Hostname: page-component-7c8c6479df-fqc5m Total loading time: 0 Render date: 2024-03-18T08:21:56.772Z Has data issue: false hasContentIssue false

Explaining the Contemporary Alignment of Race and Party: Evidence from California's 1946 Ballot Initiative on Fair Employment

Published online by Cambridge University Press:  02 September 2008

Anthony S. Chen
Affiliation:
University of Michigan, Ann Arbor
Robert W. Mickey
Affiliation:
University of Michigan, Ann Arbor
Robert P. Van Houweling
Affiliation:
University of California, Berkeley

Abstract

Why do most African Americans and other racial liberals vote Democratic, whereas most racial conservatives—largely whites—vote Republican? To what extent is this alignment of race and party attributable to the strategic choice of GOP elites to take the party in a racially conservative direction during the mid-1960s? This paper exploits a little-known ballot initiative in postwar California to shed light on the question. Proposition 11, as it was known, would have outlawed discrimination in employment if it had passed. Instead, it failed by more than a two-to-one margin. Drawing on archival and statistical evidence, including the ecological analysis of precinct-level election returns, we find that Republican voters were much more likely than Democratic voters to oppose Proposition 11, despite Republican Governor Earl Warren's well-known support for fair employment practices (FEP) legislation. We conclude that many Republican voters tended strongly toward racial conservatism well before Republican elites decided to pursue racially conservative policies in the mid-1960s. We suggest that the emergence of the contemporary alignment of race and party may have been less contingent on elite strategy and more structurally determined than the conventional wisdom allows.

Type
Research Article
Copyright
Copyright © Cambridge University Press 2008

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1. Here, “racial alignment” refers merely to the distribution of voters by race across the two parties. We accept the criticisms of the traditional concept of “realignment.” Carmines, Edward G. and Stimson, James A., Issue Evolution: Race and the Transformation of American Politics (Princeton: Princeton University Press, 1989), 2126Google Scholar; Mayhew, David R., Electoral Realignments (New Haven: Yale University Press, 2002)Google Scholar.

2. On the gradual sorting of southern whites, see, e.g., Black, Earl and Black, Merle, The Rise of Southern Republicanism (Cambridge: Harvard University Press, 2003)Google Scholar, and Green, Donald, Palmquist, Bradley, and Schickler, Eric, Partisan Hearts & Minds: Political Parties and the Social Identities of Voters (New Haven: Yale University Press, 2002), ch. 6Google Scholar. On the role of class in American party alignments, see Manza, Jeff and Brooks, Clem, Social Cleavages and Political Change: Voter Alignments and U.S. Party Coalitions (New York: Oxford University Press, 1999), chs. 2–3Google Scholar; Bartels, Larry M., Unequal Democracy: The Political Economy of the New Gilded Age (Princeton: Princeton University Press, 2008), ch. 3Google Scholar; Stonecash, Jeffrey, Brewer, Mark D., Petersen, R. Eric et al. , “Class and Party: Secular Realignment and the Survival of Democrats Outside the South,” Political Research Quarterly 53 (2000): 731–52CrossRefGoogle Scholar. On the demise of the New Deal coalition, see Fraser, Steve and Gerstle, Gary, eds., The Rise and Fall of the New Deal Order (Princeton: Princeton University Press, 1989)Google Scholar.

3. Our usage of the terms “racial liberalism” and “racial conservatism” is broadly consistent with the usage in Carmines and Stimson.

4. As Richard Valelly points out, the U.S. stands alone among democratic polities in having enfranchised, disenfranchised, and reenfranchised groups of its citizens. See Valelly, Richard, The Two Reconstructions: The Struggle for Black Empowerment (Chicago: University of Chicago Press, 2004), ch. 1CrossRefGoogle Scholar.

5. Carmines and Stimson, Issue Evolution, xii, 141, 154, 160. The issue evolution theory is extended in Stimson, James A., Tides of Consent: How Public Opinion Shapes American Politics (Cambridge: Cambridge University Press, 2004), ch. 3CrossRefGoogle Scholar.

6. Qtd. in Carmines and Stimson, Issue Evolution, 42.

7. Carmines and Stimson, Issue Evolution, xi–xii, 42, 45, 47; New York Times, 13 May 1964, 1, 22; 21 June 1964, 1; Rick Perlstein, Before the Storm: Barry Goldwater and the Unmaking of the American Consensus (New York: Hill and Wang, 2001), 363–4. The view that the year 1964 was causally “critical” for producing the contemporary sorting out of voters into parties is advanced by the vast majority of political scientists and historians writing on the subject.

8. Carmines and Stimson, Issue Evolution.

9. Robert W. Mickey, Paths Out of Dixie: The Democratization of Authoritarian Enclaves in America's Deep South (Princeton University Press, forthcoming), ch. 2.

10. Carmines and Stimson, Issue Evolution, 184–85, 190.

11. Carmines and Stimson, Issue Evolution, 134, 150–156; Thomas Byrne Edsall with Mary D. Edsall, Chain Reaction: The Impact of Race, Rights, and Taxes on American Politics, rev. ed. (New York: Norton, 1992), 7. The potential for mobilizing white ethnics via “law and order” appeals began earlier than usually appreciated. Indeed, during the summer of 1964, observers referred to crowd disturbances involving blacks as “Goldwater rallies.” Jeremy D. Mayer, “LBJ Fights the White Backlash: The Racial Politics of the 1964 Presidential Campaign,” Prologue 33 (Spring 2001), accessed online at http://www.archives.gov/publications/prologue.html. On the increasing prominence of “law and order” rhetoric tied to racial politics in Republican forums before 1964, see Perlstein, Before the Storm.

12. Carmines and Stimson, Issue Evolution, 190.

13. Carmines and Stimson, Issue Evolution, 117, 134.

14. Carmines and Stimson, Issue Evolution, 16, 134, 150–56, 161.

15. The emphasis on elites is not unique to Carmines and Stimson or Edsall and Edsall. For instance, elite racial appeals are granted equally significant causal weight in explaining a variety of electoral outcomes. See Mendelberg, Tali, The Race Card: Campaign Strategy, Implicit Messages, and the Norm of Equality (Princeton: Princeton University Press, 2001)Google Scholar.

16. This contingency is, of course, a necessary corollary of the emphasis on elite agency—different elite choices can produce different paths to different alignments.

17. Carmines and Stimson, Issue Evolution, 9, 11, 186. Carmines and Stimson declare themselves “intellectually at war … with any notion of inevitability.” They “encounter no situations in the political evolution of race where … it could only have happened as it did.” They describe their scenarios, instead, as “… more akin to Tolstoy battle scenes, where calculation, force, confusion, and chance commingle to produce an outcome, the appearance of which is only orderly after the fact.” Carmines and Stimson, Issue Evolution, 16, 18.

18. Leading statements of the occurrence of an important “white backlash” include: Edsall with Edsall, Chain Reaction; Matusow, Allen, The Unraveling of America: A History of Liberalism in the 1960s (New York: Harper & Row, 1984)Google Scholar, Abigail, and Thernstrom, Stephen, America in Black and White: One Nation, Indivisible (New York: Simon and Schuster, 1997)Google Scholar, and Rieder, Jonathan, Canarsie: The Jews and Italians of Brooklyn Against Liberalism (Cambridge: Harvard University Press, 1985)Google Scholar.

19. Political scientists and historians have employed a “white backlash” frame to interpret other episodes throughout the twentieth century, including struggles over the FEPC; school desegregation cases; white (and especially southern) mass response to the passage of landmark civil rights legislation; welfare policies and “welfare dependency;” and, most recently, affirmative action in hiring and higher education.

20. Stimson, Tides of Consent, ch. 3;Adams, Greg D., “Abortion: Evidence of Issue Evolution,” American Journal of Political Science 41 (1997): 718–37CrossRefGoogle Scholar; and Wolbrecht, Christina, The Politics of Women's Rights: Parties, Positions, and Change (Princeton: Princeton University Press, 2000)Google Scholar. Stimson argues that we must accept as reasonable the possibility that in the early 1970s, feminist activists—many of them involved in other ostensibly leftist causes—could have entered the Republican party and taken others with them. Stimson, Tides of Consent, 68.

21. Sugrue, Thomas J., The Origins of the Urban Crisis: Race and Inequality in Postwar Detroit (Princeton: Princeton University Press, 1996)Google Scholar; Hirsch, Arnold R., Making the Second Ghetto: Race and Housing in Chicago, 1940–1960 (Chicago, 1983)Google Scholar, and Hirsch, Arnold R., “Massive Resistance in the Urban North: Trumbull Park, Chicago, 1953–1966,” Journal of American History 82 (1995): 522550CrossRefGoogle Scholar; Self, Robert O., American Babylon: Race and the Struggle for Postwar Oakland (Princeton: Princeton University Press, 2003)Google Scholar; Kurashige, Scott, The Shifting Grounds of Race: Black and Japanese Americans in the Making of Multiethnic Los Angeles (Princeton: Princeton University Press, 2008)Google Scholar; Countryman, Matthew, Up South: Civil Rights and Black Power in Philadelphia (Philadelphia: University of Pennsylvania Press, 2006)Google Scholar; Biondi, Martha, To Stand and Fight: The Struggle for Civil Rights in Postwar New York City (Cambridge, MA: Harvard University Press, 2003)Google Scholar. See also Lockard, Duane, Toward Equal Opportunity: A Study of State and Local Antidiscrimination Laws (New York: Macmillan, 1968)Google Scholar; Fine, Sidney A., Expanding the Frontiers of Civil Rights: Michigan, 1948–1968 (Detroit: Wayne State University Press, 2000)Google Scholar. This paragraph draws on Daniel T. Kryder and Robert W. Mickey, “The Politics of Backlash: Consequences of a Metaphor” (unpublished manuscript in the authors' possession), as well as Anthony S. Chen, “From Fair Employment to Equal Employment Opportunity and Beyond,” (PhD diss., University of California, Berkeley, 2002), 275–95.

22. Sugrue, The Origins of the Urban Crisis, 173, 210, 222–24, 227, 267–68. See also “Crabgrass-Roots Politics: Race, Rights, and the Reaction Against Liberalism in the Urban North, 1940–1964,” Journal of American History 82 (1995): 551–78.

23. Anthony S. Chen, The Fifth Freedom: Jobs, Politics, and Civil Rights in the United States, 1941–1972 (Princeton University Press, forthcoming); also see Chen, Anthony S., “The Party of Lincoln and the Politics of State Fair Employment Practices Legislation in the North, 1945–1964,” American Journal of Sociology 112 (2007): 1713–74CrossRefGoogle Scholar, and Chen, Anthony S., “‘The Hitlerian Rule of Quotas’: Racial Conservatism and the Politics of Fair Employment Practice Legislation in New York State, 1941–1945,” Journal of American History 92 (2006): 1238–64CrossRefGoogle Scholar. The seminal volume on the politics of state civil rights legislation is Lockard, Toward Equal Opportunity. But see also Erikson, Robert S., “The Relationship between Party Control and Civil Rights Legislation in the States,” Western Political Quarterly 24 (1971): 178182CrossRefGoogle Scholar; Collins, William J., “The Political Economy of State-Level Fair-Employment Laws, 1940–1964,” Explorations in Economic History (2003): 2451CrossRefGoogle Scholar.

24. Daniel T. Kryder and Robert W. Mickey, “The Politics of Backlash”; Anthony S. Chen, “From Fair Employment to Equal Employment Opportunity and Beyond,” 275–295; Chen, The Fifth Freedom.

25. See, e.g., Lipset, Seymour Martin and Rokkan, Stein, “Cleavage Structures, Party Systems, and Voter Alignments,” in Party Systems and Voter Alignments: Cross-National Perspectives, ed. Lipset, and Rokkan, (New York: Free Press, 1967), 164Google Scholar.

26. Gerring, John, Party Ideologies in America, 1828–1996 (Cambridge: Cambridge University Press, 1998), chapters 4 and 7CrossRefGoogle Scholar; Bartels, Larry M., “What's the Matter with What's the Matter with Kansas?,” Quarterly Journal of Political Science 1 (2006): 201–26CrossRefGoogle Scholar. Also see Erikson, Robert S., Lancaster, Thomas D., and Romero, David W., “Group Components and of the Presidential Vote, 1952–1984,” Journal of Politics 51 (1989): 337–46CrossRefGoogle Scholar.

27. Lee, Taeku, Mobilizing Public Opinion: Black Insurgency and Racial Attitudes in the Civil Rights Era (Chicago: University of Chicago Press, 2002), ch. 5Google Scholar.

28. David Karol, Coalition Management: Explaining Party Position Change in American Politics, unpublished manuscript. Karol also shows that nonsouthern Democratic congressional representatives began voting (slightly) more liberally than Republicans on civil rights proposals as early as the 1940s.

29. Feinstein, Brian and Schickler, Eric, “Platforms and Parties: The Civil Rights Realignment Reconsidered,” Studies in American Political Development 22 (2008): 131CrossRefGoogle Scholar.

30. For an important critique, see Abramowitz, Alan I., “Issue Evolution Reconsidered: Racial Attitudes and Partisanship in the U.S. Electorate,” American Journal of Political Science 38 (1994): 124CrossRefGoogle Scholar.

31. Carmines and Stimson, Issue Evolution, 189.

32. Otherwise, an alternative perspective—that the alignment we have today was structurally determined by class cleavages—would still be a live option.

33. Moreover, in the twentieth century, public sector employment has been of great importance to black America.

34. Of course, relative to other democratic polities, the midpoint of the left-right dimension in the United States through most of its history has been on the right. Still, the distance between the two major parties has been persistent, and the parties have often been polarized on this dimension. Gerring, Party Ideologies in America.

35. Given the black American experience with southern state governments, as well as the fact that federalism effectively disempowers subnational governments from redistributing income, it is understandable that black America's target, and dim hope, has been the federal government. Mickey, Paths Out of Dixie, ch. 2; Peterson, Paul E., City Limits (Chicago: University of Chicago Press, 1981)CrossRefGoogle Scholar; Peterson, Paul E., The Price of Federalism (Washington: Brookings, 1995)Google Scholar; Dawson, Michael, Behind the Mule (Princeton: Princeton University Press, 1993)Google Scholar.

36. However mistily Americans now conjure up the 1963 march on Washington as a call for children of all races to play together, the march was, at the time, a march for “jobs and justice.”

37. On the specific importance of the Second World War to racial politics and the rise of the civil rights movement, see, among many others, Reed, Merl E., Seedtime for the Modern Civil Rights Movement: The President's Committee on Fair Employment Practice, 1941–1946 (Baton Rouge: Louisiana State University Press, 1991)Google Scholar, and Watson, Denton, Lion in the Lobby: Clarence Mitchell, Jr.'s Struggle for the Passage of Civil Rights Laws (New York: Morrow, 1990)Google Scholar.

38. A different incarnation of the first scenario is that Republican elites call for civic and social equality, and Democratic elites do as well. The parties then split the racially liberal vote. But this incarnation falls prey to the same problem as the previous one. The GOP could not have limited itself to supporting just civic and social equality, as it would have put them at a competitive disadvantage with Democrats. There would have been pressure for them to support all three types of equality.

39. Gerring, Party Ideologies in America, ch. 7.

40. Feinstein and Schickler, “Parties and Platforms.”

41. Meanwhile, racial conservatives would have had a difficult time finding a partisan home, and it is possible there would have been a sustained effort to launch a third party.

42. Patterson, James T., Mr. Republican: A Biography of Robert A. Taft (Boston: Houghton Mifflin Company, 1972)Google Scholar.

43. It bears mentioning that Carmines and Stimson (Issue Evolution, 150) do offer some evidence on the racial politics of Republican voters in the mid-1960s. Drawing on a time series constructed out of data from the National Election Studies and the Harris Poll, they find that Republican and Democratic identifiers held very similar racial attitudes through 1962. In fact, Republicans had been more liberal than Democrats since 1955. It was only in 1963 to 1964, after Kennedy's introduction of civil rights legislation and Goldwater's rejection of it, that Republicans became much more conservative than Democrats. However, their measure of racial attitudes is problematic because it is based on survey items about desegregation. Because racial segregation was largely identified with the South, Republicans had no problem expressing their disdain for the practice. It did not seem to implicate any Republican constituencies, which were located mostly in the North, and criticizing segregation would only exacerbate tensions between the southern and northern wing of the Democratic Party. In point of fact, criticizing segregation could conceivably heighten their partisan advantage over Democrats by exposing them for their contradictory stance. A different source of data is needed to make valid inferences about the racial attitudes of Republican voters. We enumerate the features of such data below.

44. This feature of the case is also useful for assessing the claim that the opposition of nonsouthern whites to civil rights (i.e., white backlash) was triggered by the radicalization of the civil rights agenda. Also worth pointing out is that blacks exercised the franchise in northern states before 1964.

45. Matthew Lassiter has argued that the distinction between de facto and de jure segregation has blurred over time; he also argues that it was a false dichotomy even for much of the postwar period. See Lassiter, Matthew, The Silent Majority: Suburban Politics in the Sunbelt South (Princeton: Princeton University Press, 2006)Google Scholar. We agree with many aspects of his analysis. Our reference to de jure segregation is simply meant to point out that the racial politics of Republican voters are difficult to measure if poll respondents were asked to express their opinion about practices that were identified by much of the public as distinctively southern. This identification was certainly contestable. After a number of speaking engagements in the North, Alabama Governor George Wallace remarked, with wonder and glee, “The whole United States is Southern!” Payne, Charles M., “‘The Whole United States is Southern!’ Brown v. Board and the Mystification of Race,” Journal of American History 91 (2004): 8391CrossRefGoogle Scholar.

46. Kevin Allen Leonard, “Years of Hope, Days of Fear: The Impact of World War II on Race Relations in Los Angeles,” (PhD diss., Department of History, University of California, Davis, 1992), 328; Sides, Josh, L.A. City Limits: African American Los Angeles from the Great Depression to the Present (Berkeley: University of California Press, 2003)Google Scholar; Kurashige, The Shifting Grounds of Race, 218. See also Leonard, Kevin Allen, Battle for Los Angeles: Racial Ideology and World War II (Albuquerque: University of New Mexico Press, 2006)Google Scholar.

47. Kesselman, Louis C., The Social Politics of FEPC: A Study in Reform Pressure Movements (Chapel Hill: University of North Carolina Press, 1948)Google Scholar; Ruchames, Louis, Race, Jobs, and Politics (New York: Columbia University Press, 1953)Google Scholar; Garfinkel, Herbert, When Negroes March (Glencoe, IL: Free Press, 1959)Google Scholar; Reed, Merl E., Seedtime for the Modern Civil Rights Movement (Baton Rouge: Louisiana State University Press, 1991)Google Scholar; Kryder, Daniel T., Divided Arsenal (Cambridge: Cambridge University Press, 2000)Google Scholar.

48. Patterson, Mr. Republican.

49. Ruchames, Race, Jobs, and Politics; Reed, Seedtime for the Modern Civil Rights Movement.

50. Chen, “‘The Hitlerian Rule of Quotas’,” 1238–64.

51. Los Angeles Times, 15 Mar. 1945, 2; ibid., 5 Apr. 1945, A4.

52. Assembly Final History 1945, 178.

53. Sides, L.A. City Limits, 62–66; Kurashige, The Shifting Grounds of Race.

54. Assembly Final History 1945, 178; Chen, The Fifth Freedom; Los Angeles Times, 14 June 1945, 2.

55. Los Angeles Times, 14 June 1945, 2; ibid., 18 June 1945, 2. Sponsored by Assemblyman Sam L. Collins, Warren's bill, A.B. 1399, proposed the establishment of a “Commission on Political and Economic Equality” for the “purpose of research and education in this field.” See Beach Vasey to Herbert Mitgang, 26 Nov. 1946, F3640:3657, Earl Warren Papers, California State Archives (hereafter EW-CSA). The commission would investigate the enforcement of existing statutes against discrimination and determine whether new legislation was necessary. See Los Angeles Sentinel, 11 Jan. 1945, 1.

56. Los Angeles Times, 13 Dec. 1945, 1; ibid., 6 Jan. 1945, 2; ibid., 8 Jan. 1946, 1, 8; Beach Vasey to Earl Warren 14 Jan. 1946, F3640:8401, EW-CSA.

57. Beach Vasey to Earl Warren, F3640: 6100, 28 Feb. 1945, EW-CSA.

58. William Nickerson, Jr., to Earl Warren, 7 Sept. 1945, F3640: 8452, EW-CSA. Albert Johnson called to make a similar point to Warren about the significance of FEP to the black electorate. See Beach Vasey to Earl Warren, 17 Aug. 1945, F3640: 3656, EW-CSA.

59. Beach Vasey to Thomas E. Dewey, 15 Mar. 1945, F3540: 3656, EW-CSA; Beach Vasey to Earl Warren, 24 Mar. 1945, ibid.; J. Welsh to Beach Vasey, 23 Mar. 1945, ibid.

60. Walter A. Gordon to Earl Warren, 29 Nov. 1945, F3630: 8452, EW-CSA.

61. Beach Vasey to Earl Warren, 18 Dec. 1945, F3640: 8452, EW-CSA.

62. Beach Vasey to Earl Warren, 14 Jan. 1946, F3640: 8452, EW-CSA.

63. Los Angeles Times, 16 Feb. 1946, 2. Warren's bill in 1946 was A.B. 97.

64. Los Angeles Times, 17 Jan. 1946, 2; “Big Business Opposed, GOP Knifed State FEPC Measure,” California Eagle, 24 Jan. 1946. See Chen, “The Party of Lincoln and the Politics of State Fair Employment Practices Legislation in the North,” 1736. The Hawkins bill was numbered A.B. 11, and the Evans bill was numbered A.B. 31.

65. “Group Maps Plan for FEPC Initiative Campaign,” California Eagle, 11 Nov. 1945.

66. “The Truth about the FEPC Bill,” California Eagle, 21 Feb. 1946.

67. Proposed Amendments to Constitution: Propositions and Proposed Laws Together with Arguments, General Election, 5 Nov. 1946.

68. California Eagle, 11 Oct. 1947, 7.

69. Leonard, “Years of Hope, Days of Fear.”

70. Beach Vasey to Earl Warren, 10 Oct. 1946, F3640: 8854, EW-CSA.

71. Verne Scoggins to F.A. Ferguson, 15 Oct. 1946, F3640: 8854, EW-CSA.

72. Earl Warren to Felix A. Manley, 13 Dec. 1946, F3640: 8854, EW-CSA.

73. Ibid.

74. Charles W. Fisher to Earl Warren, 2 Apr. 1945, F3640: 7769, EW-CSA.

75. Roger D. Lapham to Earl Warren, 23 May 1945, F3640: 7769, EW-CSA.

76. Los Angeles Times, 25 Aug. 1946, A4.

77. Minutes, 6 Sept. 1946, 5, Folder: 1946–1947, Box 2242, records of the California Chamber of Commerce, California State Library.

78. Los Angeles Times, 11 Oct. 1946, 2.

79. F.A. Ferguson to Earl Warren, 14 Oct. 1946, qtd. in Chen, The Fifth Freedom.

80. Women of the Pacific, “Employers! FEPC IS On Your Doorstep!” n.d. [circa 1946], F3540: 8854, EW-CSA.

81. J.M. Whitley to Earl Warren, 8 Jan. 1947, F3640: 3657, EW-CSA. On the use of constituent letters as a source of information on public opinion, see Lee, Taeku, Mobilizing Public Opinion (Chicago: University of Chicago Press, 2002)Google Scholar.

82. Los Angeles Sentinel, 29 Nov. 1945, 5

83. Los Angeles Times, 8 Jan. 1946, 8.

84. California Senate Journal, 7 Jan. 1946, 60.

85. Los Angeles Times, 16 Feb. 1946, 2.

86. Oakland Tribune, 6 Nov. 1946, 1; State of California Statement of Vote, General Election, 5 Nov. 1946.

87. Chen, The Fifth Freedom.

88. In his keynote address to the GOP national convention in 1944, Warren decried New Dealers' attempts to ensnare workers in regulation that would reduce their opportunity to work where they pleased and prevent business owners from running their firms as they saw fit (New York Times, 25 and 27 June 1944, 12; Los Angeles Times, 27 and 28 June 1944, 2).

89. Records for two counties, Kern and San Mateo, are no longer available.

90. Due to a coding error in some precincts we cannot use the Senate race as a benchmark for the entire state, although it produces nearly identical results to the Lieutenant Governor's race in the subset of correctly coded precincts.

91. For a similar use of bounds, see Sekhon, Jasjeet Singh and Herron, Michael C., “Black Candidates and Black Voters: Assessing the Impact of Candidate Race on Uncounted Vote Rates,” Journal of Politics 67 (2005): 154177Google Scholar.

92. Goodman, Leo, “Ecological Regression and the Behavior of Individuals,” American Sociological Review 18 (1953): 663664CrossRefGoogle Scholar.

93. Goodman, “Ecological Regression.”

94. This regression is based on 14,909 precincts, and the r-squared statistic is .30. Table 2 presents the associated confidence intervals, which are constructed using the simulation-based procedure outlined in Tomz, Michael and Van Houweling, Robert, “How Does Voting Equipment Affect the Racial Gap in Voided Ballots?”, American Journal of Political Science 47 (2003): 46-60CrossRefGoogle Scholar.

95. Achen, Christopher H. and Shively, W. Phillips, Cross-Level Inference (Chicago: University of Chicago Press, 1995)Google Scholar.

96. The heteroskedasticity in the error term ɛ i = e ri + (e di − e ri)T di + v i does not bias the point estimates, though it does affect their variance, an issue we address by computing heteroskedastic-consistent standard errors in our regression models.

97. Achen and Shively, Cross-Level Inference.

98. If instead the estimate of β 2 is negative, implying that b 4 > b 2, we identify the system of equations by setting b 2 = 0, thereby fixing Democratic support for Proposition 11 at a constant level while letting Republican support depend on the partisan composition or precincts.

99. This regression is based on 14,909 precincts, and the r-squared statistic is .31.

100. Tomz and Van Houweling, “How Does Voting Equipment Affect the Racial Gap in Voided Ballots?”.

101. King, Gary, A Solution to the Ecological Inference Problem: Reconstructing Individual Behavior from Aggregate Data (Princeton: Princeton University Press, 1997)Google Scholar.

102. Kenneth Benoit and Gary King, EzI: A(n Easy) Program for Ecological Inference (Version 2.7, 14 Apr 2003).

103. King's method also allows us to let the rates of support for Prop. 11 by one of the partisan groups vary with, T di, the partisan composition of the precincts, in a way analogous to the quadratic model. When we let the rate of one group or the other vary with the composition of precincts, the estimates produced by EI were nearly identical to those produced when we fixed the rates of both groups across precincts.

104. Percent urban and percent nonwhite are drawn from the 1940 census. Value-added manufacturing data are from 1947.

105. Tomz and Van Houweling, “How Does Voting Equipment Affect the Racial Gap in Voided Ballots?”.

106. We adopt this method because it is straightforward and produces the most conservative estimates of the difference between the partisan groups of the three methods we have employed. Results using King's EI are very similar to those we report.

107. We also divided the counties based on their rate of population growth between 1940 and 1950 and detected no significant difference in the behavior of either partisan group depending on the rate of growth in the counties they inhabited.

108. Zaller, John R., The Nature and Origins of Public Opinion (New York: Cambridge University Press, 1992), 13CrossRefGoogle Scholar.

109. Although there is no evidence one way or the other, we do not think that most voters interpreted his silence as outright opposition to Proposition 11. Warren could have easily professed to support the underlying principle of the initiative but oppose it because it contained certain objectionable provisions. That he remained resolutely silent would have simply deepened their puzzlement.

110. If the racial conservatism of Republicans voters does not appear elite-led, neither does the racial liberalism of Democratic elites. Our findings suggest that any Democratic efforts to make inroads among Republicans by adopting a racially conservative position on FEP would have been fraught with substantial electoral danger. It could have cost them the loyalty of racial liberals and thereby risked splitting their heterogeneous electoral base. In the case of Republicans and Democrats, then, it appears plausible that the masses were constraining partisan elites rather than being led by them. This finding is in tension with contemporary research on parties and party competition. Downsian models of party competition highlight elite choice, while historical-institutionalist modes of explanation emphasize the contingency of political outcomes and processes. It may be that the social bases of party competition have gone underappreciated, and undertheorized, in the process. More compelling spatial models of party competition incorporate the agency of activists. See Aldrich, John H., Why Parties? (Chicago: University of Chicago Press, 1995)CrossRefGoogle Scholar. However, the relationship between activists and the social bases of parties is unclear. For a critique of historical-institutionalist scholarship along these lines, see Paul Frymer and Robert W. Mickey, “Missed Opportunities in ‘Missed Opportunities’ Research: Reassessing Structure and Contingency in the New Deal Era,” paper presented at the 2007 APSA Annual Meetings, Chicago.

111. Langdon Post, California Democrats in the Earl Warren Era, an oral history conducted in 1971–1972 by Amelia R. Fry for the Regional Oral History Office, The Bancroft Library, University of California, Berkeley, California, 1976; Tarea Hall Pittman, NAACP Official and Civil Rights Worker, an oral history conducted in 1971–1972, Regional Oral History Office, The Bancroft Library, University of California, Berkeley, 1974; Richard Rodda, Bee Perspectives on the Warren Era, an oral history conduced in 1969–1972 by Amelia R. Fry and June C. Hogan for the Regional Oral History Office, The Bancroft Library, University of California, Berkeley, 1976.

112. On the politics of fair housing legislation, see Duane Lockard, Toward Equal Opportunity; Erikson, “The Relationship between Party Control and Civil Rights Legislation in the American States”; Collins, William J., “The Political Economy of State Fair-Housing Laws Prior to 1968,” Social Science History 30 (2006): 1549CrossRefGoogle Scholar; Anthony S. Chen and Robin Phinney, “Did the Civil Rights Movement Have a Direct Impact on Public Policy? Evidence from the Passage of State Fair Housing Laws,” Working Paper 2004–005, Gerald R. Ford School of Public Policy, University of Michigan.

113. Republican support for keeping the law on the books ranged from 8 percent (high school graduates) to 33 percent (more than 2 years of college), while for Democrats, support for the law ranged from 29 percent (less than high school) to 64 percent (more than 2 years of college). Wolfinger, Raymond E. and Greenstein, Fred I., “The Repeal of Fair Housing in California: An Analysis of Referendum Voting,” American Political Science Review 62 (1968), 760CrossRefGoogle Scholar, Table 8. For an important new look at the politics of fair housing in California, see Andrea Gill, “Upholding the ‘Right’ to Discriminate: The California Real Estate Association and the Fight against Fair Housing Law,” paper presented at the Policy History Conference, St. Louis, MO, 30 May 2008.

114. A pre-election poll in Detroit showed that 82 percent of Republicans planned to vote for the ordinance, compared to 64 percent of Democrats. Hahn, Harlan, “Northern Referenda on Fair Housing: The Response of White Voters,” Western Political Quarterly (1968): 489Google Scholar.

115. For instance, based on a question about state FEP legislation administered in a 1946 Gallup Poll, it would seem that Connecticut, New Jersey, New York, and Rhode Island all exhibit the same high level of public support for state FEP legislation. California's level of public support is noticeably lower, though it is comparable to the level of support in states such as Michigan, Ohio, and Illinois. Authors' calculations from Gallup Organization, Gallup Poll, No. 349 (Storrs, CT: Roper Center, 1945). In the politics of state FEP legislation, California belongs to a group of late-adopting states that included Illinois and Ohio. All three states passed such legislation belatedly because the GOP retained control over veto points through the late-1950s and early-1960s. This suggests that California is not exceptional, but rather typical, of racially conservative states.

116. Berinsky, Adam J., “American Public Opinion in the 1930s and 1940s: The Analysis of Quota-Controlled Sample Survey Data,” Public Opinion Quarterly (2006): 499529CrossRefGoogle Scholar.