Published online by Cambridge University Press: 27 January 2009
The 1980 American national elections produced two highly salient and widely-discussed departures from past tradition. For the first time in forty-eight years, a previously elected incumbent president seeking re-election was defeated. At the same time the Republicans, the traditional minority party, seized control of the upper house. Both events figured heavily in discussions of the American electorate's supposed turn to the right. Much of this emphasis on ideological change may be misplaced. A third, although much less noted, departure from tradition also occurred in 1980: for the first time since 1932, real disposable per capita income declined in the year preceding a presidential election. The importance of such a decline for the electoral fate of an incumbent president has been well documented. Numerous previous studies have indicated a powerful and consistent impact of changes in economic conditions on voting behaviour in presidential elections. The influence of economic conditions on the fate of the president's party in House elections is also well-documented. In the case of Senate elections, however, the literature is curiously silent.
1 See Tufte, Edward R., Political Control of the Economy (Princeton, N.J.: Princeton University Press, 1978).Google Scholar
2 See, for example, Kramer, Gerald H., ‘Short-Term Fluctuations in U.S. Voting Behavior, 1896–1964’. American Political Science Review, LXV (1971), 131–43CrossRefGoogle Scholar; Tufte, Edward R., ‘Determinants of the Outcomes of Midterm Congressional Elections’, American Political Science Review, LXIX (1975), 812–26CrossRefGoogle Scholar; Bloom, Howard S. and Price, H. Douglas, ‘Voter Response to Short-Run Economic Conditions’. American Political Science Review, LXIX (1975). 1240–54CrossRefGoogle Scholar; Hibbing, John R. and Alford, John R., ‘The Electoral Impact of Economic Conditions: Who Is Held Responsible?’ American Journal of Political Science, XXV (1981), 423–39.CrossRefGoogle Scholar
3 See, for example, Hinckley, Barbara, ‘House Re-elections and Senate Defeats: The Role of the Challenger’, British Journal of Political Science, X (1980), 441–60CrossRefGoogle Scholar; and Abramowitz, Alan I., ‘A Comparison of Voting for U.S. Senator and Representative in 1978’, American Political Science Review, LXXIV (1980), 633–40.CrossRefGoogle Scholar While much has been written on House-Senate comparisons based on the 1978 CPS election survey, some caution with regard to the Senate findings is clearly in order. See, for example, Mann, Thomas E. and Wolfinger, Raymond E., ‘Candidates and Parties in Congressional Elections’, American Political Science Review, LXXIV (1980), 617–32.CrossRefGoogle Scholar
4 See Cover, Albert D. and Mayhew, David R., ‘Congressional Dynamics and the Decline of Competitive Congressional Elections’, in Dodd, Lawrence C. and Oppenheimer, Bruce I., eds, Congress Reconsidered (Washington: Congressional Quarterly, Inc., 1981), p. 72.Google Scholar
5 A similar source of potential extraneous variance involves the on-year, off-year pattern of success for the in-party in congressional elections. The equations in Table 2 were re-estimated with the addition of a dummy scored 1 for on-years and 0 for off-years. The coefficient for this dummy was not significant for either the Senate or the House equations, nor were the other parameters in the equations appreciably altered.
6 The additional economic variables of inflation and unemployment were also tested, but in no case did either of these variables approach significance. The possibility of an asymmetric effect for good and bad economic conditions, as suggested by Bloom and Price in ‘Voter Response to Short-Run Economic Conditions’, was also examined for our data on the Senate but received no support.
7 See Blalock, Herbert M. Jr., Causal Inference in Nonexperimental Research (Chapel Hill: University of North Carolina Press, 1964), pp. 97–126.Google Scholar
8 See Gudgin, G. and Taylor, P. J., ‘The Decomposition of Electoral Bias in a Plurality Election’, British Journal of Political Science, X (1980), 515–21CrossRefGoogle Scholar; and Tufte, Edward R., ‘The Relationship Between Seats and Votes in Two-Party Systems’, American Political Science Review, LXVII (1973), 540–54.CrossRefGoogle Scholar
9 For an exception, see Hudson, E. A. and Weavers, M., ‘New Zealand General Elections: A Formal Analysis’, Political Science, XXIX (1977), 1–18.CrossRefGoogle Scholar
10 These analyses were repeated including measures of unemployment and inflation, with results very similar to those reported in footnote 6, i.e., nc relationship even approached significance for either variable in either body. Similarly, the examination of asymmetry was repeated for percentage of seats. The results were again completely compatible with those reported for percentage vote share.
11 See, for example, Cover, and Mayhew, , ‘Congressional Dynamics’.Google Scholar
12 This insulation in the Senate was originally far more extreme because in addition to the longer, staggered terms of senators, the selection method was by state legislature rather than by popular election.
13 If, for example, a party gains 30 per cent in a given Senate election, it is gaining only about ten seats in terms of overall Senate membership. A similar 30 per cent gain in House elections would be a 30 per cent gain in House membership since the entire membership is determined in each election.
14 The standard error of the regression equation is 12·84. Thus, the 45·56 per cent of the seats that the Democrats would have needed to capture in order to retain control of the Senate falls 0·72 standard errors above the predicted value of 36·36. The probability of a value 0·72 standard errors or more above the mean (the range necessary for Democratic control) equals 0·24.