Voting Behaviour in Britain

1
Voting Behaviour in Britain

• Lynn Bennie

2
What are the main influences on voting in
Britain?

• STUDIES OF VOTING BEHAVIOUR
• THE BUTLER AND STOKES MODEL
• VOTING BEHAVIOUR FACTORS
• DEALIGNMENT AND REASONS FOR CHANGE
• RECENT DEVELOPMENTS IN ELECTION STUDIES
• TURNOUT
• CONCLUSION 2005 and Beyond

3
STUDIES OF VOTING BEHAVIOUR

• Methods
• Analysis of election results
• Voter surveys e.g. British Election Study. We
  have data on 40 years of elections.

4
BUTLER AND STOKES

• Political Change in Britain 1964, 1966, 1970
• Class and partisan id key i.e. long-term factors
• Importance of socialisation

5
Class

• Central cleavage in society
• 1966 of the 100 constituencies with the largest
  of manual workers, Lab won 99.
• Pulzer 1967 Class central determinant, all else
  is embellishment and detail.

6
VOTING BEHAVIOUR FACTORS

• Long term structural factors
• Class
• Religion
• Sex
• Age
• (PARTY ID SOCIOLOGICAL MODELS)

7
Factors cont.

• Short term
• Issues
• Personalities
• Leaders
• Campaigns
• The media
• (RATIONAL CHOICE MODELS)

8
Social Cleavages

• Manual non-manual
• Worker employer
• Public sector private sector
• Centre periphery
• Secular religious
• Urban rural
• Material post-material

9
Ivor Crewe (in Bartle King 2005)

• There is no way of knowing, but there are ways
  of arriving at credible probabilities. One
  useful approach is to think of any election
  result as the product of long-term structural
  factors and short-term campaign factors.

10
Dealignment/ Change

• Decline in voter loyalty/class id/partisan id
• 1966 46 very strongly identified with a
  party. 2005 8
• But Total with some id 1966 94 2005 81

11
Other indicators of change

• Tactical voting 17 of Lib Dems in 05.
• Issue voting Economy, Iraq war
• Campaigns matter
• Media
• Decline in turnout
• Rise in support for other parties.
• Decline in party membership
• Rise in general cynicism

12
INCREASED VOLATILITY

• A Restless Electorate
• 1997 Swing of 10 from Con to Lab double
  post-war record of 79
• ¼ of voters did not support same party
• Rise in support for others
• Move from habitual to judgemental voting (Denver)

13
Reasons Bottom-up.

• Increase in education
• Changing occupational and industrial structures
• Decline of working class
• Increase in affluence
• Cross-cutting pressures

14
But also Top-down

• Performance of governments
• The Media
• Leading to disillusionment
• Overall shift in balance between long-term (-)
  and short-term ()

15
Turnout

• 2001 59.2. Lowest since 1918
• 2005 61.3
• Generational change evident decline in duty to
  vote/ sense of civic duty in Thatcher and Blair
  generations
• So, turn-out likely to remain low???

16
Recent BES evidence

• Harold Clarke et al 2004
• Analyse entire series of BES
• Note the general decline in sense of civic duty
• Decline of position politics and rise of valence
  politics

17
Position Politics

• Involves issues on which the public take
  different sides e.g. privatisation, trade union
  rights. It was previously argued that voters
  assessed the parties on issues and voted for the
  party that was closest to their own position.

18
Valence Politics

• General judgements on party policies (the
  economy), party performance and leaders.
  Involves issues on which everyone is (more or
  less) agreed on the end to be pursued e.g. crime,
  peace and prosperity, corruption.

19
Competence key

• general judgement on the relative competence of
  the parties to achieve desired ends.
• the politics of delivery, not fixed ideological
  positions
• Labour seen as most competent, until now.

20
Partisan id no longer important?

• Still important, but reconceptualised
• Valenced partisanship
• A storehouse of accumulated party and party
  leader performance evaluations.
• It is continuously updated as voters acquire new
  information and react to events voters
  continuously make judgements about the competence
  of parties, governments and leaders.

21
Are old models at all relevant?

• YouGov polls ask if people generally identify
  with any party and 70 say yes.
• BES reveals that over 80 of voters still
  identify with one party
• Most survey respondents in BES are stable in
  their party id
• Still possible to identify natural Lab Con
  supporters, and at a constituency level point to
  Cons and Lab heartlands.
• So, tribal voters still exist (Denver).

22
But VB more complicated now

• Greater turbulence within the electorate.
• Non-voting phenomenon is a hot topic - we could
  do more to understand this.
• In 2005, the Govt was elected with the support
  of only 21.6 of the eligible electorate i.e.
  non-voters (40) out-number those who voted for
  the winning party by nearly 2 to 1.
• Has led to much debate in parties and
  universities democratically legitimate??

23
Conclusion

• Top-level political scientists simply debate the
  importance of different factors and influences.
• Most do not/ cannot offer definitive conclusions.