216 POLITICS

in terms of his ‘impossibility theorem’. In Social Choice and Individual Values (1951)
Arrow drew attention to the problem of ‘transitivity’ that occurs when voters are
allowed to express a range of preferences for candidates or policy options, rather
than merely cast a single vote. The drawback of casting but a single vote is not only
that it is a crude all-or-nothing device, but also that no single candidate or option
may gain majority support. For instance, candidate A may gain 40 per cent of the vote,
candidate B 34 per cent, and candidate C 26 per cent. The situation could, neverthe-
less, become more confused if second preferences were taken into account.
Let us assume, for the sake of argument, that the second preferences of all
candidate A’s supporters go to candidate C, the second preferences of candidate
B favour candidate A, and the second preferences of candidate C go to candidate
B. This creates a situation in which each candidate can claim to be preferred by
a majority of voters. The first and second preferences for candidate A add up to
74 per cent (40 per cent plus B’s 34 per cent). Candidate B can claim 60 per cent
support (34 per cent plus C’s 26 per cent), and candidate C can claim 66 per cent
support (26 per cent plus A’s 40 per cent). This problem of ‘cyclical majorities’
draws attention to the fact that it may not be possible to establish a reliable link
between individual preferences and collective choices. In other words, election
results cannot speak for themselves, and politicians and political commentators
who claim to find meaning in them are, to some extent, acting arbitrarily.
Nevertheless, the latitude that this allows politicians is not unlimited, because
they know that they will be called to account at the next election. In this light,
perhaps the most significant function of elections is to set limits to arbitrary
government by ensuring that politicians who claim to speak for the public must
ultimately be judged by the public.

VOTING BEHAVIOUR
The growth of academic interest in voting behaviour coincided with the rise of
behavioural political science. As the most widespread and quantifiable form of
political behaviour, voting quickly became the focus for new techniques of
sample surveying and statistical analysis. The American Voter (Campbell et al.,
1960), the product of painstaking research by the University of Michigan,
became the leading work in the field and stimulated a wealth of similar studies,
such as Butler and Stokes’ Political Change in Britain (1969). At the high point of
the behavioural revolution, it was thought that voting held the key to disclosing
all the mysteries of the political system, perhaps allowing for laws of mass politi-
cal psychology to be developed. Even though these lofty hopes have not been
fulfilled, psephology (the scientific study of voting behaviour) still commands a
central position in political analysis. This is because voting provides one of the
richest sources of information about the interaction between individuals, society
and politics. By investigating the mysteries of voting behaviour, we are thus able
to learn important lessons about the nature of the political system, and gain
insight into the process of social and political change.
Voting behaviour is clearly shaped by short-term and long-term influences.
Short-term influences are specific to a particular election and do not allow con-
clusions to be drawn about voting patterns in general. The chief short-term influ-
ence is the state of the economy, which reflects the fact that there is usually a link
 R E P R E S E N TAT I O N , E L E C T I O N S A N D VO T I N G 217

between a government’s popularity and economic variables such as unemploy-

CONCEPT ment, inflation and disposable income. Optimism about one’s own material
Partisan circumstances (the so-called ‘feel-good’ factor) appears to be particularly crucial
dealignment here. Indeed, it is often alleged that governments attempt to create pre-election
booms in the hope of improving their chances of gaining re-election. The
Partisan dealignment is a
decline in the extent to chances that political and business cycles can be brought into conjunction are
which people align clearly strengthened by flexible-term elections that allow the government to
themselves with a party choose when to ‘go to the country’.
by identifying with it. Another short-term influence on voting is the personality and public standing
This implies that the of party leaders. This is particularly important, because media exposure portrays
‘normal’ support of
parties falls, and a leaders as the brand image of their party. This means that a party may try to rekin-
growing number of dle popular support by replacing a leader who is perceived to be an electoral liabil-
electors become ‘floating’ ity. Another factor is the style and effectiveness of the parties’ electoral
or ‘swing’ voters. As party campaigning. The length of the campaign can vary from about three weeks for
loyalties weaken, flexible-term elections to up to two years in the case of fixed-term elections, such
electoral behaviour
becomes more volatile, as those for the US president. Opinion polls are usually thought to be significant
leading to greater in this respect, either giving a candidate’s or party’s campaign momentum, or
uncertainty and, perhaps, instilling disillusionment, or even complacency, amongst voters.
the rise of new parties, or A final short-term influence, the mass media (see p. 179), may also be of
the decline of old ones. long-term significance if biased or partisan coverage reflects structural, and
The principal reasons for
partisan dealignment are therefore continuing, factors such as press ownership. However, the pattern of
the expansion of media coverage may change from election to election. For instance, under Tony
education, increased Blair’s leadership, the UK Labour Party made concerted attempts to court the
social mobility, and Murdoch press in particular, helping to explain the party’s longest period in
growing reliance on power, between 1997 and 2010. All such considerations, nevertheless, operate
television as a source of
political information. within a context of psychological, sociological, economic and ideological influ-
ences on voting. These are best examined in relation to rival models of voting.
The most significant of these are the following:

 the party-identification model

 the sociological model
 the rational-choice model
 the dominant-ideology model.

Theories of voting
Party-identification model
The earliest theory of voting behaviour, the party-identification model, is based
on the sense of psychological attachment that people have to parties. Electors are
seen as people who identify with a party, in the sense of being long-term
supporters who regard the party as ‘their’ party. Voting is therefore a manifesta-
tion of partisanship, not a product of calculation influenced by factors such as
policies, personalities, campaigning and media coverage. This model places
heavy stress on early political socialization (see p. 178), seeing the family as the
principal means through which political loyalties are forged. These are then, in
most cases, reinforced by group membership and later social experiences.
In this model, attitudes towards policies and leaders, as well as perceptions
about group and personal interests, tend to be developed on the basis of party
218 POLITICS

identification. Events are thus interpreted to fit with pre-existing loyalties and

CONCEPT attachments. This partisan alignment tends to create stability and continuity,
Class especially in terms of habitual patterns of voting behaviour, often sustained over
dealignment a lifetime. From this point of view, it should be possible to calculate the ‘normal’
vote of a party by reference to partisanship levels. Deviations from this ‘normal’
Class dealignment is the
weakening of the level presumably reflect the impact of short-term factors. One of the weaknesses
relationship between of this model is the growing evidence from a number of countries of partisan
social class and party dealignment (see p. 217). This indicates a general fall in party identification and
support. Social class may a decline in habitual voting patterns. In the USA, partisan dealignment is
nevertheless remain a reflected in a decline in the number of registered Democrats and Republicans,
significant (even the
most significant) factor and a rise in the number of Independents (up from 6 per cent in 1952 to 36 per
influencing electoral cent in 2009). In the UK, it is demonstrated by a decline in the strength of alle-
choice. The impact of giance to the Conservative Party and the Labour Party, ‘very strong’ identification
dealignment has been to with either party having fallen from 43 per cent in 1966 to 9 per cent in 2005.
undermine traditional
class-based parties
(notably, working class Sociological model
parties of the left), often
bringing about a The sociological model links voting behaviour to group membership, suggesting
realignment of the party that electors tend to adopt a voting pattern that reflects the economic and social
system. Explanations of position of the group to which they belong. Rather than developing a psycholog-
class dealignment usually
focus on changes in the ical attachment to a party on the basis of family influence, this model highlights
social structure that have the importance of a social alignment, reflecting the various divisions and tensions
weakened the solidaristic within society. The most significant of these divisions are class, gender, ethnicity,
character of class religion and region. Although the impact of socialization is not irrelevant to this
identity, such as post- model, social-base explanations allow for rationality insofar as group interests
industrialism.
may help to shape party allegiances. For many analysts, the sociological model is
best understood as an ‘interest plus socialization’ approach to voting (Denver,
2012). This has perhaps been clearest in relation to social class (see p. 153).
Not uncommonly, party systems have been seen to reflect the class system, with
the middle classes providing the electoral base for right-wing parties, and the
working classes providing the electoral base for left-wing parties. The Labour–
Conservative two-party system in the UK was traditionally understood in
precisely this light. Peter Pulzer (1967) was able to declare, famously, ‘class is the
basis of British party politics; all else is embellishment and detail’. The sociolog-
ical model, however, has been attacked on the grounds that, in focusing on social
groups, it ignores the individual and the role of personal self-interest. Moreover,
there is growing empirical evidence that the link between sociological factors
and party support has weakened in modern societies. In particular, attention has
been paid to the phenomenon of class dealignment. Evidence of class dealign-
ment can be found in most western societies. For example, absolute class voting
(the proportion of voters who support their ‘natural’ class party) fell in the UK
from 66 per cent in 1966 to 47 per cent in 1983. In 1997, the Labour Party, for
the first time, received more votes from non-manual workers than from manual
workers.

Rational-choice model
Rational-choice models of voting shift attention onto the individual, and away
from socialization and the behaviour of social groups. In this view, voting is seen
 R E P R E S E N TAT I O N , E L E C T I O N S A N D VO T I N G 219

as a rational act, in the sense that individual electors are believed to decide their
party preference on the basis of personal self-interest. Rather than being habit-
ual, a manifestation of broader attachments and allegiances, voting is seen as
essentially instrumental; that is, as a means to an end. Rational-choice models
differ in that some, following the example of V. O. Key (1966), see voting as a
retrospective comment on the party in power and how its performance has
influenced citizen’s choice. Others, such as Himmelveit et al., (1985), portray
voters as active, in the sense that they behave like consumers expressing a choice
amongst the available policy options.
The latter view stresses the importance of what is called ‘issue voting’, and
suggests that parties can significantly influence their electoral performance by
revising and reshaping their policies. It is generally accepted that this has been
one of the consequences of partisan and class dealignment. This has also been
encouraged by the pluralism and individualism that postmodernism (see p. 18)
has fostered. The weakness of rational-choice theories is that they abstract the
individual voter from his or her social and cultural context. In other words, to
some extent, the ability to evaluate issues and calculate self-interest (the essence
of instrumental voting) is structured by broader party attachments and group
loyalties.

Dominant-ideology model
Radical theories of voting tend to highlight the degree to which individual choices
are shaped by a process of ideological manipulation and control. In some
respects, such theories resemble the sociological model, in that voting is seen to
reflect a person’s position in a social hierarchy. Where these theories differ from
the sociological model, however, is in emphasizing that how groups and individ-
uals interpret their position depends on how it has been presented to them
through education, by the government and, above all, by the mass media. (The
influence of the media on political debate and party competition is examined in
greater detail in Chapter 8.)
In contrast to the earlier view that the media merely reinforce pre-existing
preferences, this suggests that the media are able to distort the flow of political
communications, both by setting the agenda for debate and by structuring pref-
erences and sympathies. The consequence of this is that, if voters’ attitudes
conform to the tenets of a dominant ideology, parties will not be able to afford
to develop policies that fall outside that ideology. In this way, far from challeng-
 Issue voting: Voting
ing the existing distribution of power and resources in society, the electoral
behaviour that is shaped by
party policies and (usually) a process tends to uphold it. The weakness of the dominant-ideology model is
calculation of personal self- that, by overstating the process of social conditioning, it takes individual calcu-
interest. lation and personal autonomy out of the picture altogether.