Hostname: page-component-5d59c44645-kw98b Total loading time: 0 Render date: 2024-02-28T19:39:10.601Z Has data issue: false hasContentIssue false

Public Opinion, Organized Labor, and the Limits of New Deal Liberalism, 1936–1945

Published online by Cambridge University Press:  04 October 2011

Eric Schickler*
University of California, Berkeley
Devin Caughey
University of California, Berkeley


The seemingly wide opening for liberal domestic policy innovation by the U.S. federal government in the early-to-mid-1930s gave way to a much more limited agenda in the late 1930s and 1940s. The latter years saw the consolidation and gradual extension of several key programs (e.g., Social Security and Keynesian macroeconomic management), but also the frustration of liberal hopes for an expansive “cradle-to-grave” welfare state marked by strong national unions, national health insurance, and full employment policies. Drawing upon rarely used early public opinion polls, we explore the dynamics of public opinion regarding New Deal liberalism during this pivotal era. We argue that a broadly based reaction against labor unions created a difficult backdrop for liberal programmatic advances. We find that this anti-labor reaction was especially virulent in the South but divided even Northern Democrats, thus creating an effective wedge issue for Republicans and their Southern conservative allies. More generally, we find that the mass public favored most of the specific programs created by the New Deal, but was hardly clamoring for major expansions of the national government's role in the late 1930s and 1940s. These findings illuminate the role played by the South in constraining New Deal liberalism while also highlighting the tenuousness of the liberal majority in the North.

Research Article
Copyright © Cambridge University Press 2011

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.)


1. Conservative resistance did have an impact on some specific New Deal proposals, such as the wealth tax and “death sentence” provision of the Public Utilities Holding Company Bill. Much of this resistance came from Northern Democrats not fully on board with Roosevelt. See Patterson, James T., Congressional Conservatism and the New Deal: The Growth of the Conservative Coalition in Congress, 1933–1939 (Lexington: For the Organization of American Historians [by] University of Kentucky Press, 1967)Google Scholar.

2. Throughout the paper, we use “Northern” as a shorthand for “non-Southern.”Google Scholar

3. Alsop, Joseph and Catledge, Turner, The 168 Days (Garden City, NY: Doubleday, 1938)Google Scholar; Dunn, Susan, Roosevelt's Purge: How FDR Fought to Change the Democratic Party (Cambridge, MA: Belknap Press, 2010)Google Scholar; Patterson, , Congressional Conservatism and the New Deal.Google Scholar

4. It is worth noting that there was no single, defined “liberal program” as of the late 1930s. As Alan Brinkley has shown, the meaning of the New Deal was still very much up for grabs even among its strongest liberal supporters. But there is little doubt that the post-1945 outcome provided a less secure place for unions and a less expansive welfare state than many New Deal liberals expected before the war; see Brinkley, Alan, The End of Reform: New Deal Liberalism in Recession and War (New York: Vintage Books, 1995)Google Scholar.

5. Fraser, Steve and Gerstle, Gary, The Rise and Fall of the New Deal Order, 1930–1980 (Princeton, NJ: Princeton University Press, 1989)Google Scholar; Orren, Karen and Skowronek, Stephen, “Regimes and Regime Building in American Government: A Review of the Literature on the 1940s,” Political Science Quarterly 113, no. 4 (1998–1999)Google Scholar.

6. Katznelson, Ira and Lapinski, John S., “Congress and American Political Development: Missed Chances, Rich Possibilities,” Perspectives on Politics 4, no. 2 (2006)CrossRefGoogle Scholar. Although other institutions clearly played a role in elite decision making about New Deal liberalism, a focus on Congress is justified given that it—more than the White House or bureaucracy—constituted the tightest constraint on liberal initiatives starting in 1937.

7. By contrast, in 1936, Democrats won the popular vote in Northern House elections by over four million votes. See Rusk, Jerrold, A Statistical History of the American Electorate (Washington, DC: CQ Press, 2001), 234Google Scholar.

8. Mayhew, David R., America's Congress: Actions in the Public Sphere, James Madison Through Newt Gingrich (New Haven, CT: Yale University Press, 2000)Google Scholar.

9. The relationship between public opinion and the actions of elected representatives should be empirically investigated rather than taken for granted. This is especially true given the highly uneven character of American democracy during this period, particularly in the South, where representation was compromised by severe restrictions on the electorate and on partisan competition. One important but largely unexplored question is the degree to which the restricted form of electoral competition in the South induced its representatives to respond to public opinion (or subsets thereof). See Caughey, Devin, “The Mass Basis of the ‘Southern Imposition’: Labor Unions, Public Opinion, and Representation, 1930s–1940s” (paper presented at the American Political Science Association Annual Meeting, Seattle, WA, Sept. 1–4, 2011), which explores this and related questions.Google Scholar

10. Brinkley, , The End of Reform.Google Scholar

11. Patterson, , Congressional Conservatism and the New DealGoogle Scholar; Katznelson, Ira, Geiger, Kim, and Kryder, Daniel, “Limiting Liberalism: The Southern Veto in Congress, 1933–1950,” Political Science Quarterly 108, no. 2 (1993)Google Scholar.

12. Lichtenstein, Nelson, “Politicized Unions and the New Deal Model: Labor, Business, and Taft-Hartley,” in The New Deal and the Triumph of Liberalism, ed. Milkis, Sidney and Mileur, Jerome (Boston: University of Massachusetts Press, 2002)Google Scholar.

13. Plotke, David, Building a Democratic Political Order: Reshaping American Liberalism in the 1930s and 1940s (Cambridge: Cambridge University Press, 1996)CrossRefGoogle Scholar.

14. Harris, Howell John, The Right to Manage: Industrial Relations Policies of American Business in the 1940s (University of Wisconsin Press: Madison, 1982)Google Scholar; Fones-Wolf, Elizabeth A., Selling Free Enterprise: The Business Assault on Labor and Liberalism, 1945–60 (Urbana: University of Illinois Press, 1994)Google Scholar.

15. Leuchtenburg, William Edward, Franklin D. Roosevelt and the New Deal, 1932–1940 (New York: Harper & Row, 1963), 35, 188–89Google Scholar.

16. Plotke, , Building a Democratic Political Order, 153–6Google Scholar; Dark, Taylor E., The Unions and the Democrats: An Enduring Alliance (Ithaca, NY: Cornell University Press, 1999)Google Scholar.

17. For leftist critiques of the Wagner Act, see Tomlins, Christopher, The State and the Unions (New York: Cambridge University Press, 1985)Google Scholar; Goldfield, Michael, “Worker Insurgency, Radical Organization, and New Deal Labor Legislation,” American Political Science Review 83, no. 4 (1989)Google Scholar; Gordon, Colin, New Deals: Business, Labor, and Politics in America, 1920–1935 (New York: Cambridge University Press, 1994)Google Scholar. For compelling refutations of such arguments, see Farhang, Sean and Katznelson, Ira, “The Southern Imposition: Congress and Labor in the New Deal and Fair Deal,” Studies in American Political Development 19 (2005): 56Google Scholar; Zieger, Robert H., The CIO: 1935–1955 (Chapel Hill: University of North Carolina Press, 1995)Google Scholar.

18. See Lichtenstein, “Politicized Unions and the New Deal Model,” 124. Given the generally pro-labor occupants of the White House, the group most dissatisfied with this arrangement was business. In 1939, when a Roper poll sampled business executives, asking which New Deal program they disliked the most, the resounding answer was the Wagner Act. See Amenta, Edwin, et al. , “Bring Back the WPA: Work, Relief, and the Origins of American Social Policy in Welfare Reform,” Studies in American Political Development 12 (1998)Google Scholar, and Harris, , The Right to Manage: Industrial Relations Policies of American Business in the 1940sGoogle Scholar.

19. Farhang and Katznelson, “The Southern Imposition”; Lichtenstein, Nelson, “From Corporatism to Collective Bargaining,” in The Rise and Fall of the New Deal Order, 1930–1980, ed. Fraser, Steve and Gerstle, Gary (Princeton, NJ: Princeton University Press, 1989)Google Scholar; Lichtenstein, “Politicized Unions and the New Deal Model.”

20. For a more positive assessment of labor's place in the emergent New Deal order, see Plotke, Building a Democratic Political OrderGoogle Scholar. As Orren and Skowronek observe, the literature on the 1940s has had a “glass half full” / “glass half empty” aspect: the 1933 to 1950 period encompassed both major triumphs and disappointments for labor-oriented liberalism. The political terrain had shifted with the Wagner Act and related victories; the battle now was over the meaning of those victories for future politics. In other words, to say that labor's place in the postwar political economy did not match the aspirations of New Deal labor-oriented reformers of the late 1930s does not challenge the notion that the Wagner Act constituted a significant political transformation; Orren and Skowronek, “Regimes and Regime Building in American Government.”

21. See, for example, Atleson, James B., Labor and the Wartime State: Labor Relations and Law During World War II (Urbana: University of Illinois Press, 1998)Google Scholar.

22. See especially Katznelson, , Geiger, , and Kryder, , “Limiting Liberalism”; Farhang and Katznelson, “The Southern Imposition.”Google Scholar

23. Farhang and Katznelson, “The Southern Imposition,” 2. Anthony Badger uses the term “schizophrenic” to describe the Democratic Party in this period; Badger, Anthony J., The New Deal: The Depression Years, reprint of the 1989 ed. (Chicago: Ivan R. Dee, 2002), 271Google Scholar.

24. For an early, innovative analysis of the impact of sectional political economy on congressional decision-making with respect to labor and other policy domains, see Bensel, Richard, Sectionalism and American Political Development: 1880–1980 (Madison: University of Wisconsin Press, 1984)Google Scholar.

25. Lichtenstein, , “Politicized Unions and the New Deal Model.”Google Scholar

26. Brinkley criticizes previous accounts of the New Deal as excessively state-centered and calls for attention to the influence of a “much broader array of social and political forces.” He also claims, however, that the “questions of political economy . . . that preoccupied liberals in the latter years of the New Deal” were relatively insulated from mass-level pressures, existing “largely within the world of elites—intersecting networks of liberal policymakers, journalists, scholars and intellectuals.” Hence he focuses primarily on elite-level debates. See Brinkley, , The End of Reform, 1213Google Scholar.

27. See also Hamby, Alonzo L., Beyond the New Deal: Harry S. Truman and American Liberalism (New York: Columbia University Press, 1973)Google Scholar. For a critique of the claim that New Deal reformers seriously contemplated a fundamental restructuring of the economy, see Brown, Michael, “State Capacity and Political Choice: Interpreting the Failure of the Third New Deal,” Studies in American Political Development 9, no. 1 (1995)Google Scholar.

28. On consolidation, see Plotke, , Building a Democratic Political OrderGoogle Scholar; Hamby, , Beyond the New Deal.Google Scholar For accounts of how the Second World War changed the liberal agenda, see Amenta, Edwin and Skocpol, Theda, “Redefining the New Deal: World War II and the Development of Social Provision in the United States,” in The Politics of Social Policy in the United States, ed. Weir, Margaret, Orloff, Ann Shola, and Skocpol, Theda (Princeton, NJ: Princeton University Press, 1988)Google Scholar; Mayhew, David R., “Wars and American Politics,” Perspectives on Politics 3, no. 3 (2005)Google Scholar.

29. Fones-Wolf, , Selling Free EnterpriseGoogle Scholar; Harris, , The Right to ManageGoogle Scholar; Jacobs, Meg, “‘How About Some Meat?’: The Office of Price Administration, Consumption Politics, and State Building from the Bottom Up, 1941–1946,” Journal of American History 84, no. 3 (1997)Google Scholar; Lichtenstein, “From Corporatism to Collective Bargaining”; Bell, Jonathan, The Liberal State on Trial: The Cold War and American Politics in the Truman Years (New York: Columbia University Press, 2004)Google Scholar.

30. Polenberg, Richard, “The Decline of the New Deal, 1937–1940,” in The New Deal: The National Level, ed. Braeman, John, Bremner, Robert H., and Brody, David (Columbus: Ohio University Press, 1975), 255Google Scholar.

31. See, e.g., Newman, Katherine S. and Jacobs, Elisabeth, “Brothers’ Keepers?,” Social Science and Modern Society 44, no. 5 (2007)Google Scholar; Erskine, Hazel, “The Polls: Government Role in Welfare,” Public Opinion Quarterly 39 (1975)Google Scholar; Schiltz, Michael, “Public Attitudes Toward Social Security, 1935–1965,” (Washington, DC: Social Security Administration, 1970)Google Scholar; Jacobs, “‘How About Some Meat?’”; Ellis, Christopher and Stimson, James A., “Symbolic Ideology in the American Electorate,” Electoral Studies 28 (2009)Google Scholar. Page and Shapiro's comprehensive survey of American public opinion extends back to the mid-1930s, but many of the polls we examine were not available for direct analysis at the time that book was written; see Page, Benjamin I. and Shapiro, Robert Y., The Rational Public: Fifty Years of Trends in Americans’ Policy Preferences (Chicago: University of Chicago, 1992), 4246Google Scholar.

32. See, for example, Cantril, Hadley, “The Issues—As Seen by the American People,” Public Opinion Quarterly 8, no. 3 (1944)Google Scholar; Zieger, The CIO; Brinkley, , The End of Reform, Chapter 9Google Scholar.

33. Berinsky, Adam J. and Schickler, Eric, “The American Mass Public in the 1930s and 1940s [Computer file].” Individual surveys conducted by the Gallup Organization, Roper Organization, NORC, and The Office of Public Opinion Research [producers], 1936–1945: Roper Center for Public Opinion Research, University of Connecticut [distributor] (2011)Google Scholar; and Berinsky, Adam J., et al. , “Revisiting Public Opinion in the 1930s and 1940s,” PS: Political Science and Politics 44(3) (2011): 515520Google Scholar.

34. Eisinger, Robert M. and Brown, Jeremy, “Polling as a Means Toward Presidential Autonomy: Emil Hurja, Hadley Cantril and the Roosevelt Administration,” International Journal of Public Opinion Research 10, no. 3 (1998)Google Scholar; Eisinger, Robert M., The Evolution of Presidential Polling (New York: Cambridge University Press, 2003)Google Scholar. The FDR Presidential Library in Hyde Park, NY, contains a wealth of largely unstudied material devoted to the work of Hurja and Cantril.

35. Converse, Jean M., Survey Research in the United States: Roots and Emergence (Berkeley: University of California Press, 1987)Google Scholar; Igo, Sarah, The Averaged American: Surveys, Citizens, and the Making of a Mass Public (Cambridge, MA: Harvard University Press, 2007)Google Scholar.

36. Igo, , The Averaged American.CrossRefGoogle Scholar

37. Converse, Survey Research in the United States.Google ScholarPubMed

38. Lewis, George F. Jr., “The Congressmen Look at the Polls,” Public Opinion Quarterly 4, no. 2 (1939)Google Scholar.

39. See, for example, Gallup, George, “Public Opinion Polls Important in Democracy,” Washington Post, August 6, 1939Google Scholar.

40. Polling may have had particularly complicated implications for Southern politicians. In a political system that had substantial authoritarian features yet also claimed to be democratic and representative, polls may have posed a challenge for political elites. Polls had the potential to construct a “public” in the South that expressed views distinct from those of the political elites who claimed to speak for the region. A key question for future research is how Southern elites thought about and responded to the onset of opinion polling in the region. We thank Richard Bensel for suggesting this line of inquiry.Google Scholar

41. These works include Baum, Matthew A. and Kernell, Samuel, “Economic Class and Popular Support for Franklin Roosevelt in War and Peace,” Public Opinion Quarterly 65 (2001)Google Scholar; Caldeira, Gregory A., “Public Opinion and The U.S. Supreme Court: FDR's Court-Packing Plan,” American Political Science Review 81, no. 4 (1987)Google Scholar; Newman, and Jacobs, , “Brothers’ Keepers?”; Helmut Norpoth, Victor Lange, and Michael Morzenti, “Voting in Wartime: The 1944 Election” (paper presented at the Midwest Political Science Association Annual Meeting, Chicago, 2009)Google Scholar; Schiltz, “Public Attitudes Toward Social Security”; Schlozman, Kay Lehman and Verba, Sidney, Injury to Insult: Unemployment, Class, and Political Response (Cambridge: Harvard University Press, 1979)Google Scholar; Weatherford, M. Stephen and Sergeyev, Boris, “Thinking about Economic Interests: Class and Recession in the New Deal,” Political Behavior 22 (2000)Google Scholar.

42. For a description of quota-sampling practices, see Berinsky, Adam J., “American Public Opinion in the 1930s and 1940s: The Analysis of Quota-Controlled Sample Survey Data,” Public Opinion Quarterly 70, no. 4 (2006)CrossRefGoogle Scholar.

43. See Roper, Elmo, “Sampling Public Opinion,” Journal of the American Statistical Association 35, no. 210, Part 1 (1940)CrossRefGoogle Scholar.

44. These figures are typical of the polls we have examined through the early 1940s. By the mid-1940s, however, Gallup adjusted his gender quotas to interview equal numbers of men and women. This change in the composition of the sample makes it difficult to track real changes in opinion over time.Google Scholar

45. Race was included as a separate weighting category for those surveys that had at least 20 African-American respondents in the sample (nearly all of the Gallup, Roper, and NORC polls had more than 20 African Americans; several OPOR polls did not). When race was used as a weighting variable for the purposes of creating cell weights, whites were weighted using the full stratification table (gender by region by occupation/education), but, due to small sample sizes, blacks were only weighted on the basis of gender.Google Scholar

46. Raking matches cell counts to the marginal distribution of the variables used in the weighting scheme, but ignores information available in the joint distribution of the weighting variables. We rely on raking weights because we lack cross-tabular information on phone ownership (e.g., phone by gender).Google Scholar

47. Ruggles, Steven et al. , “Integrated Public Use Microdata Series: Version 3.0” (Minneapolis: Minnesota Population Center, 2004).Google Scholar

48. Berinsky, Adam J. and Schickler, Eric, “The American Mass Public in the 1930s and 1940s,” (National Science Foundation, Collaborative Research Grant, 2006)Google Scholar; Berinsky, Adam et al. , “Revisiting Public Opinion in the 1930s and 1940s.”Google Scholar

49. One might argue that the unweighted marginals were more politically relevant, given that politicians would see these numbers and that the samples were intended to approximate the voting public. There is some truth to this view. To the extent that polls had a direct impact on politicians’ views, the unweighted numbers are highly informative. However, observers at the time did note that the polls tended to underestimate Roosevelt's vote share, which suggests that politicians likely also saw the potential for bias in the polls. To the extent that politicians found that the results in the voting booth—or other sources of constituent opinion—conflicted with the polling results, it is not clear that they would accept the poll numbers as decisive. More importantly, leaving aside any direct effects of polls on politicians’ views, the weighted data should give us a more solid basis for making inferences about the underlying contours of public opinion, which formed a backdrop for politicians’ maneuvering (and were a target of that maneuvering). Berinsky, Adam J. and Schickler, Eric, “The American Mass Public in the 1930s and 1940s [Computer file].”Google Scholar

50. In other words, one must assume that conditional on the weighting variables, individuals not included in the sample are “missing at random.”Google Scholar

51. Cantril, , “The Issues,” 83–97Google Scholar.

52. Schickler, Eric and Pearson, Kathryn, “Agenda Control, Majority Party Power, and the House Committee on Rules, 1937–52,” Legislative Studies Quarterly 34, no. 4 (2009)CrossRefGoogle Scholar.

53. The resolution was rejected on the House floor despite the support of nearly all of the Republicans and half of the Southern Democrats.Google Scholar

54. Whereas Roosevelt's veto of Smith–Connally was overridden, Truman's Case bill veto was upheld (but provisions of the bill were incorporated in Taft-Hartley the following year).Google Scholar

55. Schickler, Eric, Pearson, Kathryn, and Feinstein, Brian D., “Congressional Parties and Civil Rights Politics from 1933 to 1972,” Journal of Politics 72 (2010)Google Scholar; Schickler, Eric, “New Deal Liberalism and Racial Liberalism in the Mass Public, 1937–1952” (paper presented at the Annual Meeting of the American Political Science Association, Washington, DC, Sept. 2–5, 2010)Google Scholar.

56. For evidence that economic liberalism, Democratic partisanship, and racial liberalism had become related at the mass level among Northern whites by the early 1940s, see Schickler, “New Deal Liberalism and Racial Liberalism.”Google Scholar

57. On the centrality of organized labor to liberal aspirations, see e.g., Greenstone, J. DavidLabor in American Politics (New York: Knopf, 1969)Google Scholar. The Right to Manage, Harris's important study of business responses to labor union gains, notes the presence of an anti-labor mood in the late 1930s but does not examine opinion data. Harris points to conservative wins in state elections in 1938 that gave rise to the enactment of anti-union legislation in Pennsylvania, Michigan, and Wisconsin (p. 37). Harris also argues that business groups capitalized on the 1945 to 1946 strike wave in attempting to rally public support for Taft-Hartley.

58. The press narrowly beat out bankers as the group most associated with abusing power: 42 percent cited the press, as compared to 38 percent citing bankers and 11 percent citing veterans.Google Scholar

59. These numbers are unweighted due to the lack of individual-level data for this survey. The same survey found 61 percent of respondents wanted Congress to “pass laws to curb labor organizations,” with just 21 percent opposed; “Fortune Quarterly Survey: XII,” Fortune, April 1938. Unless otherwise stated, all the analyses in this paper exclude respondents who answered, “don't know.”Google Scholar

60. Thirty-seven percent of the respondents identified a group.Google Scholar

61. The remaining respondents answered, “don't know.”Google Scholar

62. Patterson, , Congressional Conservatism and the New Deal, 166Google Scholar.

63. In February 1939, the Supreme Court ruled that sit-down strikes were illegal.Google Scholar

64. We used the Proquest Historical Newspapers database, searching for front-page stories mentioning all of the following keywords: “sit down,” “strike or strikers,” and “General Motors.” The Tribune was the first to give heavy coverage to the strikes, but the Post and Times included fifty-three front-page stories in January and another fifty-three in February. The Los Angeles Times was slower to focus on the strikes but included eleven front-page stories on the sit-down strikes in February (after just two in January).Google Scholar

65. As noted above, there were very few poll questions about unions prior to the sit-down strikes. The few exceptions include a July 1936 Roper survey, which asked: “Do you believe all wage earners should belong to a labor union?” According to Roper, 29 percent chose “all workers,” as compared to 8 percent for “most,” 23 percent for “some,” and 25 percent for “none.” We do not have the individual-level data for this survey, so these numbers are unweighted. The crosstabs presented by Roper suggest that the most hostile region was the Plains—just 18 percent favored all or most workers belonging to unions. The friendliest area was the Mountain States (68.5% favored all or most workers belonging to unions). A March 1936 Gallup survey asked “Are you in favor of stronger labor unions?” A small majority (53 percent vs. 47 percent) answered in the affirmative. Again, the lack of individual level data means that we cannot weight this item and, unfortunately, Gallup did not repeat the same question later. A July 1936 Gallup survey asked a question that was repeated later on: “Are you in favor of labor unions?” A healthy 76 percent majority responded in favor, but again this is a few months before the period for which we have individual-level data. Southern respondents mirrored the national distribution on this item. When the question was asked later, there was continued widespread support for unions, but the margin was smaller than in this first poll.Google Scholar

66. Since party identification is not available in most of the early surveys, presidential vote choice is used in classifying Democrats and Republicans. For polls with party identification, the results are similar regardless of which measure is used. The figures provide the percentage giving the liberal response of those with an opinion (that is, with “don't knows” removed from the analysis). An important caveat to remember when interpreting plots over longer periods of time is that partisan groups are not entirely stable. In polls conducted between November 1936 and November 1940, for example, partisanship is defined with reference to FDR's landslide reelection of 1936, whereas during the period 1940 to 1944 the reference point was the much closer 1940 election.Google Scholar

67. Since the overwhelming majority of white Southerners backed Roosevelt, the estimated levels of support are extremely similar if one focuses on Southern FDR voters only rather than all Southern whites.Google Scholar

68. Support for the strikers did not vary much across regions, except for the east central states, which were the least supportive (with a plurality favoring GM). These states—Michigan, Ohio, Indiana, and Illinois—were the ones most directly exposed to the strikes, of course.Google Scholar

69. The unweighted numbers—which are what politicians likely attended to—were even more one-sided: a 48 percent to 29 percent GM advantage.Google Scholar

70. The South and North look similar on this measure.Google Scholar

71. Bell, Daniel, “Industrial Conflict and Public Opinion,” in Industrial Conflict, ed. Kornhauser, Arthur, Dubin, Robert, and Ross, Arthur M. (New York: McGraw-Hill, 1954)Google Scholar. Polls conducted amidst the 1936 election showed that most voters expected the Roosevelt administration to adopt a more moderate course over the next four years, and that a plurality favored such a course.

72. See the analyses of popular responses to the tactics used by the civil rights, anti-war, and women's movement in Page and Shapiro, The Rational Public.Google Scholar

73. Zieger, The CIO.Google Scholar

74. Gallup, George, “Causes of the Swing to the Right Analyzed,” Washington Post, November 14, 1938Google Scholar; Goodman, Walter, The Committee: The Extraordinary Career of the House Committee on Un-American Activities (New York: Farrar, Straus, and Giroux, 1968)Google Scholar; Wreszin, Michael, “The Dies Committee,” in Congress Investigates, ed. Schlesinger, Arthur and Bruns, Roger (New York: Chelsea House, 1975)Google Scholar. After Murphy's defeat, Roosevelt appointed him to be U.S. attorney general and later to the Supreme Court. As attorney general, Murphy was a key figure in the Justice Department's increasingly assertive stance in favor of civil rights; McMahon, Kevin J., Reconsidering Roosevelt on Race: How the Presidency Paved the Road to Brown (Chicago: University of Chicago Press, 2004)Google Scholar.

75. Lazarsfeld and his collaborators’ classic “opinion leadership” studies, for example, were based on panel surveys with dense sampling of particular communities, “which made it possible to locate changes almost as soon as they occurred and then to correlate change with the influences reaching the decision-maker”; Katz, Elihu, “The Two-Step Flow of Communication: An Up-To-Date Report on an Hypothesis,” Public Opinion Quarterly 21, no. 1 (1957): 64Google Scholar. See also Katz, Elihu and Lazarsfeld, Paul F., Personal Influence: The Part Played by People in the Flow of Mass Communications (Glencoe, IL: Free Press, 1955)Google Scholar. Zaller's work on elite cues and opinion flows also relies on panel data or a well-timed series of repeated items, as well as detailed information on respondents’ media consumption and political information; see Zaller, John R., The Nature and Origins of Mass Opinion (New York: Cambridge University Press, 1992)Google Scholar.

76. Of course, the reactions of polling organizations could be considered part of the effect of the investigations themselves.Google Scholar

77. It appears that Gallup went into the field right before the first Dies hearing focused on CIO communism. Unfortunately, the data file does not indicate the date each individual respondent was interviewed. The timing of the survey question's inclusion suggests that Gallup may have known about the upcoming hearing and included the item on union communism in anticipation of news coverage of Dies's work.Google Scholar

78. The “closed shop” refers to the requirement that an employer only hire union members; the “union shop” refers to the slightly looser stipulation that any non-union workers must join the union upon being hired. An “open shop” is one in which there is no requirement that workers join or contribute dues to a labor union.Google Scholar

79. While the figure excludes “don't know” responses, a majority of all respondents—including “don't knows”—opposed the closed and union shops in the surveys.Google Scholar

80. In the first two surveys asking about the closed shop, Southern whites appear less conservative than Northern Republicans. By 1942, however, the two groups appear indistinguishable.Google Scholar

81. Maintenance of membership required that employees who were members of the union at the time a labor agreement was made, or who later joined the union, had to remain members until the agreement expired. While Atleson grants that maintenance of membership offered important benefits to unions, he charges that WLB decision-making processes and criteria gave rise to a more bureaucratic union movement that failed to capitalize on shop-floor militancy; Atleson, Labor and the Wartime State. As Workman argues, however, the WLB was clearly seen by participants at the time as a union ally, and its decisions gave a much-needed boost to unions that were in a vulnerable position; Workman, Andrew, “Creating the National War Labor Board: Franklin Roosevelt and the Politics of State Building in the Early 1940s,” Journal of Policy History 12 (2000)CrossRefGoogle Scholar. See also Lichtenstein, “Politicized Unions and the New Deal Model”; Zieger, The CIO.

82. The question wording was: “Do you favor or oppose ‘maintenance of membership’—that is, requiring a person who joins a union to continue to belong to that union in order to hold his job?”Google Scholar

83. Kornhauser rightly argues that these poll questions were worded in a manner that might have generated greater anti-labor responses (since they did not mention alternative solutions to labor troubles other than an anti-strike law); see Kornhauser, Arthur, “Are Public Opinion Polls Fair to Organized Labor,” Public Opinion Quarterly 10 (Winter 1946–47)CrossRefGoogle Scholar. The absolute percentage opposing strikes or unions in response to any particular question may not be significant, but the general tenor across the range of items—including open-ended questions—clearly indicates the breadth of the popular anger toward unions.

84. Glaberman, Martin, Wartime Strikes: The Struggle Against the No Strike Pledge in the UAW During World War II (Detroit: Bewick Editions, 1980)Google Scholar; Lichtenstein, Nelson, Labor's War at Home: The CIO in World War II (Cambridge: Cambridge University Press, 1987)Google Scholar. The latter work portrays the wildcat strikers and militant shop stewards of the war years as “heroic figures, a vibrant, combative opposition not only to the warfare state, but to management and union bureaucracy alike”; Lichtenstein, Nelson, Labor's War at Home: The CIO in World War II: With a New Introduction by the Author (Philadelphia: Temple University Press, 2003)Google Scholar, xxii. Lichtenstein's later work, however, gives greater weight to the constraints confronting union leaders and to the fragile gains during the war years; see, for example, Lichtenstein, “Politicized Unions and the New Deal Model.”

85. Since the composition of Northern Republicans changed over time—going from a hard core of Landon supporters in 1936 to 1937 to a larger group of Willkie and Dewey voters during the war years—it is possible that if one focused on the same individuals over time, one would see even Northern Republicans moving to the right on labor (despite their conservative starting point). Unfortunately, the polling data are too limited to allow us to assess this possibility directly.Google Scholar

86. Brinkley, , The End of Reform, 225–26.Google Scholar

87. Ibid. Hillman had been one of the founders of the CIO and led the Amalgamated Clothing Workers of America. He was also the first chair of the CIO-PAC (founded in 1942). Hillman was a key link between the Roosevelt administration and the CIO. Lewis had also been a co-founder of the CIO, but withdrew his United Mine Workers from the organization in 1942.Google Scholar

88. Fewer than 1 percent listed Lewis as the “most approved” figure in business or industry. We do not have the individual-level data for this survey, so the results are unweighted.Google Scholar

89. The question read: “Which of these people do you feel have been on the whole helpful to labor and which harmful? . . . John L. Lewis”; Roper Fortune Poll, April 1940.Google Scholar

90. The “negative” count excludes a handful of respondents who said that Lewis is “out to get Roosevelt” (since this could presumably be liberals angry at Lewis for endorsing Willkie in 1940)Google Scholar.

91. Cantril, Hadley and Strunk, Mildred, Public Opinion, 1935–1946 (Princeton, NJ: Princeton University Press, 1951), 561Google Scholar.

92. The same survey asked about a National Association of Manufacturers endorsement. While the response was still negative, it was less overwhelming than in the case of a CIO endorsement (36.5% negative, 21% positive).Google Scholar

93. For an insightful account of Republican Robert Taft's 1950 reelection campaign against a CIO-backed Democratic candidate, see Calkins, Fay, The CIO and the Democratic Party (Chicago: University of Chicago Press, 1952)Google Scholar. Calkins highlights the difficulties faced by Democrats given their reliance on CIO resources and the organization's unpopularity with many Northern swing voters. In the South, the position of the CIO was even more precarious. Daniel Powell, CIO-PAC regional director for the South during this period, reminisced many years later:

From the beginning [liberal Southern] politicians saw PAC as a source of money, campaign workers, and votes. Most of them wanted PAC support, but they wanted it as quietly as possible. Jim Folsom, running for governor of Alabama in 1946, was the only major candidate in the South I can recall who wanted a public endorsement from PAC in that year. In the 1949 Virginia gubernatorial primary, United States Senator Harry Byrd charged the secretly endorsed PAC candidate with having CIO support. Two nights later, our candidate went on statewide radio to deny he had PAC or CIO support. The broadcast was paid for in part by PAC money. Occasionally some unscrupulous candidate, while seeking PAC covert support, would suggest that we publicly endorse his opponent.

See Powell, Daniel A., “PAC to COPE: Thirty-Two Years of Southern Labor in Politics,” in Essays in Southern Labor History: Selected Papers, Southern Labor History Conference, 1976, ed. Fink, Gary M. and Reed, Merle E. (Westport, CT: Greenwood Press, 1976), 247Google Scholar. The more conservative AFL, on the other hand, remained largely “respectable” in the South during this period; Biles, Roger, The South and the New Deal (Lexington: University Press of Kentucky, 1994), 101Google Scholar.

94. A mere 33 percent of African Americans, by contrast, reported that they had become less favorable toward unions during the war years; 36 percent reported their attitude becoming more favorable.Google Scholar

95. When one breaks the data down by occupation, manual workers were more favorable to unions than were other groups. But even among laborers in the North, a slim plurality reported having a less favorable attitude toward unions during the war (38%, as compared to 32% who had a more favorable attitude).Google Scholar

96. Indeed, while it is tempting to blame the war for the fading of liberal hopes, one could argue that the impending war was likely a major reason Democrats held the White House in 1940. It would have been more difficult for Roosevelt to justify a third-term campaign in the absence of the conflict, and Roosevelt had not left his party with a compelling successor. The polling data also suggest that increased concerns about the European conflict contributed to Roosevelt's support. For an analysis suggesting that Republican Thomas Dewey would have likely won in 1944 if not for the war, see Norpoth, Lange, and Morzenti, “Voting in Wartime.”Google Scholar

97. Zieger, The CIO.Google Scholar

98. The public dissatisfaction was reflected in letters to members of Congress as well as in the polls. McNaughton, Frank, Time magazine's Congress correspondent during the war, included several reports about the onslaught of anti-strike and anti-labor letters received by members from all sections of the countryGoogle Scholar. See Frank McNaughton Papers, Harry S. Truman Library, February 27 and March 20, 1942.

99. Workman, “Creating the National War Labor Board.”Google Scholar

100. Riddick, Floyd M., “The First Session of the Seventy-Eighth Congress, January 6–December 21, 1943,” American Political Science Review 38, no. 2 (1944)Google Scholar. See also McNaughton Papers, February 2, 1942; March 20, 1942; January 1, 1946.

101. Roosevelt's move sparked an angry congressional response, including an investigation that Democratic leaders unsuccessfully sought to block; Schickler and Pearson, “Agenda Control.”Google Scholar

102. Truman had proposed a more balanced anti-strike bill—which also was opposed by organized labor, but was not as draconian as the Case bill. Polls indicated that the public did approve of the Case bill veto, suggesting that they likely favored Truman's more moderate proposal.Google Scholar

103. Schickler, and Pearson, , “Agenda Control.”Google Scholar

104. We used the Proquest Historical Newspapers database, searching for front-page stories mentioning all of the following keywords: (“investigation” or “inquiry” or “investigate”) and (“house” or “senate” or “congress” or “congressional”) and (“union” or “CIO” or “C.I.O.” or “National Labor Relations”) and (“strike” or “sit-down” or “communis*”) and (“hearing” or “oversight” or “committee investig*”).Google Scholar

105. An important caveat is that this electorate largely excluded Southern African Americans, as did much of the earliest Gallup polling data. See the discussion below on African American opinion.Google Scholar

106. Smith-Connally did restrict political contributions by unions, but the formation of the CIO-PAC helped labor maneuver around these restrictions.Google Scholar

107. A case in point is an October 1941 Gallup poll—taken amid efforts by Roosevelt and Democratic leaders to bury an anti-strike bill pushed by the conservative coalition—in which 59 percent of respondents stated that they wanted unions to have less power “than at present,” as compared to 30 percent who favored the same amount of power and 11 percent who wanted unions to have more power. Among Roosevelt voters in the North, 49 percent wanted less union power, while 37.5 percent favored the same amount of union power and 13.5 percent favored greater union influence.Google Scholar

108. Brinkley, , The End of Reform.Google Scholar

109. Commenting on this problem in 1943, The Nation noted that business interests had fanned public distrust of labor unions during the war mobilization, concluding that “should the N.A.M. [National Association of Manufacturers] succeed in keeping public opinion in its present temper, it will utilize the transition from war to peace-time production . . . as the occasion for delivering the long-planned knock-out of the labor movement”; Hochman, Julius, “Let's Look at Labor: The Opportunity for Leadership,” The Nation, September 11, 1943, 291Google Scholar.

110. We code issues concerning minimum wages and maximum hours as non-labor union issues. These policies affected workers as a whole, rather than being targeted at unions per se. Separating these policies out does not affect the trends in Figure 4. We exclude civil rights policy questions from this analysis, but see Schickler, “New Deal Liberalism and Racial Liberalism” on civil rights opinion. Schickler's analysis of the limited survey items on civil rights suggests, as one might expect, that African Americans and Southern whites took sharply divergent positions. Perhaps more surprisingly, Northern Democrats emerge as more supportive of civil rights than their Republican counterparts by the late 1930s and civil rights liberalism and economic liberalism appear to become linked together in the North during the same period.Google Scholar

111. Even so, the mix of questions being asked and the aggregate responses to them tell us something about the larger political context. As the nation mobilized for and began fighting the Second World War, pollsters tended to inquire about wartime strikes and other topics on which very few respondents supported the pro-labor position. Although the question changes make it hazardous to interpret the temporal trends in absolute support of labor, they also reflect real changes in the salient political considerations that informed citizens’ evaluation of labor policy. War mobilization increased the salience of considerations unfavorable to unions, which may help explain why support for pro-labor policies (e.g., the closed shop) decreased sharply.Google Scholar

112. The figures for African Americans include residents of both regions when available, but many surveys contained few if any Southern blacks.Google Scholar

113. One might ask how Southern whites as a whole compare to FDR voters in the region. Across all of the labor items, Southern white FDR voters on average look virtually identical to Southern whites as a whole. On non-labor policies, Southern white FDR voters are a bit more liberal than Southern whites as a whole at the start of the period (about 4 points, on average), but the gap narrows to about 1 point from 1943 to 1945.Google Scholar

114. There are several possible (and not mutually exclusive) explanations for the increasing moderation of Republican voters. One is that Republicans gradually accommodated themselves to portions of the New Deal once they were enacted. Another is that the party's ranks were swelled by moderates who voted for FDR in his 1936 landslide but voted Republican in the 1940 election. A third explanation is that after 1937 the appetite for liberal initiatives diminished among all voters, including Democrats, causing Republicans to look less conservative by comparison.Google Scholar

115. See, for example, Frymer, Paul, Black and Blue: African Americans, the Labor Movement, and the Decline of the Democratic Party (Princeton: Princeton University Press, 2008)Google Scholar; Kersch, Ken I., “The New Deal Triumph as the End of History? The Judicial Negotiation of Labor Rights and Civil Rights,” in The Supreme Court and American Political Development, ed. Kahn, Ronald and Kersch, Ken I. (Lawrence: University Press of Kansas, 2006)Google Scholar. White Democratic union members tended to be relatively supportive of civil rights initiatives, such as fair employment practices legislation, during this period; Schickler, “New Deal Liberalism and Racial Liberalism.”

116. Even when Southern African Americans were included in surveys, one cannot necessarily take their stated opinions at face value, particularly if white interviewers conducted the survey. For example, a 1942 NORC study demonstrated that African Americans in Memphis gave less liberal responses, particularly on racial policy questions, when they were interviewed by whites rather than by African Americans. Interestingly, the race of the interviewer made little difference when African Americans were interviewed in New York City; Hyman, Herbert H. et al., Interviewing in Social Research (Chicago: University of Chicago Press, 1954)Google Scholar.

117. The online appendix, available at, includes these figures. One item on which there is a large gap is on preference for the AFL versus the CIO. Southern African Americans were far more likely to prefer the CIO than were Northern African Americans. This should not be surprising given that the AFL was highly racially exclusionary in the South while the CIO was a lonely voice for racial inclusion in the region.Google Scholar

118. African Americans constituted 3.7 percent of the adult population in the North in 1940. It is worth noting that in our weighted estimates for national opinion using Gallup data presented above, we essentially allow Northern African Americans to “stand in” for Southern African Americans since we weight African Americans to their share of the national population and the vast majority of the African American respondents in the Gallup surveys lived in the North (while approximately 70% of the African American population, as of 1940, was Southern).Google Scholar

119. See Topping, Simon, Lincoln's Lost Legacy: The Republican Party and the African American Vote, 1928–1952 (Gainesville: University of Florida Press, 2008)CrossRefGoogle Scholar on the GOP's uneven efforts to win back African American voters.

120. More generally, we find that when Democratic partisanship was cued by an explicit reference to Roosevelt or the New Deal, Southern whites tended to be more liberal than when such cues were not included.Google Scholar

121. When respondents were instead asked in July 1938 whether Roosevelt should become more conservative or “continue along present lines,” the margin for a conservative turn was smaller but still substantial: 60 percent vs. 40 percent nationally. Once again, Southern whites were less conservative on this indicator (53% vs. 47% in favor of a conservative shift), while Northerners were in favor of a more conservative course by a sizeable 63 percent vs. 37 percent.Google Scholar

122. The December 1945 survey used a split sample. Half were offered the “halfway between” option and half were not offered that response option.Google Scholar

123. Ellis, and Stimson, , “Symbolic Ideology in the American Electorate.”Google Scholar

124. The wording of the questions remains similar but not identical over time. In the earliest polls (1936–37) the most common wording is, “Do you favor government ownership of the railroads?” The polls at the end of the period (1945) tend to use the formulation, “Do you think the government should own the following things in this country? . . . Railroads.” Several other variations were also used, but unfortunately they tended to be used only in specific years, making it more difficult to establish continuity over time. We believe it is unlikely, however, that wording alone accounts for the response trends. See online appendix for exact question wording.Google Scholar

125. The convergence in support for public ownership across the three sectors is also striking. Evidently, in the height of the Depression, power companies and banks—which had, after all, recently been in deep trouble—were seen as more legitimate targets for government ownership than were railroads. But by 1945, the public was equally unenthusiastic about government ownership in each of these areas.Google Scholar

126. Gallup began polling in Britain and France in the late 1930s. While the samples in those countries are presumably affected by many of the same problems as Gallup's American samples, a brief comparison of attitudes in those countries with the U.S. opinion data is instructive. For example, as of 1945, while American opposed nationalizing the railroads 63 percent to 20 percent, British respondents asked a similar question by Gallup in the same year favored nationalization, 54 percent to 26 percent. When British respondents were asked about public control of the coal mines in 1944, 60 percent favored public control, with 16 percent opposed and 24 percent unsure; Gallup, George H., ed. The Gallup International Public Opinion Polls, Great Britain, 1937–1975 (New York: Random House, 1976)Google Scholar, 103, 22, 88–89. Similarly, when French respondents were asked about nationalization of the mines in 1944, 60 percent favored the measure and 19 percent opposed it, with the remainder undecided. This was twice the level of American support for nationalization recorded in a similarly worded item asked by Gallup in 1945. A 1945 French Gallup survey asked about nationalizing the banks and found 70 percent of respondents in favor. Support for nationalization remained substantial in 1946, though lower than the 1945 level; Gallup, George H., ed. The Gallup International Public Opinion Polls, France, 1939, 1944–1975 (New York: Random House, 1976)Google Scholar, 11, 21, 44. By contrast, support for nationalizing the banks was under 30 percent in the U.S. in a 1945 Gallup survey.

127. This accords with the arguments of Schiltz, “Public Attitudes Toward Social Security,” and Campbell, Andrea Louise, How Policies Make Citizens: Senior Political Activism and the American Welfare State (Princeton, NJ: Princeton University Press, 2003)CrossRefGoogle Scholar.

128. The July 1938 item was: “Do you approve of the present social security laws which provide old age pensions and unemployment insurance?” In August 1938, 92 percent responded favorably when asked: “Do you believe in government old age pensions?”Google Scholar

129. A majority of voters in both North and South also favored increasing the amount spent on Social Security, with support similar in both regions.Google Scholar

130. The question wording was as follows: “At present some groups are not included under Social Security. Do you think the Social Security Program should be changed to include the following groups? . . .”Google Scholar

131. Quadagno, Jill, The Color of Welfare (New York: Columbia University Press, 1994)Google Scholar; Lieberman, Robert C., Shifting the Color Line: Race and the American Welfare State (Cambridge, MA: Harvard University Press, 1998)Google Scholar; though see Davies, Gareth and Derthick, Martha, “Race and Social Welfare Policy: The Social Security Act of 1935,” Political Science Quarterly 112, no. 2 (1997)Google Scholar, for a non-racial explanation for occupational exclusions.

132. For a detailed analysis of attitudes toward Social Security and other social welfare programs, see Schiltz, “Public Attitudes Toward Social Security.” Schiltz analyzes support across a range of demographic categories, but does not focus on partisan divisions.Google Scholar

133. Interestingly, attitudes toward farmers and farm assistance remained quite positive during the war. A promising line of research would be to compare wartime attitudes toward unions with attitudes toward farmers. Wartime price-control policies often placed the interests of labor and farmers in direct opposition to one another, greatly undermining the chances for a “green–red” coalitional alignment. The administration viewed demands for higher farm prices as at least as great—or perhaps an even greater—threat to price controls as were wage demands from workers. Yet Congress proved far more receptive to the farmers’ claims. The rural bias in House and Senate districts no doubt helped the farmers, but it is worth exploring how the mass public viewed demands from farmers, as compared to labor. It may be that the mythology of the yeoman farmer shielded agricultural interests from the same backlash faced by labor.Google Scholar

134. Porter, David L., Congress and the Waning of the New Deal (Port Washington, NY: Kennikat Press, 1980)Google Scholar; Schickler and Pearson, “Agenda Control.”

135. We used the Proquest Historical Newspapers database, searching for front-page stories from January 1, 1939, through December 31, 1940, mentioning all of the following keywords: “Woodrum” and (“investigation” or “inquiry” or “investigate”) and (“relief” or “WPA” or “W.P.A.”) and (“house” or “senate” or “congress” or “congressional”). This yielded thirty-three front-page New York Times stories. A few of the stories concerned an earlier Senate committee investigation of the WPA's alleged promotion of New Deal candidates in the 1938 election. These disclosures generated pressure for the Hatch Act's passage. Woodrum—dubbed the “Lord High Executioner” of the WPA in the wake of his attack on a Writers’ Project production of the Mikado—played a leading role in most of the coverage. Dorris, Henry N., “House Votes WPA $100,000,000 More; Bars Higher Grant,” The New York Times, April 1, 1939Google Scholar.

136. Another 9 percent responded, “it depends.”Google Scholar

137. This heightened Southern opposition to relief spending may have reflected intensifying wartime concern about disrupting the region's labor market by undermining the dependence of low-wage laborers on employers.Google Scholar

138. This echoes the conclusion in Newman and Jacobs, “Brothers’ Keepers?” Newman and Jacobs argue that “the public supported the constraints that legislators put on benefit programs, and might have—left to their own devices—been even harsher” in terms of restrictive and exclusionary provisions (p. 8).Google Scholar

139. Free, Lloyd A. and Cantril, Hadley, The Political Beliefs of Americans: A Study of Public Opinion (New York: Simon and Schuster, 1968)Google Scholar; Stimson, James A., Tides of Consent (Cambridge: Cambridge University Press, 2004)Google Scholar.

140. There are a handful of survey questions concerning national health insurance during this period. These items, which are included in Figure 4a, indicate widespread support for the government providing access to medical care to those who cannot afford it. The surveys also indicate support for expanding Social Security to include some form of medical insurance. However, when respondents are offered the choice between medical care provided through a plan set up by the government and care provided through private insurers, the mass public is closely divided, and seemingly small question wording differences give rise to a substantially different tenor of results. See, for example, the July 1945 Gallup survey, which included a split-sample design asking two different health insurance questions; Schiltz, “Public Attitudes Toward Social Security,” 131–33. The broad sentiment in favor of expanded access likely ran up against the abstract concerns about spending and the government's role noted above (as reflected in the decreased support for government ownership of key industries and rising concerns about communism and socialism), creating a situation in which the framing of the issue would likely have a big impact on the shape of public opinion. There is far more polling regarding health insurance after 1945; a sustained examination could help illuminate the extent to which the AMA's success in Congress was also reflected in battles over public opinion. For a preliminary discussion, see Robert W. Mickey and Eric Schickler, “Battles over National Health Insurance during the New and Fair Deals and Their Legacies,” unpublished manuscript (2008). A systematic comparison of health politics in the United States and Britain is provided by Jacobs, Lawrence R., The Health of Nations: Public Opinion and the Making of American and British Health Policy (Ithaca, NY: Cornell University Press, 1993)Google Scholar. Jacobs finds widespread public support for direct governmental provision of health care in Britain by the early 1940s. While Jacobs's analysis of American public opinion data focuses on the late 1950s and 1960s (leading up to the passage of Medicare), he argues that there was far greater ambivalence in the United States with regard to national health insurance in the 1940s than was the case in Britain. We suspect that detailed analysis of the available opinion data on national health insurance from the 1940s would confirm Jacobs's argument.

141. An important caveat is that Southern blacks are so underrepresented in the samples that in most cases we can say little about them.Google Scholar

142. For accounts of business efforts to shape public opinion, see Fones-Wolf, , Selling Free EnterpriseGoogle Scholar; Jacobs, “‘How About Some Meat?’”; Tedlow, Richard S., “The National Association of Manufacturers and Public Relations during the New Deal,” Business History Review 50, no. 1 (1976)Google Scholar.

143. For a general discussion of Congress members’ actions in the “public sphere,” see Mayhew, , America's Congress.Google Scholar

144. Schickler, and Pearson, , “Agenda Control.”Google Scholar

145. See, for example, Bell, , The Liberal State on TrialCrossRefGoogle Scholar; Jacobs, “‘How About Some Meat?’”.

146. Plotke, , Building a Democratic Political OrderCrossRefGoogle Scholar.

147. Bell, , The Liberal State on Trial, 2526Google Scholar. Some scholars have asserted that in the 1930s and 1940s, “Anti-New Deal sentiment could gain only limited traction when expressed in opposition to labor, universalistic social programs, or the development of a national security state, all of which enjoyed wide support”; Lowndes, Joseph E., From the New Deal to the New Right: Race and the Southern Origins of Modern Conservatism (New Haven, CT: Yale University Press, 2008)Google Scholar, 12. Our opinion data, along with Republicans’ eagerness to exploit anti-union sentiment in campaigns, introduces a major caveat to such claims.

148. The correspondence between trends at the mass level and in Congress provides suggestive evidence that despite the South's restrictive one-party regime, Southern representatives were sensitive to changes in the preferences of (at least some of) their white constituents. This responsiveness—termed “dynamic representation” by Stimson, James A., Mackuen, Michael B., and Erikson, Robert S., “Dynamic Representation,” American Political Science Review 89, no. 3 (1995)Google Scholar—suggests that the characterization of the American South as an “authoritarian” regime requires refinement; for such characterizations, see Farhang and Katznelson, “The Southern Imposition”; Gibson, Edward L., “Boundary Control: Subnational Authoritarianism in Democratic Countries,” World Politics 58 (2005)Google Scholar; Mickey, Robert W., “The Beginning of the End for Authoritarian Rule in America: Smith v. Allwright and the Abolition of the White Primary in the Deep South, 1944–1948,” Studies in American Political Development 22, no. 2 (2008)Google Scholar. For further examination of this theme, see Caughey, “The Mass Basis of the ‘Southern Imposition’.”

149. Farhang, and Katznelson, , “The Southern Imposition,” 1.Google Scholar

150. Given the hostility toward labor unions in rural farm areas in both the North and the South, one might ask whether an agricultural political economy is not a better explanation for the South's opposition to pro-labor policies than its fear of many unions’ racial liberalism. A systematic comparison of rural–urban cleavages in public opinion is beyond the scope of this paper, but it is worth noting that the disaffection of rural representatives with an increasingly urban-oriented New Deal is a prominent theme in Mayhew's seminal analysis of party loyalty in Congress as well as in Patterson's work; Mayhew, David R., Party Loyalty Among Congressmen: The Difference Between Democrats and Republicans, 1947–1962 (Cambridge, MA: Harvard University Press, 1966)CrossRefGoogle Scholar; Patterson, Congressional Conservatism and the New Deal. Of course, there were also fundamental political differences between farm areas in and out of the South. This perspective does suggest, however, that the most anomalous feature of rural Southerners’ position with regard to labor may have been their unusually high level of support for measures like the Wagner Act, which was largely the product of their loyalty to Roosevelt and the Democratic Party, rather than their subsequent opposition.

151. Cf. Brinkley, , The End of Reform.Google Scholar

152. Stimson, James A., Public Opinion in America: Moods, Cycles, and Swings, 2nd ed. (Boulder, Colo.: Westview Press, 1999)Google Scholar. See also Erikson, Robert S., MacKuen, Michael B., and Stimson, James A., The Macro Polity (New York: Cambridge University Press, 2008)Google Scholar; Wlezien, Christopher, “The Public as Thermostat: Dynamics of Preferences for Spending,” American Journal of Political Science 39, no. 4 (1995)Google Scholar.

153. Ellis, and Stimson, , “Symbolic Ideology in the American Electorate,” 398.Google Scholar

154. While rank-and-file workers clearly had positive symbolic meaning, unions—along with union leaders and even union activists—were viewed far more skeptically.Google Scholar

155. Campbell, , How Policies Make CitizensCrossRefGoogle Scholar.

156. Relief policies proved more controversial with the public, but distrust of the WPA did not translate into a more general drive to scale back most of the New Deal welfare state. Given the longstanding tendency for the unemployed to be unorganized, it may not be surprising that relief workers failed to become a strong political constituency; Verba, Sidney and Schlozman, Kay Lehman, “Unemployment, Class Consciousness, and Radical Politics: What Didn't Happen in the Thirties,” Journal of Politics 39 (1977)CrossRefGoogle Scholar.

157. Bell, “Industrial Conflict and Public Opinion.”Google Scholar

158. Vogel, David, “The Power of Business in America: A Re-Appraisal,” British Journal of Political Science 13 (1983)CrossRefGoogle Scholar.

159. Our findings thus may also support elements of the Hartzian interpretation of American political culture: Republicans’ ability to campaign against government excesses and labor radicalism so soon after the onset of a calamitous Depression indicates the tenacity of traditional limited-government Lockean liberalism even amidst widespread acceptance of specific government interventions in the economy; Hartz, Louis, The Liberal Tradition in America (New York: Harcourt-Brace, 1955)Google Scholar.