The Floating Voter and the Liberal View of Representation
| Published date | 01 June 1969 |
| Author | A. H. Birch,J. G. Blumler,Alison Ewbank,R. J. Benewick |
| Date | 01 June 1969 |
| DOI | 10.1111/j.1467-9248.1969.tb00634.x |
| Subject Matter | Article |
THE FLOATING VOTER
AND
THE LIBERAL
VIEW
OF
REPRESENTATION’
R.
J.
BENEWICK
AND A.
H.
BIRCH
University
of
Hull
AND
J.
G.
BLUMLER
AND
ALISON EWBANK
University oj’leeds
THE
surveys of voting behaviour and political opinions made between
1950
and
1962
yielded a fair amount of data about the role of the voter in the
British representative system. The significance of these findings (assuming,
for the moment, that they were accurate) depends on the extent to which they
enhance our understanding of the nature of the representative system and its
place in the process of government, an understanding which has itself been
influenced by various liberal views about how the political system ought to
operate. In this paper we propose, first, to outline some of the main findings;
second, to relate them briefly to liberal views about representation; and
third,
to
report some of the findings of a survey which lead
us
to modify
some of the conclusions which seem to follow from earlier work.
EARLIER SURVEYS
The following are some of the main findings of earlier surveys, which will
be reported in the present tense for the sake of simplicity.
1.
Voting behaviour is highly stable. At any given time about 90 per cent.
of potential voters know how they will vote2 and the majority of them say
they vote the same way at election after election.3 The Conservative and
Labour Parties are each assured of the support of about one-third of the
electorate regardless of the party’s performance in office
or
in opposition.4
I
We are greatly indebted to Denis McQuail of Southampton University, who participated in
this
project during its early stages. We would
also
like to thank the Sub-Department of Computa-
tion of the University of
Hull
for its generous assistance. The survey was financed by grants from
the Institute
of
Electoral Research and the University of
Hull,
for
which the authors are very
grateful.
National Opinion Polls find that about
92
per ccnt.
of
respondents normally indicate
a
preference for one party or another. Constituency surveys have reported figures of between
85
per cent. and
93
per cent.
3
See,
for instance, Mark Abrams, ‘Social Trends and Electoral Behaviour’,
British Journal
of
Sociology,
Vol.
13,
1962,
pp.
230-2.
4
See Mark Abrams, ‘Opinion Polls and Party Propaganda’,
Public
Opinion Quarterly,
Vol.
28,
1964,
pp.
13-19.
Political
Studies,
Vol.
XMI,
No.
2
(1969,177-195).
12
178
THE FLOATING VOTER
2.
Political loyalties are largely inherited. Most people accept the political
beliefs of their parents and change them, if at all, only in the light of experi-
ence.’
3. There is a clear relationship between occupation and voting behaviour,
too familiar to need repetition.2
4.
Opinions about questions of policy are not closely related to voting
behaviour. Jean Blonde1 has shown that there is little or no partisanship
in relation to moral issues or questions of foreign or defence policy:
electors are divided in their views but there
is
virtually no relationship
between this division and the division between Labour and Conservative
supporters.3 On economic and social issues there is a relationship between
opinions and party loyalties but it is not a very close one. The Greenwich
and Bristol surveys4 and the
Socialist Commentary
surveys reached fairly
similar conclusions and showed in addition that Labour voters were more
likely to disagree with the policies of their party than were Conservative
voters. Not surprisingly, in view of this, only
a
minority of voters appear to
have party policies uppermost in their mind when they go to the polls. In
the Bristol survey of 1951 only 31 per cent. of the voters, when asked why
they had voted as they had, mentioned anything that could be classified as a
policy or issue. And the authors of the report
on
this survey gave reasons
for their own conclusions that no more than about 10 per cent. of the
voters had actually made their choice on the basis of an issue or set of
issues.6The survey results all point in the same direction and the overall con-
clusion would seem to be that between
40
and 50 per cent. of the voters at a
general election vote for the party of their choice either in spite of its
policies or in ignorance of them.’
5. Election results turn on the behaviour of a relatively small number of
floating voters. As we have already seen, approximately two-thirds
of
the
electorate are committed supporters of the two major parties. The remaining
one-third of the electorate is composed
of
waverers who in fact seldom do
change their vote,8 Liberals, abstainers (the largest sub-category) and actual
changers. Yet the influence
of
the latter is magnified out of all proportion by
the effect of the ‘cube law’ in the British electoral system, which greatly
inflates the parliamentary majority of the most successful party.
I
See
Mark Abrams, ‘Social Trends and Electoral Behaviour’,
op.
cit.,
pp.
23942.
For
a summary,
see
A.
H.
Birch,
The British System
of
Government
(London,
1967),
pp.
91-2.
For
an
analysis
of
working-class Conservatism, see R.
T.
McKenzie and Allan Silver,
Angels
in
Marble
(London,
1968).
Milne and
H.
C.
Mackenzie,
Marginal Sear,
1955
(London,
1958),
pp.
117-20.
R. S.
Milne and H. C. Mackenzie,
Straighr Fight
(London,
1954)
p.
139.
See
P.
G.
J.
Pulzer,
Political Represenration and Elections
in
Britain
(London,
1967),
pp.
1 14-1
5.
3
Jean Blondel,
Voters, Parries andLeaders
(London,
1963),
pp.
75-87.
4
M. Benney, A.
P.
Gray and R.
H.
Pear,
How
People Vote
(London,
1956),
pp.
140-1
;
R.
S.
5
Mark Abrams
and
Richard Rose,
Must Labour Lose?
(London,
1960),
pp.
11-20,
*
See
R.
S.
Milne and
H.
C.
Mackenzie,
Straight Fight,
pp.
130-1
;
Marginal Sear,
pp.
41-2.
To continue reading
Request your trial