Published online by Cambridge University Press: 14 July 2011
Why has the association between class and party declined over time? Contrary to conventional wisdom that emphasizes the fracturing of social structures and blurring of class boundaries in post-industrial society, it is argued here that class divisions in party preferences are conditioned by the changing shape of the class structure and the effect of parties’ strategic ideological responses to this transformation on the choices facing voters. This thesis is tested using British survey data from 1959 to 2006. We demonstrate that increasing class heterogeneity does not account for the decline of the class–party association, which occurs primarily as a result of ideological convergence between the main parties resulting from New Labour's shift to the centre.
2 Heath, Anthony, Jowell, Roger and Curtice, John, How Britain Votes (Oxford: Pergamon Press, 1985)Google Scholar.
3 Evans, Geoffrey, Heath, Anthony and Payne, Clive, ‘Class and Party Revisited: A New Model for Estimating Changes in Levels of Class Voting’, Journal of Elections, Public Opinion & Parties, 6 (1996), 157–174Google Scholar; Goldthorpe, John H., ‘Modelling the Pattern of Class Voting in British Elections, 1964–92’, in Geoffrey Evans, ed., The End of Class Politics? Class Voting in Comparative Context (Oxford: Oxford University Press, 1999), pp. 59–83CrossRefGoogle Scholar.
4 Evans, Geoffrey, Heath, Anthony and Payne, Clive, ‘Class: Labour as a Catch-All Party?’ in Geoffrey Evans and Pippa Norris, eds, Critical Elections: British Parties and Voters in Long-term Perspective (London: Sage 1999), pp. 87–101Google Scholar.
5 For influential statements of these various influences on the decline of class divisions, see: Martin Lipset, Seymour. Political Man: The Social Bases of Politics (London: Heinemann, 1981), p. 521Google Scholar; Inglehart, Ronald and Rabier, J-R., ‘Political Realignment in Advanced Industrial Society: From Class-Based Politics to Quality-of-Life Politics’, Government & Opposition, 21 (1986), 457–479CrossRefGoogle Scholar; Nichols Clark, Terry and Martin Lipset, Seymour, ‘Are Social Classes Dying?’ International Sociology, 6 (1991), 397–410CrossRefGoogle Scholar; Kitschelt, Herbert, The Transformation of European Social Democracy (Cambridge: Cambridge University Press, 1994)CrossRefGoogle Scholar; Dunleavy, Patrick and Husbands, Christopher. T., British Democracy at the Crossroads (London: Allen & Unwin, 1985)Google Scholar; Saunders, Peter, A Nation of Home Owners (London: Unwin Hyman, 1990)Google Scholar; Franklin, Mark N., The Decline of Class Voting in Britain: Changes in the Basis of Electoral Choice, 1964–1983 (Oxford: Clarendon Press, 1985)Google Scholar; Franklin, The Decline of Class Voting in Britain; Rose, Richard and McAllister, Ian, Voters Begin to Choose: From Closed-Class to Open Elections in Britain (London: Sage, 1986)Google Scholar.
6 Beck, Ulrich and Beck-Gernsheim, E, Individualization: Institutionalized Individualism and Its Social and Political Consequences (London: Sage, 2001)Google Scholar.
7 Dalton, Russell, Citizen Politics: Public Opinion and Political Parties in Advanced Industrial Democracies, 5th edn (Chatham, N.J.: Chatham House, 2008), p. 156Google Scholar.
10 Butler, David and Kavanagh, Dennis, The British General Election of 1983 (Basingstoke, Hants.: Macmillan, 1983), p. 8Google Scholar.
13 Przeworski, Adam and Sprague, John, Paper Stones: A History of Electoral Socialism (Chicago: University of Chicago Press, 1986), p. 179Google Scholar.
15 Mair, Peter, Muller, Wolfgang and Plasser, Fritz, ‘Introduction: Electoral Challenges and Party Responses’, in Peter Mair, Wolfgang Muller and Fritz Plasser, eds, Political Parties and Electoral Change (London: Sage, 2004), p. 4Google Scholar.
16 Mair, Muller and Plasser, ‘Introduction’, p. 2.
17 See Webb, Paul, ‘Party Responses to the Changing Electoral Markets in Britain’, in Mair, Muller and Plasser, eds, Political Parties and Electoral Change, pp. 20–48Google Scholar; and Jane Green, ‘A Test of Core Vote Theories: The British Conservatives, 1997–2005’, British Journal of Political Science, forthcoming.
18 Budge, Ian, ‘Party Policy and Ideology: Reversing the 1950s’, in Geoffrey Evans and Pippa Norris, eds, Critical Elections (London: Routledge, 1999)Google Scholar; Bara, Judith, ‘The 2005 Manifestos: A Sense of Déja Vu?’ Journal of Elections, Public Opinion and Parties, 16 (2006), 265–281CrossRefGoogle Scholar.
20 Kirchheimer, Otto, ‘The Transformation of the Western European Party Systems’, in Joseph LaPalombara and Myron Weiner, eds, Political Parties and Political Development (Princeton, N.J.: Princeton University Press, 1966), pp. 177–200Google Scholar.
23 Converse, Phillip E., ‘The Shifting Role of Class in Political Attitudes and Behaviour’, in E. E. Maccoby, T. M. Newcomb and E. L. Hartley, eds, Readings in Social Psychology, 3rd edn (New York: Holt, 1958), pp. 388–399Google Scholar.
24 Rose and McAllister, Voters Begin to Choose; Franklin, The Decline of Class Voting in Britain.
25 Goldthorpe, John H., Lockwood, David, Bechhoffer, Frank and Platt, Jennifer, The Affluent Worker: Political Attitudes and Behaviour (Cambridge: Cambridge University Press, 1968)Google Scholar.
26 See Evans, The End of Class Politics, for an overview.
28 Crewe, Ivor, ‘On the Death and Resurrection of Class Voting: Some Comments on How Britain Votes’, Political Studies, 34 (1986), 620–638CrossRefGoogle Scholar; Heath, Anthony, Jowell, Roger and Curtice, John, ‘Trendless Fluctuation: A Reply to Crewe’, Political Studies, 35 (1987), 259–277CrossRefGoogle Scholar; Dunleavy, Patrick, ‘Class Dealignment in Britain Revisited’, West European Politics, 10 (1987), 400–419CrossRefGoogle Scholar.
29 Most prominently, Franklin, Mark N., Mackie, Tom, Valen, Henry et al. , Electoral Change: Responses to Evolving Social and Attitudinal Structures in Western Countries (Cambridge: Cambridge University Press, 1992)Google Scholar.
30 As Clark notes in a review of class-voting research: ‘most social phenomena are complex, related to multiple causes and consequences, and seldom explained by any single factor … Yet in key papers the many potential causes of voting are ignored, and only a few factors are analysed statistically’. See Nichols Clark, Terry, ‘What Have We Learned in a Decade on Class and Party Politics?’ in Terry Nichols Clark and Seymour Martin Lipset, eds, The Breakdown of Class Politics: A Debate on Post-Industrial Stratification (Boulder, Co.: Westview, 2001), pp. 6–39Google Scholar.
31 See, for example, Evans, Heath and Payne, ‘Modelling the Class/Party Relationship 1964–87’.
32 Evans, Heath and Payne, ‘Class: Labour as a Catch-All Party?’ Oskarsson, Maria, ‘Social Structure and Party Choice’, in Jacques Thomassen, ed., The European Voter (Oxford: Oxford University Press, 2005), pp. 84–105CrossRefGoogle Scholar; Elff, Martin, ‘Social Divisions, Party Positions, and Electoral Behaviour’, Electoral Studies, 28 (2009), 297–308CrossRefGoogle Scholar.
34 The only exception is our measure for 1959, which is a combination of recalled vote choice and a follow-up question for people who did not vote or were unable to recall how they voted, about which party they would have chosen in 1959 if they had voted. This gives a ‘non-partisan’ rate of 7.9 per cent, very similar to the rates for the other three surveys in the 1960s.
35 For this we were able to use only the BES 1963–1997 surveys, two BSA surveys from 1995 and 1996 that ask about vote intention, and two BSA surveys from 2001 and 2005 that ask about vote choice in those election years. These tables are available on request from the authors.
36 In later years in Scotland and Wales respondents are also given the option of Scottish National Party (SNP) or Plaid Cymru, respectively.
37 These two measures also give very similar proportions of party support and non-identification for the five years that BES and BSA overlap (1983, 1987, 1997, 2001 and 2005), with no differences in support for any party (or no party) greater than 3 per cent.
38 Occupational class is derived from the 19-category socio-economic group (SEG). Members of the armed forces are counted as being in the lower service class (officers), foremen and technician class (NCOs) or skilled manual class (others). A small number of respondents did not have an SEG classification, but were able to be classified due to information on their employment status and managerial status.
39 It should also be noted that these two groups are essentially indistinguishable from one another in terms of party choice.
40 For example, it is well documented that own-account (petty bourgeois) manual workers are much less likely to vote Labour than other manual workers. An increase in the relative number of own-account workers within the manual ‘class’ would be expected to reduce Labour support within the class of manual workers as a whole. However, this would not mean that there had been a change in the tendency of working-class people to support Labour – only that there had been a relative increase in the size of a category of manual workers who have consistently been less supportive of Labour. The Goldthorpe schema also benefits from being the only measure of class position to have been extensively validated (see, for example: Evans, Geoffrey, ‘Testing the Validity of the Goldthorpe Class Schema’, European Sociological Review, 8 (1992), 211–232CrossRefGoogle Scholar; Evans, Geoffrey and Mills, Colin, ‘Identifying Class Structure: A Latent Class Analysis of the Criterion-Related and Construct Validity of the Goldthorpe Class Schema’, European Sociological Review, 14 (1998), 87–106CrossRefGoogle Scholar).
41 The income figures generally refer to household income for any respondent, although in a few of the earlier surveys it is the income of the head of household. As the question on income in all the BESs and BSAs is categorical, we are not precisely placing respondents on an income scale, but rather fitting nearest categories into each of the quintiles and two top deciles. Finally, we should note that income quintiles are referenced to the individual survey sample, so the bottom quintile refers to the bottom income quintile in the survey, not necessarily in the population. In practice, since these are close to representative surveys, this does not give large discrepancies.
42 The combination category of ‘non-Christian’ and ‘no religion’ is due to the very large overlap between non-Christian and non-white respondents, and the very small number of non-Christian respondents in early surveys.
43 Analyses not shown here but available from the authors include public/private sector location. Unfortunately, this question was not asked of respondents in the earlier BES studies, so the time span covered is somewhat truncated. In the surveys for which they were available, their inclusion had no substantive impact on changes in the class–party association.
44 Budge, Ian, Klingemann, Hans-Dieter, Volkens, Andrea and Bara, Judith, Mapping Policy Preferences: Estimates for Parties, Governments and Electors 1945–1998 (Oxford: Oxford University Press, 2001)Google Scholar; Hans-Klingemann, Dieter, Volkens, Andrea, Bara, Judith, Budge, Ian and McDonald, Michael D., Mapping Policy Preferences II: Estimates for Parties Governments and Electors in the OECD, EU and Central and Eastern Europe 1990–2003 (Oxford: Oxford University Press, 2006)Google Scholar.
46 It is important to note that voters do perceive the changes that are identified by the CMP measures. The BES surveys ask the following question: ‘Considering everything the parties stand for, would you say there is a good (1964–1970)/great (1974) deal of difference between the parties, some difference, or not much difference?’, later amended to: ‘Considering everything the Conservative and Labour parties stand for, would you say there is a great deal of difference between them, some difference, or not much difference?’. Answers to these questions map closely onto the pattern of changes observed in the CMP: the proportion of BES respondents who perceive the parties to be a ‘good deal’ (or a ‘great deal’) different in each election and the difference between the Labour and Conservative parties in left–right manifesto positions correlate at 0.87. This indicates that people are aware of the major differences between the main parties, and although there is no explicit reference to a left–right ideological dimension in these questions, it would appear that voters are reacting to party polarization in the same terms as the manifesto project.
47 The total number of non-missing cases on the dependent variable is 101,788. We model only those cases that have non-missing data on the control variables that we wish to include in later models, and as household income in particular has a non-response rate of around 15 per cent this reduces our available cases.
48 There is evidence that change over time in British party support has been driven by generational change; see for example: Butler and Stokes, Political Change in Britain; Tilley, James, ‘Political Generations and Partisanship in the UK, 1964–1997’, Journal of the Royal Statistical Society: Series A, 165 (2002), 121–135CrossRefGoogle Scholar.
49 This is mainly due to the invariant effect that these factors have on party choice. Apart from some slight differences in the earliest period for the effects of education, and some changes in how income affects party choices, there is little change in the impact of these factors over time. There are of course large changes in the proportions of the population living in council housing, with higher education, and who are members of trade unions, but the effect of being a trade unionist in the 2000s is not dissimilar to the effect in the 1960s.
50 Unfortunately, not all years have full information on every control variable, and so this comparison refers to two models run on all years that have the full set of control variables (70,000 cases). The February 1974 BES is missing religion, the 1983 BES is missing income, the 1997 BSA is missing private schooling, the 2001 BES is missing housing tenure and private schooling and the 2005 BES is missing private schooling. The graphs below do include these years, as we have modelled them separately with the full range of control variables that are available (and, in the case of the 1983 BES, a subjective measure of income). These estimates of the class coefficients are in all cases very similar to those estimates from adjacent surveys.
52 The number of observations at the individual and survey level is lower than previously as we are unable to include the February 1974 BES, the 1983 BES, the 1997 BSA, the 2001 BES and the 2005 BES in a pooled model due to the missing data on key independent variables mentioned earlier.
53 As previously, these refer to a white male born in the 1930s with a minimum education in the state system, who does not belong to a trade union, is in the third income quintile, has no religion and is an owner-occupier.
54 Note that we do not include neither the main effects of party divergence, nor the two-way interactions of class–party divergence and post-1974–party divergence. We have no theoretical reason for including them as we are arguing that the only impact of party divergence is on the class coefficients themselves. Equally, we should note that running a full model with all interactions and main effects gives almost identical results to those presented here.
55 Sanders, David, ‘The Impact of Left–Right Ideology’, in Geoffrey Evans and Pippa Norris, eds, Critical Elections: British Parties and Voters in Long-Term Perspective (London: Sage, 1999), pp. 181–206Google Scholar; Adams, James, Green, Jane and Milazzo, Caitlin, ‘Has the British Public Depolarized along with Political Elites? An American Perspective on British Public Opinion’, Comparative Political Studies, 45 (2012)CrossRefGoogle Scholar.
57 McCarty, Nolan, Poole, Keith T. and Rosenthal, Howard, Polarized America: The Dance of Ideology and Unequal Riches (Cambridge, Mass: MIT Press, 2006)Google Scholar.
58 Bartels, Larry M., Unequal Democracy: The Political Economy of the New Gilded Age (Princeton, N.J.: Princeton University Press, 2008)Google Scholar.
59 Huber, John D., and Stanig, Piero, ‘Voting Polarization on Redistribution across Democracies’ (paper presented at the 2006 Annual Meetings of the American Political Science Association, Philadelphia, 2006)Google Scholar.
62 Campbell, Angus, Converse, Philip E., Miller, Warren E. and Stokes, Donald E., The American Voter (Chicago: University of Chicago Press, 1960), pp. 234–240Google Scholar; Converse, Philip E., ‘The Nature of Belief Systems in Mass Publics’, in David E. Apter, ed., Ideology and Discontent (New York: The Free Press, 1964)Google Scholar.
63 Curtice, John, ‘Elections and Public Opinion’, in A. Seldon, ed., Blair's Britain 1997–2007 (Cambridge: Cambridge University Press, 2007), pp. 35–53Google Scholar.
64 Franklin, The Decline of Class Voting in Britain.
65 Heath, Anthony, Jowell, Roger, Curtice, John, Evans, Geoffrey, Field, Julia and Witherspoon, Sharon, Understanding Political Change: The British Voter 1964–1987 (Oxford: Pergamon, 1991)Google Scholar. The Liberals fielded only 311 and 332 candidates in 1966 and 1970 respectively, but 517 in February 1974 and 619 in October 1974. There were also large increases in the number of nationalist candidates in Scotland and Wales over this period.
66 Evans, Heath and Payne, ‘Modelling the Class/Party Relationship 1964–87’.
67 Franklin, Mackie, Valen et al., Electoral Change: Responses to Evolving Social and Attitudinal Structures in Western Countries; Evans, Geoffrey, ed., The End of Class Politics? Class Voting in Comparative Context (Oxford: Oxford University Press, 1999)CrossRefGoogle Scholar; Elff, Martin, ‘Social Structure and Electoral Behavior in Comparative Perspective: The Decline of Social Cleavages in Western Europe Revisited’, Perspectives on Politics, 5 (2007), 277–294CrossRefGoogle Scholar.