Senate elections in the United States, 1920-94

British Journal of Political Science

Volumn 30, Issue 3, 2000, Pages 483-506

  Highton, Benjamin ^a  

^a UNIVERSITY OF CALIFORNIA   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

EID: 0034393108     PISSN: 00071234     EISSN: None     Source Type: Journal    
DOI: 10.1017/S000712340000020X     Document Type: Article

References (45)

1. Assessing the effect of state partisan composition on Senate election outcomes is possible because interstate differences in partisanship have continued to be apparent in America. I also investigate the effects of intrastate changes with regard to the rise of the Republican party in the South.
2. The only study I have found that does not model the effect of state partisan composition is Peverill Squire, 'Challengers in US Senate Elections', Legislative Studies Quarterly, 14 (1989), 531-47.
3. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
4. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
5. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
6. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
7. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
8. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
9. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
10. See, for example, Alan Abramowitz, 'Explaining Senate Elections', American Political Science Review, 82 (1988), 383-403; David Brady, Brian Gaines and Douglas Rivers, 'The Incumbency Advantage in the House and Senate: A Comparative Institutional Analysis' (unpublished manuscript, Stanford University, 1994); James E. Campbell and Joe A. Sumners, 'Presidential Coattails in Senate Elections', American Political Science Review, 84 (1990), 513-24; Barbara Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Defining the Parameters of Subpresidential Voting', American Political Science Review, 64 (1970), 836-42; V. O. Key Jr, Politics, Parties, and Pressure Groups, 5th edn (New York: Thomas Y. Crowell, 1964); David Ian Lublin, 'Quality, not Quantity: Strategic Politicians in US Senate Elections, 1952-1990', Journal of Politics, 56 (1994), 228-41; Charles Stewart III, 'A Sequential Model of US Senate Elections', Legislative Studies Quarterly, 14 (1989), 567-601; Mark C. Westlye, Senate Elections and Campaign Intensity (Baltimore, Md: The Johns Hopkins University Press, 1991). The period of time covered in these studies ranges from ten to thirty-eight years.
11. However, my analysis reveals that even over periods as short as ten years, substantial change in the effect of partisan composition may be observed.
12. The single exception to the common practice of assuming that the effect of state partisan composition on Senate election results is invariant over time is Warren Lee Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models', American Political Science Review, 67 (1973), 1213-34. Kostroski analysed elections over a twenty-two year period, from 1948 to 1970, and found that the 'importance of party has undergone a sharp, secular decline' (p. 1229). However, as I have explained elsewhere in Benjamin Highton, 'Parties, Candidates, and Issues in US Senate Elections' (unpublished manuscript, University of California, Berkeley, 1998), Kostroski employs a flawed model, making his conclusions dubious. In addition, Kostroski uses a crude indicator of state partisan composition that he calls 'base party vote', which is the value of 'the lowest percent of votes for House candidates cast in that state during the five election years of which the "target year" [the year of the Senate election] ... constitutes the midpoint year' (Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models,' p. 1226). I will demonstrate that this sort of measure tends to underestimate the effect of state partisan composition on election outcomes.
13. The single exception to the common practice of assuming that the effect of state partisan composition on Senate election results is invariant over time is Warren Lee Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models', American Political Science Review, 67 (1973), 1213-34. Kostroski analysed elections over a twenty-two year period, from 1948 to 1970, and found that the 'importance of party has undergone a sharp, secular decline' (p. 1229). However, as I have explained elsewhere in Benjamin Highton, 'Parties, Candidates, and Issues in US Senate Elections' (unpublished manuscript, University of California, Berkeley, 1998), Kostroski employs a flawed model, making his conclusions dubious. In addition, Kostroski uses a crude indicator of state partisan composition that he calls 'base party vote', which is the value of 'the lowest percent of votes for House candidates cast in that state during the five election years of which the "target year" [the year of the Senate election] ... constitutes the midpoint year' (Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models,' p. 1226). I will demonstrate that this sort of measure tends to underestimate the effect of state partisan composition on election outcomes.
14. The single exception to the common practice of assuming that the effect of state partisan composition on Senate election results is invariant over time is Warren Lee Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models', American Political Science Review, 67 (1973), 1213-34. Kostroski analysed elections over a twenty-two year period, from 1948 to 1970, and found that the 'importance of party has undergone a sharp, secular decline' (p. 1229). However, as I have explained elsewhere in Benjamin Highton, 'Parties, Candidates, and Issues in US Senate Elections' (unpublished manuscript, University of California, Berkeley, 1998), Kostroski employs a flawed model, making his conclusions dubious. In addition, Kostroski uses a crude indicator of state partisan composition that he calls 'base party vote', which is the value of 'the lowest percent of votes for House candidates cast in that state during the five election years of which the "target year" [the year of the Senate election] ... constitutes the midpoint year' (Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models,' p. 1226). I will demonstrate that this sort of measure tends to underestimate the effect of state partisan composition on election outcomes.
15. David R. Mayhew, 'Congressional Elections: The Case of the Vanishing Marginals,' Polity 6 (1974), 295-317.
16. From 1946 to 1994, 77 per cent of Senate incumbents seeking re-election were successful. The corresponding number for House incumbents is 92 per cent (Norman J. Ornstein, Thomas E. Mann and Michael J. Malbin. Vital Statistics on Congress 1997-1998 (Washington, DC: Congressional Quarterly Inc., 1998).
17. Hinckley, 'Incumbency and the Presidential Vote in Senate Elections: Denning the Parameters of Subpresidential Voting', p. 840.
18. Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models'; Michael Krashinsky and William J. Milne, 'The Effects of Incumbency in US Congressional Elections, 1950-1988', Legislative Studies Quarterly, 18 (1993), 321-44.
19. Kostroski, 'Party and Incumbency in Postwar Senate Elections: Trends, Patterns, and Models'; Michael Krashinsky and William J. Milne, 'The Effects of Incumbency in US Congressional Elections, 1950-1988', Legislative Studies Quarterly, 18 (1993), 321-44.
20. See Highton, 'Parties, Candidates, and Issues in US Senate Elections', Appendix 2.A.
21. Brady, Gaines and Rivers, 'The Incumbency Advantage in the House and Senate'.
22. Gary C. Jacobson, The Politics of Congressional Elections, 4th edn (New York: Addison Wesley Longman, 1997), p. 24.
23. Abramowitz, 'Explaining Senate Elections'; Squire, 'Challengers in US Senate Elections'.
24. Abramowitz, 'Explaining Senate Elections'; Squire, 'Challengers in US Senate Elections'.
25. Campbell and Sumners, 'Presidential Coattails in Senate Elections'.
26. John R. Hibbing and John R. Alford, 'Economic Conditions and the Forgotten Side of Congress: A Foray into US Senate Elections', British Journal of Political Science, 12 (1982), 505-16; Lublin, 'Quality, not Quantity' ; Dennis M. Simon, Charles W. Ostrom Jr and Robin F. Marra, 'The President, Referendum Voting, and Subnational Elections in the United States,' American Political Science Review, 85 (1991), 1177-92.
27. John R. Hibbing and John R. Alford, 'Economic Conditions and the Forgotten Side of Congress: A Foray into US Senate Elections', British Journal of Political Science, 12 (1982), 505-16; Lublin, 'Quality, not Quantity' ; Dennis M. Simon, Charles W. Ostrom Jr and Robin F. Marra, 'The President, Referendum Voting, and Subnational Elections in the United States,' American Political Science Review, 85 (1991), 1177-92.
28. John R. Hibbing and John R. Alford, 'Economic Conditions and the Forgotten Side of Congress: A Foray into US Senate Elections', British Journal of Political Science, 12 (1982), 505-16; Lublin, 'Quality, not Quantity' ; Dennis M. Simon, Charles W. Ostrom Jr and Robin F. Marra, 'The President, Referendum Voting, and Subnational Elections in the United States,' American Political Science Review, 85 (1991), 1177-92.
29. Lublin, 'Quality, not Quantity'.
30. Several sources provided the necessary data: Robert A. Diamond, ed. Guide to US Elections (Washington, DC: Congressional Quarterly Inc., 1975), Richard M. Scammon and Alice V. McGillivray, eds, America Votes 20: A Handbook of Contemporary American Election Statistics (Washington, DC: Congressional Quarterly Inc., 1993), and US Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970, Bicentennial Edition, Part 1 (Washington DC, USBC, 1975).
31. Several sources provided the necessary data: Robert A. Diamond, ed. Guide to US Elections (Washington, DC: Congressional Quarterly Inc., 1975), Richard M. Scammon and Alice V. McGillivray, eds, America Votes 20: A Handbook of Contemporary American Election Statistics (Washington, DC: Congressional Quarterly Inc., 1993), and US Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970, Bicentennial Edition, Part 1 (Washington DC, USBC, 1975).
32. Several sources provided the necessary data: Robert A. Diamond, ed. Guide to US Elections (Washington, DC: Congressional Quarterly Inc., 1975), Richard M. Scammon and Alice V. McGillivray, eds, America Votes 20: A Handbook of Contemporary American Election Statistics (Washington, DC: Congressional Quarterly Inc., 1993), and US Bureau of the Census, Historical Statistics of the United States, Colonial Times to 1970, Bicentennial Edition, Part 1 (Washington DC, USBC, 1975).
33. Other than Minnesota, only Virginia had more than one race excluded for this reason. In both 1970 and 1976, Harry Byrd Jr, who won a special election as a Democrat in 1966 to replace his father Harry Byrd, ran as an independent in Virginia. Both parties nominated candidates to oppose him in 1970 while only the Democrats contested the 1976 race. Byrd Jr won these elections. In Minnesota the Farmer-Labor party was a vigorous alternative to the two major parties between 1922 and 1942, winning five elections and finishing second three times. All Minnesota Senate elections during this period were deleted from the sample for analysis.
34. Gary Jacobson introduced this variable in his analysis of the effects of campaign spending in congressional elections (Gary C. Jacobson, 'The Effects of Campaign Spending in Congressional Elections', American Political Science Review, 72 (1978), 469-91).
35. Lublin, 'Quality, not Quantity, p. 230.
36. Jacobson, 'The Effects of Campaign Spending in Congressional Elections,' p. 471.
37. Andrew Gelman and Gary King, 'Estimating Incumbency Advantage without Bias', American Journal of Political Science, 34 (1990), 1142-64.
38. Because my concern is to estimate the overall incumbency advantage, I do not include indicators that tap characteristics of challengers and their campaigns. This approach parallels that of Gelman and King, 'Estimating the Incumbency Advantage without Bias.'
39. See, for example, Mayhew, 'Congressional Elections: The Case of the Vanishing Marginals',
40. and Gelman and King, 'Estimating Incumbency Advantage without Bias'.
41. Jacobson, The Politics of Congressional Elections, 4th edn, chap. 3, provides a detailed review of the literature and evidence regarding the incumbency advantage in House elections.
42. Christopher H. Achen, Interpreting and Using Regression, Sage University Papers, Quantitative Applications in the Social Sciences, Number 07-029 (Beverly Hills, Calif.: Sage Publications, 1982), pp. 68-77.
43. The evidence supporting this claim was provided in the previous section.
44. The corresponding standard errors are 1.0 and 0.07.
45. Moreover, the possibility that there was no increase in the incumbent advantage during this time can confidently be rejected (p<0.01).