R2; in fact, in some of the scattergrams the R2 actually declined.
31 Tufte, “The Relationship between Seats and Votes,” 542–43.
32 As Tufte has noted, the cube law is a misnomer: in practice the parabola depicted by this law tends to follow a seat-vote ratio of 2.5:1 rather than 3:1 (ibid., 544–45). For applications of the cube law in modified form to Canada see Quaker, Terrence H., “Seats and Votes: An Application of the Cube Law to the Canadian Electoral System,” this JOURNAL 1 (1968), 336–344; and Barry Kay, “Improving upon the Cube Law: A Regional Swing Model,” paper presented at the annual meeting of the Canadian Political Science Association, Victoria, British Columbia, 1990.
33 There are other ways: increasing the number of parties and candidates in each constituency and ensuring that the winning vote pluralities for the largest party are as close as possible will also do the trick, which is what likely occurred in the 1993 federal election in Ontario. Outcomes are much less predictable, however, so it may not be the ideal strategy. See Spafford, “The Electoral System of Canada,” for an analysis of the logic underpinning this phenomenon.
34 Lovink, J. A. A., “On Analyzing the Impact of the Electoral System on the Party System in Canada,” this JOURNAL 3 (1970), 497–521. His patronage factor is probably better described as a porkbarrel factor, insofar as benefits from regional policies and the like devolve on groups rather than individuals. On this distinction see Noel, S. J. R., “Dividing the Spoils: The Old and the New Rules of Patronage in Canadian Politics,” Journal of Canadian Studies 22 (1987), 72–75.
35 Blake, Donald E., “LIP and Partisanship: An Analysis of the Local Initiatives Program,” Canadian Public Policy 2 (1976), 17–32; and Macnaughton, Bruce D. and Winn, Conrad J., “Economic Policy and Electoral Self Interest: The Allocations of the Department of Regional Economic Expansion,” Canadian Public Policy 7 (1981), 318–327. See also Savoie, Donald J., The Politics of Public Spending in Canada Toronto:University of Toronto Press, 1990).
36 Johnston and Ballantyne, “Geography and the Electoral System.”
37 Irvine, William, Does Canada Need a New Electoral System? Kingston:Institute of Intergovernmental Relations, 1979), 15–17.
38 On Belgium and the Netherlands see, respectively, Covell, M., “Stability and Change in the Belgian Party System,” and Wolinetz, S., “The Netherlands: Continuity and Change in a Fragmented Party System,” both in Wolinetz, S., ed., Parties and Party Systems in Liberal Democracies London:Routledge, 1988).
39 Lijphart et al., “Separation of Powers and Cleavage Management.”
40 See the contributions to McRae, K. D., ed., Consociational Democracy: Political Accommodation in Segmented Societies Toronto:McClelland and Stewart, 1974).
41 Rogowski, Ronald, Rational Legitimacy: A Theory of Political Support Princeton:Princeton University Press, 1974).
42 Thomas, Paul G., “Parties and Regional Representation,” in Bakvis, , ed., Representation, Integration and Political Parties in Canada, 203.
43 Cabinet representation does not directly follow representation in the government caucus. In fact, often the underrepresentation of a province or region in the caucus will be compensated for by increased representation in cabinet. In the case of Quebec, until the mid-1960s it was overrepresented in cabinet relative to the country as a whole, but it did lag behind Ontario. See Schwartz, Mildred A., Politics and Territory: The Sociology of Regional Persistence in Canada Montreal:McGill-Queen's University Press, 1974), 67–82. From 1968 through 1988, however, its overrepresentation was consistently greater than that for Ontario (Roselle Sheryl Hryciuk, “The Federal Cabinet as an Intrastate Federal Institution” [MA thesis, Dalhousie University, 1993], 20).
44 Trudeau, Federalism and the French Canadians, 119.
45 The capacity of the Quebec caucus to act in a much more disciplined fashion compared to its Ontario counterpart in both the Trudeau and Mulroney cabinets has been well documented. See Simpson, Jeffrey, “The Two Trudeaus: Federal Patronage in Quebec, 1968–84,” Journal of Canadian Studies 22 (1987), 96–110; and Thomas, “Parties and Regional Representation,” 179–273.
46 This point underscores the manner in which the electoral system can interact with features of central institutions. This is true not only in Canada. As one of the anonymous referees for this JOURNAL pointed out, in the case of the American South, “until recently, the crucial variable was not the level of support for the Democrats but the consistency of that support which ensured long-service careers for southern members of Congress which in turn meant that they held a disproportionate number of key committee chairs.”
47 Thomas, “Parties and Regional Representation,” 243.
49 The distribution formula became subject to the unanimity amendment provision (s. 41) of the 1982 Constitution Act.
50 For a direct comparison between the Canadian Liberal party and the Belgian Christian Social party on these points see David Rayside, M., “The Impact of the Linguistic Cleavage on the ‘Governing’ Parties of Belgium and Canada,” this Journal 11 (1978), 61–97.
51 On the “intrastate” federalism dimension, that is, the representation of territorial interests within central institutions, see Smiley, D. V. and Watts, R. L., Intrastate Federalism in Canada, Research Studies of the Royal Commission on the Economic Union and Development Prospects for Canada, Vol. 39 Toronto:University of Toronto Press, 1985), esp. chap. 5.
52 Johnston and Ballantyne, “Geography and the Electoral System,” 861.
53 For discussion of the algorithm and technique see McLain, H. D., “Drawing Contours from Arbitrary Data Points,” The Computer Journal 17 (1974), 318–324. We used the DWLS smoothing routine found in the graphics module of the Sys-tat (Version 5.01) statistical package (see Leland Wilkinson, SYGRAPH: The System for Graphics [Evanston, 111.: SYSTAT, 1990], 236). The vote-seat ratio is based on percentage votes for the largest party in each election.
54 Macnaughton, and Winn, , “Economic Policy and Electoral Self Interest,” 318–327; and Bakvis, H., Regional Ministers: Power and Influence in the Canadian Cabinet Toronto:University of Toronto Press, 1991), 82–114.
55 Cairns, “The Electoral System and the Party System in Canada,” 67.
56 For example, Lijphart, Democracy in Plural Societies and Electoral Systems and Party Systems.