1 For the United States, see Hibbs, Douglas A., The American Political Economy: Macroeconomics and Electoral Politics in the United States (Cambridge: Harvard University Press, 1987)Google Scholar. For European countries, see Lewis-Beck, Michael, Economics and Elections: The Major Western Democracies (Ann Arbor: University of Michigan Press, 1988).Google Scholar

2 See Archer, Keith and Johnson, Marquis, “Inflation, Unemployment and Canadian Federal Voting Behaviour,” this Journal 21 (1988), 569–84Google Scholar; and Clarke, Harold D. and Kornberg, Allan, “Support for the Canadian Federal Progressive Conservative Party since 1988: The Impact of Economic Evaluations and Economic Issues,” this Journal 25 (1992), 29–53.Google Scholar

3 Monroe, Kristen and Erickson, Lynda, “The Economy and Political Support: The Canadian Case,” Journal of Politics 48 (1986), 616–47CrossRefGoogle Scholar; and Clarke, Harold D. and Zuk, Gary, “The Politics of Party Popularity, Canada 1974–1979,” Comparative Politics 20 (1987), 299–315.CrossRefGoogle Scholar

4 Happy, J. R., “Voter Sensitivity to Economic Conditions: A Canadian-American Comparison,” Comparative Politics 19 (1986), 45–56CrossRefGoogle Scholar. See also, by the same author, “Economic Performance and Retrospective Voting in Canadian Federal Elections,” this Journal 22 (1989), 377–87Google Scholar, and “The Effects of Economic and Fiscal Performance on Incumbency Voting: The Canadian Case,” British Journal of Political Science 22 (1992), 117–30Google Scholar. Finally, see Carmichael, Calum M., “Economic Conditions and the Popularity of the Incumbent Party in Canada,” this Journal 23 (1990), 713–26.Google Scholar

5 Happy, , “Voter Sensitivity to Economic Conditions,”Google Scholar “Economic Performance and Retrospective Voting” and “The Effects of Economic and Fiscal Performance”;; and Carmichael, “Economic Conditions.”;

6 Kramer, Gerald H., “Short-Term Fluctuations in the U.S. Voting Behavior, 1896–1964,” American Political Science Review 65 (1971), 131–43CrossRefGoogle Scholar; Stigler, George J., “General Economic Conditions and National Elections,” American Economic Review 63 (1973), 160–67Google Scholar; Bloom, Howard S. and Price, H. Douglas, “Voter Response to Short-Run Economic Conditions: The Asymmetric Effect of Prosperity and Recession,” American Political Science Review 69 (1975), 1240–54CrossRefGoogle Scholar; Fair, Ray C., “The Effects of Economic Events on Votes for the President,” Review of Economics and Statistics 60 (1978), 159–73CrossRefGoogle Scholar, and “The Effect of Economic Events on Votes for President: 1984 Update,” Political Behavior 10 (1988), 168–79Google Scholar; Lewis-Beck, Michael and Rice, Tom W., “Forecasting Presidential Elections: A Comparison of Naive Models,” Political behavior 6 (1984), 9–12CrossRefGoogle Scholar, and Forecasting Elections (Washington: Congressional Quarterly Press, 1992)Google Scholar; and Campbell, James E. and Wink, Kenneth A., “Trial-Health Forecasts of the Presidential Vote,” American Politics Quarterly 18 (1990), 251–69.CrossRefGoogle Scholar

7 Happy's various articles corne up with contradictory findings with respect to unemployment. In his first two articles (“Voter Sensitivity to Economic Conditions,” and “Economic Performance and Retrospective Voting”;), he reports that unemployment does not seem to have any impact on the vote. In his most recent study (“The Effects of Economic and Fiscal Performance”;), however, he finds that unemployment has a negative effect on incumbency voting. Carmichael's results indicate that, at least for the elections held between 1945 and 1972, bad economic conditions benefited the incumbent party, a pattern opposite to the one found by Happy (see Carmichael, , “Economic Conditions”Google Scholar;). These conflicting results stem in part from different methodological choices. For instance, Happy looks at provincial election outcomes and at provincial economic indicators while Carmichael examines regional election outcomes and relates them to national economic indicators.

8 We start in 1953 because of the constraints imposed by our measure of economic performance. We assume that voters assess performance in relative terms, compared to past economic conditions. The way we have constructed our variables (see Appendix A) requires data going back seven years before the year of the election. Since data on unemployment are available only from 1946 on, we could not incorporate the election of 1949 into our analysis.

9 The same point is made by Dungan, Peter and Murphy, Steve, “Elections and Economics,”Google Scholar Institute for Policy Analysis, University of Toronto, PEAP memo 92–9, November 9,1992.

10 Nominal disposable income has also been tested but was not significant.

11 Clarke, Harold D., Mishler, William and Whitely, Paul, “Recapturing the Falklands: Models of Conservative Popularity, 1979–1983,”; British Journal of Political Science 20 (1990), 66CrossRefGoogle Scholar; see also Mosley, Paul, “Popularity Functions and the Role of the Media: A Pilot Study of the Popular Press,”; British Journal of Political Science 14 (1984), 117–28.CrossRefGoogle Scholar

12 Alt, James E., The Politics of Economic Decline: Economic Management and Political Behavior in Britain since 1964 (New York: Cambridge University Press, 1979).Google Scholar

13 Clarke, Harold D. and Stewart, Marianne, “Dealignment of Degree: Partisan Change in Britain,”; Journal of Politics 46 (1984), 689–717.CrossRefGoogle Scholar

14 The five-year period appears as a reasonable compromise in light of previous works (Alt, in “The Politics of Economic Decline,”; uses three-year moving averages to measure “critical”; rates of inflation, whereas Mosley, in “Popularity Functions,”; employs a longer, seven-year definition) and of the various time frames found in the economic literature.

15 For new short-lived governments, the question of course is whether it is the new or the previous government that is held responsible for economic performance. We have assumed that responsibility is gradually shifted from the previous to the new government. To give a concrete example: for the period of 1963 to 1965, we have assumed that the previous Conservative government was held entirely responsible for economic conditions in the second quarter of 1963, immediately following the election, that it was still held partly responsible in the third quarter, but that responsibility then progressively shifted to the Liberals. As a consequence, change in economic conditions was multiplied by a weight shifting from −1 to −1/2 to 0 to 1/2 to 1 between 1963:2 to 1964:2. We thank one of the referees for suggesting such an approach.

16 Recent studies by Clarke and Kornberg and by Mackuen, Erickson and Stimson suggest that prospective as well as retrospective judgments about the state of the economy may affect party support. See Clarke, and Kornberg, , “Support for the Canadian Federal Progressive Conservative Party”Google Scholar;; and Mackuen, Michael B., Erickson, Robert S. and Stimson, James A., “Peasants or Bankers? The American Electorate and the U.S. Economy,” American Political Science Review 86 (1992), 597–611CrossRefGoogle Scholar. The absence of time-series data on prospective evaluations precludes the testing of the relative influence of retrospective and prospective judgments.

17 In the two elections lost by Louis Saint-Laurent (1957) and Pierre Trudeau (1979), the Liberal party obtained more votes but less seats than the Conservatives.

18 Lemieux, Vincent and Crête, Jean, “Quebec,” in Penniman, H. R., ed., Canada at the Polls, 1979 and 1980 (Washington: American Enterprise Institute, 1981)Google Scholar; see also Crête, Jean and Simard, Carole, “Conjoncture économique et élections: une étude des élections au Québec,” in Crête, Jean, ed., Le comportement électoral au Québec (Chicoutimi: Gaétan Morin, 1984).Google Scholar

19 The leader variable ranges from +1 to −1. It takes the value of +1 if the Liberal leader is from Quebec and no other leader is from that province (elections of 1953 and 1957); of +0.5 if the Liberal leader is from Quebec and a minor party leader is also from Quebec (elections of 1968, 1972, 1974, 1979 and 1980); of 0 if no leader is from Quebec (election of 1958); of −0.5 if the Liberal leader is not from Quebec and a minor-party leader is from that province (elections of 1962, 1963, 1965); of −1 if the Liberal leader is not from Quebec and a major party leader is from Quebec (elections of 1984 and 1988). The coding is based on the assumption that the presence of a leader from Quebec has a greater impact on the vote when that leader heads a major party than when he heads a minor party. It is debatable whether Fabien Roy, the leader of the Social Credit, had any substantial impact on the vote in 1979 and particularly in 1980. But coding the leader variable for 1979 and 1980 as +1 rather than +0.5 leaves our results virtually intact.

20 See Maddala, G. S., Econometrics (New York: McGraw-Hill, 1977), 200Google Scholar. We thank one of the referees for suggesting such a test.

21 We could also note that the F test with respect to seat share is significant at the .05 level, which strengthens the case for distinguishing these elections.

22 See Appendix B for a presentation of the data set.

23 We should note that there is no indication that collinearity is a problem in our analysis. The correlations between unemployment and inflation, unemployment and income growth and inflation and income growth are respectively .43, .09 and −.17. We also ran regressions which included only inflation or income (plus the leader variable). These two variables remained nonsignificant.

24 Johnston, Richard, Public Opinion and Public Policy (Toronto: University of Toronto Press, 1986), 138.Google Scholar

25 In 1968, inflation came up on top of the list of problems Canadians wished the government to tackle. See Meisel, John, Working Papers on Canadian Politics (Montreal: McGill-Queen's University Press, 1972)CrossRefGoogle Scholar. Likewise, Clarke et al. report that inflation was the dominant issue of the 1974 election. See Clarke, Harold, Jenson, Jane, LeDuc, Lawrence and Pammett, Jon, Political Choice in Canada (Toronto: McGraw-Hill Ryerson, 1979), 246Google Scholar. For an excellent overview of the importance of economic issues in federal elections, see Archer and Johnson, “Inflation, Unemployment and Canadian Federal Economic Behaviour.”;

26 As Archer and Johnson aptly put it, “dramatic increases in inflation prior to 1974 gave high salience to the issue, but no party was able to use it to capture large numbers of supporters of the other parties” (see Archer, and Johnson, , “Inflation, Unemployment and Federal Voting Behaviour,” 582–83).Google Scholar

27 Archer and Johnson note that “inflation in 1984 had declined to levels not seen in more than a decade, yet few respondents were prepared to credit the incumbent party for that performance, whereas almost half thought the surge in unemployment was politically significant” (see ibid., 583).

28 As the leader variable shifts from +0.5 to −1 (see Appendix B), the predicted drop in the Liberal vote is eight percentage points.

29 The equation used for the prediction includes all elections held before 1974 except that of 1958, which is problematic in terms of retrospective assessments. For the same reason, the election of 1980 is excluded in the forecasting period.

30 See Greene, Jay P., “Forewarned before Forecast: Presidential Election Forecasting Models and the 1992 Election,” PS: Political Science and Politics 26 (1993), 17–21.Google Scholar

31 See Sankoff, David and Mellos, Koula, “La régionalisation électorale et l'amplification des proportions,” this Journal 6 (1973), 380–99.Google Scholar

32 Clarke, and Kornberg, , “Support for the Canadian Federal Progressive Conservative Party since 1988,” 53.Google Scholar