Contemporary Party Politics

Contemporary Party Politics
This page intentionally left blank

Contemporary Party Politics
Robin T. Pettitt
© Robin T. Pettitt 2014
All rights reserved. No reproduction, copy or transmission of this

publication may be made without written permission.
No portion of this publication may be reproduced, copied or transmitted

save with written permission or in accordance with the provisions of the
Copyright, Designs and Patents Act 1988, or under the terms of any licence
permitting limited copying issued by the Copyright Licensing Agency,
Saffron House, 6-10 Kirby Street, London EC1N 8TS.
Any person who does any unauthorized act in relation to this publication
may be liable to criminal prosecution and civil claims for damages.
The author has asserted his right to be identified as the author of this work
in accordance with the Copyright, Designs and Patents Act 1988.
First published 2014 by

PALGRAVE MACMILLAN
Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,

registered in England, company number 785998, of Houndmills, Basingstoke,
Hampshire RG21 6XS.
Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

175 Fifth Avenue, New York, NY 10010.
Palgrave Macmillan is the global academic imprint of the above companies

and has companies and representatives throughout the world.
Palgrave® and Macmillan® are registered trademarks in the United States,

the United Kingdom, Europe and other countries
ISBN: 978-0-230-23779-7 hardback

ISBN: 978-0-230-23780-3 paperback
This book is printed on paper suitable for recycling and made from fully
managed and sustained forest sources. Logging, pulping and manufacturing
processes are expected to conform to the environmental regulations of the
country of origin.
A catalogue record for this book is available from the British Library.
A catalog record for this book is available from the Library of Congress.
Til min Regnbue
This page intentionally left blank
Contents
List of Figures x
List of Tables xi
Acknowledgements xii
1 Introduction 1
Defining ‘political party’ 4
The three faces of party organisations 7
The structure of parties 9
The role of parties in democratic states 14
The crisis of parties 16
Outline of the book 19
2 Party Systems 21
The importance of understanding party systems 21
Early approaches to party systems 25
Party system typologies 27
The effective number of parties 33
Explaining party system variation 36
Which party system is best? 40
Conclusion 42
3 Theories of Party Development 43

The ‘internal’ and ‘external’ origins of parties 43
An alternative view of the emergence of political parties 47
Understanding party types 50
Cadre parties to mass parties 50
Challenging the mass party model: the catch-all party 52
Organisational consequences of the catch-all party:
the electoral-professional party model 53
The cartel party 57
Conclusion: Beyond the cartel – beyond party types? 58
4 Ideology 60
On ideology and party politics 60
Political cleavages 64
Party families 67
vii
viii Contents
Ideology in contemporary European party politics 70

Party ideology beyond Europe 76
United States 76
India 79
Africa 81
Conclusion 82
5 Party Members, Activists and Supporters 84

Defining party member, activist and supporter 85
The rise (and fall?) of party members 88
Why members? 91
Party membership incentives: supply 92
Leadership incentives: demand 94
The costs of a membership organisation 97
Explaining and reversing decline 99
The effects of membership decline 100
Conclusion 101
6 Candidate Selection 103

The importance of candidate selection 103
Selecting candidates 005
The political system 106
National legislation 108
The candidate selection process: impact of different
methods 112
Territorial (de-) centralisation and in/exclusiveness of the
selectorate 113
Supply and demand 118
Supply 119
Demand 121
Conclusion 123
7 Policy Making 124

Who should be involved? 124
The normative argument against intra-party democracy 125
The normative argument for intra-party democracy 126
Pragmatic arguments against intra-party democracy 130
Pragmatic argument for intra-party democracy 132
Who can be involved in policy making? 133
Who wants to be involved in policy making? 136
Who are involved in policy making? 137
Conclusion 141
Contents ix
8 Campaigning 142
How campaigning has changed 142
First era 144
Second Era 145
Third Era 147
The limits of change 149
Drivers of change 153
Social changes 153
Technological change 154
Party-specific issues 155
Impact of changes 157
Financing 158
Conclusion 161
9 Government 162
Deciding to aim for government office 162
Getting into government – coalitions 169
Factors impacting on coalition formation 173
Portfolio allocation 175
Conclusion 176
10 The Internationalisation of Party Politics 178

Party Internationals 178
The First, Second, Fourth and Socialist Internationals 179
The Liberal International 180
International Democratic Union 181
Communist International (Comintern) 181
Europarties 185
The Development of Europarties 185
Limits to the development of Europarties 189
Parties and globalization 194
The internationalisation of campaign consultancy 197
Conclusion 199
11 The Future of Party Politics 201

Future direction in the study of political parties 201
Are parties in crisis? 205
Bibliography 208
Index 221
List of Figures
1.1 Confidence in political parties: Not very much/None at

all (percentage) 17
2.1 Streams of interaction 23
2.2 Election results in Denmark (2011) and the UK (2010) 24
2.3 Segmentation and polarisation 29
2.4 ENEP and ENPP in the UK and New Zealand 35
2.5 The making of multi-party systems through overlapping
political divides 39
3.1 Party origins and institutionalisation 49
4.1 Downs’ voter distribution and party placement 63
4.2 The electoral support for party families 73
6.1 Funnel of candidate selection 105
6.2 The candidate selection context 106
6.3 Territorial (de-) centralisation and in/exclusiveness of the
selectorate 115
7.1 Curvilinear disparity 131
9.1 Support for the Social Democrats and the Socialist
People’s Party in Denmark after the 2011 Election 167
9.2 Coalitions in a minority situation for hypothetical
100-seat legislature 170
10.1 Relations between Europarties and European Parliament
Groups 189
10.2 Americanization vs. Modernization 199
x
List of Tables
6.1 Illustration of registration requirements for candidates

for the lower chamber 110
6.2 Illustrations of candidate selection procedures 116
7.1 Levels of policy and potential for consultation 135
8.1 Eras of campaigning 143
10.1 Brief description of existing Europarties 186
10.2 Key organisational features of Europarty statutes 190
xi
Acknowledgements
One does not complete a project such as this without incurring consider-
able debt. On the academic side I would like to gratefully acknowledge
the work of several anonymous reviewers who have commented on the
text at various stages from proposal to final draft. They have undoubt-
edly helped improve the manuscript significantly. I would also like to
thank the people at Palgrave Macmillan for encouraging me in my
endeavours. I owe particular thanks to Steven Kennedy for seeing the
book through its early stages and to Stephen Wenham for seeing it
through to completion. Stephen Wenham in particular has been very tol-
erant of my many broken deadlines.
I would also like to thank Pascal Perrineau and CEVIPOF – Paris for
hosting me during some of my most productive months working on the
book. Rejane Senac facilitated my initial introduction to CEVIPOF for
which I am very grateful. Odile Gaultier-Voituriez was ever helpful in
getting me access to key pieces of literature.
Most importantly I need to thank my family for their tolerance and
forgiveness – especially Rainbow Murray for her love and support
throughout the slow birth of this book. Jeg elsker dig. Speaking of slow
births, it would not do to neglect the most important development during
the making of this book: the arrival of my beautiful daughter Alicia Hope
Murray-Pettitt. This book will not mean very much to you for the fore-
seeable future – but at least there will be one more thing in the house for
you to gnaw on.
xii
Chapter 1
Introduction
Political parties are the most important organisations in modern demo-

cratic states. That is the simplest and most straightforward reason for
reading, and indeed writing, a book dedicated to the study of political
parties. There are many reasons why parties are important. If we want to
know why so few women become legislators; if we want to know why
one person and not another gets to lead a country; or how much of a
choice we have on the ballot paper, we need to look to political parties
for most of our answers. The importance of political parties is captured
by Schattschneider’s oft quoted claim that ‘modern democracy is unthink-
able save in terms of parties’ (1942, p. 1). No modern democracy has
been made to work without political parties. Modern democracy is party
democracy. One of the few attempts to build a ‘no-party democracy’,
Uganda, is really little more than a thinly veiled one-party state and
struggles to live up to most definitions of democracy.
The dominance of parties in the modern world is also illustrated by the
fact that, not only do they sit at the centre of all democratic countries;
they are also a feature of many non-democratic systems. There are two
major examples of this. The first is Communist single-party states, most
notable the former Soviet Bloc in Eastern Europe and the Soviet Union.
In these countries the only party allowed to exist was the Communist
Party. The Communist Party sat at the centre of all political power and
reached far into people’s lives. It was therefore a tool used by Communist
leaders for political and social control of the population. Even today a
Communist Party dominates political life in the world’s most populous
nation, The People’s Republic of China.
The second example is the Ba’ath Party in Iraq and Syria. As with the
Communist Parties in the Soviet Bloc and China, the Ba’ath Party was the
only legal party in Syria and Iraq and used as a way of controlling all
aspects of the state.
It could even be argued that understanding political parties is now
more important than ever. With the end of the Cold War there was a
major expansion of democratic government. Many military dictatorships
and Communist single-party states became multi-party democracies,
vastly increasing the number of political parties in the world. Indeed,
with the end of the Cold War, liberal democracy, which is also party
1
2 Contemporary Party Politics
democracy, was seen as having seen off its final competitor (Com-
munism) for the position of most desirable political system. The future,
argued people like Francis Fukuyama, belonged to liberal (party based)
democracies. If the future is indeed in the hands of liberal democracies,
then understanding the organisations at the heart of this political system
must be a prime concern.
Indeed, when looking around the world, there are only a handful of
countries, such as Saudi Arabia and the Central African Republic, that do
not have political systems based around political parties. The fact that so
few countries have political systems devoid of parties in some form
underlines the importance of this type of organisation. In short, political
parties are at the heart of all democratic and a great many non-democ-
ratic countries.
The fact that political parties should be such a dominant feature of
both democratic and non-democratic systems of government is hardly
surprising. Political parties organise people with roughly similar view-
points and give them a ‘label’ (that is, the party’s name). Under that label,
parties then present a programme for what they want to do should they
acquire power. This gives voters, the media and other observers a way of
judging what political trends are moving in a society. Imagine a parlia-
ment of several hundred members, without parties. It would be well
neigh impossible to establish a working majority for a government.
Voters would have a difficult time judging who was responsible for
certain actions, and therefore faced with a difficult task when deciding
whom to support at an election.
Further, political parties provide a critical link between citizens and
governments through their membership organisations. Political parties
have both a civil society element (the membership organisation) and a
government element (parliamentarians and ministers). Through that they
can give their leaders a level of popular legitimacy that personal indi-
vidual rule would find it difficult to match. This works both in demo-
cratic multi-party democracies and in one-party authoritarian regimes.
Even if there is no democratic element in a country’s system, dictators
have often found it expedient to at least pretend to have popular legiti-
macy through a party organisation.
Political parties are the vehicles through which both voters and cam-
paigners can be mobilised behind some cause – which in turn is key to
winning elections in democratic countries. Indeed, Michels wrote that:
[a] class which unfurls in the face of society the banner of certain defi-
nite claims and which aspires to the realization of a complex of ideals
[...] needs an organization. Be the claims economic or be they political,
organization appears the only means for the creation of a collective
Introduction 3
will. Organization [...] is the weapon of the weak in their struggle with
the strong (1915, p. 21)
In other words if people who do not have significant personal resources

want to achieve their aims, the best way of doing so is through collective
action in the form of an organisation.
Now, obviously such political organisations need not take the form of
a political party. Trade unions, single issue organisations, terrorist groups
and rebel armies are all examples of organisations pursuing ‘certain defi-
nite claims’. However, in a democratic polity the most effective and direct
way of realising a ‘complex of ideals’ is by gaining control of the levers of
government power. That means succeeding in getting elected to legislative
assemblies or other positions of power, which in turn is most easily done
through a political party. Nowhere in the democratic world do independ-
ents – politicians without party affiliation – dominate elected positions.
Instead, political parties do. As Aldrich (2011, p. ix) writes ‘for politi-
cians to win more of what they seek to win, more often, and over a
longer period’ they need political parties.
In short, political parties are central to the workings of democratic
politics and are also a feature of the politics of many non-democratic
countries. No other type of political organisation can claim to have the
impact and influence that political parties have. They are simply the most
important political organisation in the modern world.
However, while political parties are a feature of both democratic and
undemocratic regimes this book will focus on party politics in demo-
cratic countries. ‘Democratic’ is obviously a highly contested term and
there is no clear separation of democratic and non-democratic politics. It
may at times be fairly easy to tell a clearly democratic country from a
clearly autocratic one. Putting for example Sweden next to say Saudi
Arabia the differences would be clear enough. However, there is a lot of
grey between those two polar opposites. In essence, the focus of this
book will be on parties operating in what the Economist Intelligence
Unit’s ‘Democracy index 2011’ labels as either ‘full democracies’ or
‘flawed democracies’ (EIU, 2011, pp. 4–6). The EIU defines a full
democracy as:
Countries in which not only basic political freedoms and civil liberties
are respected, but these will also tend to be underpinned by a political
culture conducive to the flourishing of democracy. The functioning of
government is satisfactory. Media are independent and diverse. There
is an effective system of checks and balances. The judiciary is inde-
pendent and judicial decisions are enforced. There are only limited
problems in the functioning of democracies. (EIU, 2011, p. 30)
They define a flawed democracy as countries that:
have free and fair elections and even if there are problems (such as
infringements on media freedom), basic civil liberties will be respected.
However, there are significant weaknesses in other aspects of democ-
racy, including problems in governance, an underdeveloped political
culture and low levels of political participation. (EIU, 2011, p. 30)
There are several reasons why this book will focus on parties in demo-
cratic countries. One is linked to how one might define a political party,
which will be explored further below. Another is the close link between
political parties and democracy. It was explored above how modern
democracy is party-democracy. There is in the modern world an intrinsic
link between political parties and democracy. We did see that some dicta-
torships are based around a political party. However, dictatorships do not
need party politics to ‘function’. Democracy does. Hence, we can under-
stand dictatorships without necessarily paying a great deal of attention to
party politics. We cannot understand democracies without understanding
party politics. Considering how important democracy as a form of gov-
ernment has become since the end of the Cold War, the need to under-
stand the party politics side of democracy is overwhelming.
A final reason is that a competitive party system, that is, where mul-
tiple parties are allowed to compete for power freely, unlike for example
communist single-party states, is one of the defining features of modern
democracy. Hence, party politics is at its most diverse and thus inter-
esting and worthy of study in a democracy. If there is only one party in
existence our task is very simple and therefore not that interesting. In a
system with several parties all jockeying for position and power, the task
is that much more complicated, and, therefore, also that much more
worthwhile.
This chapter will next explore the definition of what a political party
is. The chapter will then explore various aspects of party organisations,
before looking at the main roles performed by political parties in demo-
cratic states. It will then give an overview of the idea that political parties
are currently in a state of crisis. The chapter will end with an overview of
the contents of this book.
Defining ‘political party’
Because parties are to be found in every democratic country, and in a

good few undemocratic ones, as a group they are marked by a huge
variety. Parties are influenced by the environment they operate in and are
Introduction 5
as diverse as their host countries. For that reason it can often be difficult,
but also endlessly fascinating, to explore them as a group. Indeed, merely
defining what a party is can be problematic. Some are old well-estab-
lished institutions, with long histories and strongly hierarchical structure
where each member knows where they fit in the organisation and who is
in charge – largely because they can read about it in the party’s rules.
Examples of such parties would be the British, German and French
socialist parties. Others, like the Republicans and Democrats in the
United States have even longer histories, but are much more loosely
organised with much less in the way of a hierarchical structure or clear
lines of command. Yet others, such as List Pim Fortuyn in the
Netherlands and Silvio Berlusconi’s Forza Italia (Forward Italy) and Il
Popolo della Libertà (The People of Freedom) were set up as their foun-
ding leader’s personal political projects. Some parties, usually smaller
fringe parties, have a very clear political ideology were as others are more
amorphous and designed to offend as few people as possible. The que-
stion is to what extent we can come up with a definition that captures all
these very varied organisations. We usually know a political party when
we see it, but coming up with a clear definition is not easy considering
this great diversity.
Indeed, definitions of parties have varied and changed over time. In a
text first published in 1770 Edmund Burke defined a party as ‘a body of
men [sic] united, for promoting by their joint endeavours the national
interest, upon some particular principle in which they are all agreed’
(Burke, 1981, p. 317). It is no doubt true that most parties would argue
that what they do is in the national interest. However, this is not unique
to political parties. Arguably the Church of England and other state reli-
gion organisations would claim to be doing the same. In addition, the
national interest could also be said to be the focus of Political Action
Committees (PAC) in the United States. A PAC will spend money in
support of for example a particular candidate, but will not be running
candidates itself. They do so with a view to promote a particular cause or
principle, but are by their very definition not political parties. Their
defined role in US legislation is to raise and spend money in the name of
a cause. Any number of campaign organisations, such as Amnesty
International and Greenpeace also campaign on very specific principles,
but we would not regard them as political parties. In short, ‘a body of
men’, that is, an organisation of some sort, pursuing a particular prin-
ciple is not something that is unique to political parties.
Burke’s definition suggests political parties are linked to political prin-
ciples, something which will be explored in Chapter 4. These principles
are sometimes seen as being derived from underlying conflicts in society –
conflicts in which the interests of distinct groups are seen to clash. In the
classic left–right view of ideology, these groups are defined as ‘classes’,

and, according to some, parties are the organisational expression of this
class conflict. So, Berntson (1974, p. 16) defines a party as an organisa-
tion that represents a social class, or an alliance of social classes, in the
struggle against other social classes for control over the actions of the
state. However, as with having an ideology, the defence of the interests of
particular groups in society is not unique to political parties. Trade
unions and numerous other interest organisations would also be covered
by that definition.
However, Berntson may still have part of the answer – the part related
to fighting to gain control over the actions of the state. One often cited
definition of a party comes from Downs who wrote that ‘in the broadest
sense, a political party is a coalition of men [sic] seeking to control the
governing apparatus by legal means [...] By legal means, we mean duly
constituted elections or legitimate influence’ (Downs, 1957, p. 24). The
problem with this definition is that Downs’ view of ‘legal means’ is suffi-
ciently broad to also include organisations that would not normally be
regarded as parties, such as trade unions. True, trade unions do not seek
to gain direct control over the levers of government in the sense that the
chief purpose of trade union officials is not to become members of parlia-
ment or government ministers. However part of a trade union’s purpose
is to achieve at least indirect control over government action through
‘legitimate influence’. Hence, including ‘legitimate influence’ still makes
Downs’ definition too broad. However, if we restrict ‘legitimate means’
to mean ‘duly constituted elections’ we may be close to a definition that
captures what is unique to political parties: fielding candidates in elec-
tions.
Indeed, this ‘electoral’ view of a political party is found elsewhere.
Epstein defines a party as ‘any group, however loosely organized, seeking
to elect governmental office holders under a given label’ (1967, p. 9).
However, not all parties have government ambition, at least in the short
or medium term, something which will be explored in Chapter 9. Hence,
we may want to replace ‘governmental office holders’ with ‘members of
legislative assemblies’. This is the definition which Bille settles on: ‘a
political party is a group of people who, under a common label, field can-
didates to popularly elected political assemblies and who typically are
capable of having candidates elected’ (1997, p. 17). It would probably
make most sense to exclude the last part (‘and who typically are capable
of having candidates elected’). Even the most electorally unsuccessful of
parties is still a party if it fields candidates to legislative assemblies,
because this is what makes parties unique: fielding candidates to popu-
larly elected legislative assemblies. Yes, parties do many other things than
that, but they share all of those things with many other types of organisa-
Introduction 7
tions. What sets parties apart from other political organisations is that
they field candidates for popularly elected assemblies. The definition of a
political party used here will therefore be a shortened version of what
Bille came up with:
A political party is a group of people who, under a common label,

fields candidates to popularly elected assemblies.
This definition, derived from several prominent scholars, also adds to the
reasons for focusing on democratic countries. If the central defining
feature of a political party is that it fields candidates in elections, then
there needs to be such elections in place for us to talk about a party in the
sense defined above. Clearly, this definition only makes sense in a demo-
cratic country. This would suggest that for example the Communist
Parties of Cuba, China and North Korea are not ‘true’ parties. There is
no currently existing definition of a party that could include ‘parties’ such
as those in control of Cuba, China and North Korea, but exclude organi-
sations that are usually not seen as parties, such as trade unions or other
interest organisations. It could perhaps be argued that if an organisation
calls itself as a party, then we should see it as a party. However, there are
organisations, such as the Indian National Congress and the Union pour
un Mouvement Populaire (Union for a Popular Movement) in France
which do not have party in the title, but are usually seen as parties.
Hence, for the purpose of this book, it seems to make most sense to
restrict ourselves to the definition given above, and focus on democratic
countries for the reasons discussed.
The three faces of party organisations
Now that we know how to identify a party and distinguish it from other
political organisations it is necessary to consider what a party looks like
as an organisation. How are they organised and what elements are they
made up of internally?
As noted earlier, political parties are endlessly varied. Each party will
have found its own response to its political circumstances, which in turn
will be different for each country and for each party within that country.
However, it is still possible to make some generalisations about the struc-
ture of political parties. Very generally it is possible to argue that a party
is made up of three ‘faces’ (Katz and Mair, 1993).
The first is what is referred to as the party in public office. This
element of the party consists of its publically elected officials (e.g., parlia-
mentarians) and members of the government (i.e., ministers). Then there
is the party on the ground. This is basically the extra-parliamentary mem-

bership organisation. This part could also be expanded to include regular,
but non-member, supporters. Finally, there is the party in central office.
This is the party’s central co-ordinating bodies. Its leading officials are
traditionally elected by the mass membership and it is staffed by fulltime
professionals.
The concept of three faces has a number of implications for how we
analyse political parties. First, it suggests that viewing a political party as
a unitary actor, akin to an individual human being, is misleading. Talking
about Party X doing something does not accurately reflect what is really
going on. What it actually means is that someone within Party X has
taken a decision about a particular course of action – a decision which
may or may not be supported by other people in the party. So, if we see a
headline along the lines of ‘The Moderate Party ignored its members and
proposed cuts to the defence budget’ this will be a massive simplification
of events. A party does not ‘ignore its members’ – the members are part
and parcel of the party. In addition, ‘the members’ are unlikely to be a
homogenous mass with identical views on any given issue. What would
be more accurate would be something along the lines of: ‘Moderate Party
leader Birgitte Nyborg and her allies in the party’s national executive
decided to propose cuts to the defence budget against the wishes of her
opponents on the national executive and their supporters among the local
activists’. ‘The party ignored its members’ is clearly much simpler, but not
very accurate. A party is an organisation made up of people with dif-
ferent levels of influence; with more or less divergent notions about what
the party should be doing; and with different levels of commitment to the
party.
In addition, the three faces of a party can and often do have different
priorities and short- to medium-term goals (even if they agree on long-
term goals – that is, the party’s basic ideological foundations). It could be
argued that because the party in public office owes its existence to the
electorate, this face of a party is likely to focus on satisfying short-term
voter demands. The party on the ground on the other hand does not have
the perks that come with public office (a salary, prestige, an office and
staff) and is, therefore, less concerned with electoral expedience. This face
of the party is motivated by ideological conviction and will therefore be
more radical than the party in public office. The extent to which this is
the case is open to debate and will be explored further in see Chapter 7.
However, it does illustrate that seeing a party as a single unitary actor
with a single purpose can be problematic.
Finally, the idea that different elements in the party may have diver-
gent and potentially even contradictory goals raises the question of who
decide which goals to pursue. Which of the three faces is in control of the
Introduction 9
party? The most popular idea is that the balance of power between the
three faces has changed over the years. The argument is that the party in
central office and the party on the ground have become subservient to the
party in public office, especially in the context of policy making. This did
not happen quietly or without a great deal of, to some extent ongoing,
resistance from the party in central office and in particular the party on
the ground, but is a point of view which is broadly accepted. This change
has been particularly profound in the traditional mass-membership party
(see Chapter 3). In this kind of party, the party in central office was there
to organise and also represent the party on the ground. However, over
time the party in central office increasingly became the tool of the party
in public office, used to manage and control the party on the ground. The
result is that, not only has the party in public office largely taken control
of the organisation, but the party as a whole is increasingly epitomised by
the party in public office. What this means is that when commentators
and a great many academics write that Party X has done something or
other, what they are actually saying is that the party in public office, or
even just the leader(ship) of the party in public office, has taken a deci-
sion on something. In short a party is a multifaceted organisation which
will act with greater or lesser degrees of unity. Hence, if we are to truly
understand party behaviour we must do so keeping in mind that parties
are often internally divided to a greater or lesser extent.
The structure of parties
However, the idea of the three faces of political parties only gives us a
very general view of how parties are structured internally. It tells use very
little about how these elements operate in practice. Saying something
general about how parties operate is very difficult. The practical day-to-
day running of a party is where the diversity of political parties really
comes into play. Parties have found their own answers to how to organise
and these answers will be affected by the history of a party, the people in
it and the national context a party finds itself in.
What we can say is that parties tend to be strongly affected by a
country’s electoral system. At the local level most party leaders will try to
ensure that there is a local branch in every electoral district to organise
and mobilise members and supporters, at the very least at election time.
Hence, at a very basic level, the number of branches a party has will be
partly affected by how many electoral districts a country is divided into.
Parties will vary greatly in the extent to which party leaders encourage
members and supporters to be active in the periods between elections.
Some party leaders, especially in the United States, prefer local activity to
be at a minimum when there is not an election on, so as avoid local inter-

ference with the work of the party in public office. This is why parties in
the United States have been referred to as ‘empty vessels’ (Katz and
Kolodny, 1994). Traditionally in the United States only the party in
public office existed continuously between elections. The two other faces
would to a large extent be built up as an election came closer, and then be
allowed to fade away afterwards. European parties have usually put far
more emphasis on the continuous existence of all three faces in-between
elections.
However, regardless of whether the local elements of a party are tem-
porary or permanent, they will be based around the electoral districts of a
country. One key exception to this pattern was the old Communist
Parties from before 1989. Many Communist Parties were organised
around ‘cells’. Cells were not necessarily based on geography, but rather
on an occupational basis. The cell ‘unites all party members who work at
the same place. There are factory, workshop, shop, office and administra-
tive cells’ (Duverger, 1964, p. 26). This fits with the idea that, at least in
principle, Communist parties were less oriented towards the electoral
arena. Their main purpose was not so much to enter parliament and
potentially government, and more the complete overthrow of the existing
order, through uniting the proletariat, or working class, against their
oppressors in the bourgeoisie. However, the reality of working in a demo-
cratic political system caused a number of Western European communist
parties to orientate themselves increasingly toward the electoral structure
of the country.
How parties are structured above the local level will vary enormously
from party to party, even within a single country. At the local level all the
parties in a country will have to conform to the structure of the electoral
system of that country, and will therefore vary less between them (but
will obviously vary from country to country as each country will have
different electoral systems). So for example, the number and size of elec-
toral districts or constituencies will be decided by the national govern-
ment. If the parties of a country want to be represented in all local
electoral districts (which is typically the case) the number of local
branches will be decided for them. However, how a party then organises
itself above the local level will be decided by each party individually.
Hence, the following should be seen as a very general summary of
what the trends are, and there will be a great deal of divergence from
these trends. The most comprehensive study of party organisation can be
found in Katz and Mair’s (1992) edited book Party Organisations: A
Data Handbook. While that book is from the early 1990s, parties are
somewhat conservative organisations. This means that they tend to
change slowly, and the lessons in that book remain valid today.
Introduction 11
There are a number of constituent bodies that can be found in most

political parties. Typically there will be some form of national conference
where local branches are represented. These are often major events in a
party’s calendar and will attract sometimes 100s or 1,000s of delegates
from local branches, as well as party leaders and the party’s members of
legislative assemblies (such as members of parliament). The exact role of
this conference varies, but at least formally speaking it is often referred to
as the party’s highest authority. What that means in practice is frequently
left unclear, and the actual powers of the conference tend to be limited.
Nevertheless, the party conference will have great significance. Many
parties hold up such delegates’ conferences as an example of their com-
mitment to democracy, not only nationally in their country, but internally
in their organisation as well. It also tends to be the only place where rep-
resentatives from all three faces of a party will come together in a single
physical space. On a day-to-day basis, the people belonging to each of a
party’s three faces, will be spread out across the country and will not reg-
ularly come across each other. The party conference therefore has an
important unifying function – it is the embodiment of the party as a
single unified entity, which is part of the reason why most parties have
such events, even if their actual power is limited.
The frequency of the conference varies from party to party. Often it is
annual, but in some parties it can be several years apart: for example:
• The British Labour Party has its conference every year.

• The Danish Social Democrat’s ‘Congress’ is every four years, but they
then have ‘themed conferences’ in between.
• Both the Republicans and the Democrats in the United States have
‘conventions’ every four years.
• The New Patriotic Party of Ghana has an ‘Annual Delegates
Conference’.
• The Bharatiya Janata Party (BJP) of India has a ‘Plenary Session’ at
least once every three years.
To deal with matters in between the main party conference, parties tend
to have some form of national committee or executive. The exact powers
and membership will vary, but it is often the case that the national com-
mittee is seen as the highest authority after the party conference.
Membership is often made up of representatives from the different parts
of the party: that is, the parliamentary party, local and/or regional
branches, local and regional government politicians and sometimes other
affiliated organisations such as trade unions. In some parties there will be
a national committee which meets every few months, as well as a smaller
executive which meets on a more regular basis. Regardless of the exact
nature of the national committee, its general role is to provide leadership

on a more detailed ‘day-to-day’, or at least ‘month-to-month’, level than
is possible for the party conference.
In addition to the purely geographical and hierarchical structure of a
party there is the party in public office. Most parties will organise their
publically elected legislators into party groups locally, regionally and
nationally. These groups exist alongside the geographical structure of the
party, but will often have a guaranteed level of representation at the party
conference, as well as national, regional and local committees.
There is certainly a huge level of variety within these general trends.
However, very broadly speaking, most parties will at least have the ele-
ments outlined. It makes sense to have local branches organised
according to electoral districts, and any party will need some form of
national, regional and local coordinating committees. These are simply
organisational necessities. Strictly speaking one could do away with the
party conference, but many parties find it useful to have a relatively fre-
quent gathering of representatives from across the organisation. This can
add democratic legitimacy to the party and provide a forum for net-
working and sharing of good practice between regional and local offi-
cials. Ultimately, if nothing else, it is an opportunity for the leading
activists in the party to get together with the national party ‘celebrities’
such as notable parliamentarians and ministers. This may seem trivial,
but this author’s own experience at the annual conference of several
parties suggests that the social side is a significant part of why delegates
go to these events. In the final reckoning it can be a reward for all the
hard work put into running local branches and election campaigns.
However, while it is true that above the local level parties tend to vary
significantly in the way they organise themselves, there are still dis-
cernible patterns. These patterns have been linked to the ideology of a
party.
According to Maurice Duverger the organisation of socialist parties
tend to be more formalised than conservative parties. He argues that in
socialist parties, the relationships between the different elements of a
party, their rights and responsibilities, are usually much more detailed
and formalised, or ‘articulated’, than in conservative parties. Hence, a
socialist party is much more likely than a conservative party to have a
detailed rulebook which is adhered to relatively strictly. This has been
exemplified in this author’s own experience. No activist in the British
Labour Party would be unfamiliar with the party rulebook. They might
not know it in intimate detail, although many do, but they would cer-
tainly know of its existence and importance. By contrast, this author has
come across members of the British Conservative Party, including a par-
liamentary candidate, who did not know the party had a rulebook. This
Introduction 13
may at first not seem to matter much: why should one care about how
detailed a party’s rulebook is?
Duverger argues that this is important. One key reason why it is
important is that parties that are relatively informal, or weakly articu-
lated, are inherently undemocratic. If, say, the power of the leader is left
vague and informal it is difficult to accuse the leader of overstepping his
or her authority. Similarly, if the rights of grassroots members are unspec-
ified how can they demand their due influence, when that due is unclear?
On the other hand, a very formalised party with a high degree of organi-
sational articulation is not inherently democratic. A party can be articu-
lated in a democratic or an undemocratic direction. However, because the
rules are very detailed and everybody’s rights and responsibilities (or the
lack thereof) are clear and unambiguous it is easy to tell whether a party
is democratic or not. Many communist parties, especially those with
strong links to the Communist Party of the Soviet Union, tended to be
very top led through a system known as ‘democratic centralism’.
Officially, democratic centralism meant that there would be open debate
about a topic before a vote was taken. Once a decision had been made,
all members were expected to stick to that decision religiously with no
dissent expressed. In effect it often meant that the leadership was in
control of the organisation. The leadership issues instructions and the
members followed. The party’s hierarchical nature was very clear from
the rulebook, so everybody knew their place.
Compare this to the British Conservative Party which only acquired a
formal rulebook after William Hague became leader in 1997, and even
then it seems to have been largely ignored. The Conservative Party has
traditionally been very top led. The members had only a very marginal
influence on the work of the parliamentary party, and until the election
of Ian Duncan Smith in 2001, had no influence on the selection of the
party leader. Indeed, for much of the party’s history the leader was not so
much selected as ‘emerged’ out of informal deals and compromises
between senior party figures. So, the Conservative Party is not very dem-
ocratic, but the lack of internal democracy is achieved in a different, and
informal, way compared to the highly articulated communist parties of
old.
In short, the more detailed the rulebook the easier it is to tell who is in
control. The vaguer it is, the more the leadership is able to assert their
control through informal means and the inherent power that comes with
having access to the resources associated with high office. In informal
organisations, members have few tools with which to temper that power.
As a general trend, left wing parties have detailed rulebooks, and right
wing parties do not.
The role of parties in democratic states
Now that we know more about what parties are and what they look like,
it is worth considering what they do. As was argued at the beginning of
this chapter, parties are extremely important in the running of democratic
states. The reason for this importance is the breadth and depth of the
roles that parties fulfil.
Perhaps the most important role of political parties is that they create
‘linkage’ between governments and voters.
The political party is the one agency that can claim to have as its very
raison d’être the creation of an entire linkage chain, a chain of connec-
tions that runs from the voters through the candidates and the elec-
toral process to the officials of government. (Lawson, 1988, p. 16)
Linkage is a very broad term, but Lawson is probably the writer who is
most well known for having dealt with this subject in the context of
political parties. Lawson argues that there are several kinds of linkage a
party can engage in.
• Participatory linkage: is focused on giving citizens a way of partici-

pating actively in the governing of the country beyond merely voting
in elections. This kind of linkage will be notable for a strong party on
the ground and a high level of intra-party democracy.
• Electoral linkage: here the leadership keeps a tight control over the
elected representatives. Elected representatives are expected to adhere
to the views of the party’s supporters. However, those views are inde-
pendently, and more or less correctly, determined by party leaders.
The role of party members is not to participate in policy making, but
only to campaign at election time. Hence, the only real opportunity
for supporters to express their views and participate is at election
time. This has also been referred to as ‘responsive’ linkage (Rommele,
Farrell and Ignazi, 2005, p. vii).
• Clientelistic linkage: this is where the focus is on ‘buying’ votes in
exchange for favours. That is, in return for voting for a party a con-
stituency can expect to have certain state investments made in local
infrastructure.
• Directive linkage: in this case the party is used as a means through
which those in power can control the behaviour of citizens. This can
be through education (indoctrination), coercion or both.
A significant part of the debate in the context of linkage is about the

extent to which parties provide each of these kinds of linkage and also
Introduction 15
which kind they should provide. That is, the debate is both descriptive
and prescriptive. The evidence on the descriptive side is very varied and
often provides evidence of all four types of linkage at various times and
places. The prescriptive side, that is, what kind of linkage parties should
provide often argues that parties ought to provide forms of linkage that
foster more active and inclusive forms of democracy.
Lawson argues that depending on what kinds of linkage a party pro-
vides it may be working either for or against the development of a
healthy democracy. Participatory linkage is always good for democracy.
Electoral/responsive linkage may or may not contribute to democracy. It
very much depends on how accurate the responsiveness is. Clientelistic
linkage certainly contribute towards the maintenance of power for a
party, but depending on how that power is used, may work for or against
democracy. Directive linkage using education will again help a party
maintain power, but that can be good or bad for democracy depending
on its use. Coercion is always bad for democracy.
However, even though the idea of linkage has attracted a great deal of
attention the roles that parties fulfil go beyond that. Pedersen (1989, p.
268) summarises the roles of parties as follows:
• Mobilisation and socialisation: parties mobilise citizens, not just to

vote, but also to become active in politics. They identify and recruit
members and show those members what being active in politics is
about. That is, they socialise them into the role of political activists,
which may lead to positions of responsibility in the organisation.
• Selection and training of political leaders. Through mobilising citizens
and recruiting them into the party structure they also work to identify
potential candidates for high office. These people will often receive
training and support, both as candidates and as elected legislators. By
doing so parties also act as gatekeepers to high office. It is very rare
for anyone to make it very far in politics without the support of a
party. Hence, parties control who gets the chance to obtain positions
of power in the political structure of a country.
• Interest articulation: all parties, to greater or lesser extent, work to
represent the interests of various groups in society.
• Political programme formulation: one of the key purposes of parties
is to create, present, and if they get the chance, carry out a pro-
gramme of political action. In doing so they prioritise between com-
peting demands in society, and to some extent present a vision for
the good society. Political programmes will often contain informa-
tion about what kind of society the party would like to see built, in
addition to the more day-to-day concern of allocating scarce
resources.
The crisis of parties
It should be clear by now that political parties are highly complex organi-
sations that fulfil a wide range of crucial tasks in democratic societies.
However, despite how important they are, parties have never been very
popular. The modern manifestation of this unpopularity is the notion that
political parties are in crisis.
Parties have a long history, with the deepest roots to be found in
Western Europe and the United States. For as long as parties have existed
they have been viewed by many with deep suspicion, often tipping over
into outright hostility. The problem is neatly encapsulated by the title of
Scarrow’s (2006) overview of the rise of political parties: ‘the unwanted
emergence of party based politics’. As early as 1835 de Tocqueville wrote
that ‘parties are an evil inherent in free governments’ (see White, 2006, p.
7). The antipathy towards political parties has deep roots in political phi-
losophy. One prime example of this is Rousseau’s work on the ‘general
will’. According to Rousseau the ‘general will’ is what should be the
guide for collective action. The general will can be reached when every
individual in a society, without any kind of outside interference, considers
what the right way forward is. The general will must be the collective
opinion of all individuals in society, acting as individuals: ‘If, when the
people, being furnished with adequate information, held its deliberations,
the citizens had no communication one with another, the grand total of
the small differences would always give the general will, and the decision
will always be good’ (Rousseau, 1966, p. 23). However, it is important
that every individual acts independently, and without reference to any
sub-group within society. Any kind of faction will undermine the general
will:
when factions arise [...] the differences become less numerous and give
a less general result [...] It is therefore essential, if the general will is to
be able to express itself, that there be should be no partial society
within the State, and that each citizen should think only his own
thoughts. (Rousseau, 1966, p. 23)
While not often cited, the idea that parties undermine the collective will
and exacerbate divisions is widely held. Hence, from the very beginning
of their existence, parties were viewed as something negative, if also
unavoidable.
This negative view of political parties is continued in the first compre-
hensive study of political parties, Ostrogorski’s Democracy and the
Organization of Political Parties, first published in English in 1902.
Ostrogorski does not object to groupings of parliamentarians – that is,
Introduction 17
100
90
80
70
60
50
40
30
20
10
0
USA
France
GB
Italy
Spain
Canada
Japan
Mexico
South Africa
Australia
Norway
Sweden
Argentina
Poland
Switzerland
Brazil
Chile
India
Romania
Taiwan
Ukraine
Ghana
New Zealand
Germany
Figure 1.1 Confidence in political parties: Not very much/None at all (percentage)
Source: World Value survey 2005–8.
the party in public office. What he does object to is the existence of extra-
parliamentary organisations – that is, the party on the ground. Their
excessive control over members of parliament was alleged to undermine
the smooth running of a parliamentary democracy.
This antipathy towards political parties continues today. As we can see
from Figure 1.1 there is a widespread lack of confidence in political
parties in all the countries for which data is presented here. There are
clearly some variations from country to country; however, nowhere is
lack of confidence lower than 50 per cent. In some countries the lack of
confidence is almost total.
In addition to this outright hostility to political parties from both
scholars and the population of the democratic world at large, doubts
about the role of parties in modern politics is reflected in a growing field
of study: the crisis of political parties. The idea that political parties are
in crisis is not new, but it is widespread. Daalder (1992, p. 269) writes
that ‘we all talk about the crisis of party’. Mair (2003, p. 6) wrote: ‘That
political parties are in crisis, and potentially on the verge of serious
decline, is now more or less accepted wisdom among commentators
throughout the established democracies’. This was a theme he expanded
on three years later (Mair, 2006, p. 50), but with even more glum conclu-
sions than in the 2003 piece: ‘parties may be able to fill public office, but
having abandoned their representative role, they may no longer be able to
justify doing so’. The party crisis idea has evidently been a regularly
occurring theme in the literature for decades.
The idea that parties are in crisis is based on two broad issues. The
first is that parties across the democratic world are said to have seen a
significant decline in their membership organisations. The idea of parties
having a mass membership organisation is so closely intertwined with
views of what a party ought to look like that this decline is seen as a sign
that parties are in crisis. The issue of party members will be explored
further in Chapter 5.
The second element in the party crisis argument is based on claims that
parties are no longer fulfilling the roles they should be fulfilling in demo-
cratic societies. This is the essence of Mair’s argument. Leaders of polit-
ical parties may still have a key role in governing countries in the
democratic world. However, increasingly party leaders are withdrawing
into the world of government institutions and away from civil society.
Party leaders govern, but they no longer represent (Mair, 2006). This
echoes Lawson’s work on linkage. After having worked on the idea that a
key function of parties it to provide linkage between rulers and the ruled
she went on to argue that they are increasingly failing to do so effectively.
They are losing members, their top leaders are becoming increasingly iso-
lated from the rest of society and are failing to reflect and act on the
political demands of the voters.
However, while the idea of a party crisis and party failure may be
long standing and widespread, it is argued that a lot of the literature on
this topic is shot through with underlying normative assumptions
(Daalder, 1992). When authors talk of parties being in crisis they do so
on the basis of more or less subjective assumptions about what parties
ought to be doing. Early ‘party crisis’ ideas questioned the very legiti-
macy of parties (as outlined above) based on ideas about how democ-
racy ought to function. Later there were questions over certain kinds of
party – that is, a concern over the rise of fascist and communist parties.
The last of the ‘parties in crisis’ elements is the argument that democ-
racy has moved beyond the need for parties, that parties are now redun-
dant.
Daalder’s argument suggests that when people talk about parties
being in crisis, what they really mean is that parties are no longer living
up to some pre-set idea of what a democratic state should look like.
Hence, if we are to fully understand the claims that parties are in crisis
we would need to evaluate these underlying normative assumptions. In
essence, whether one agrees with the party crisis idea is less an objec-
tive evaluation, but a normative one based on what one thinks parties
ought to be doing. We will return to this theme in the conclusion to this
book.
Introduction 19
Outline of the book
The rest of the book will be divided into 10 chapters. Chapter 2 will deal
with party systems. Perhaps the most important factor parties have to
cope with in their political activities is other parties. The party system a
party operates in is a key environmental factor in our understanding of
party life. Hence, we will look first at party systems.
Chapter 3 will look at how parties have developed and, according to
some, moved through different stages from one party type to another.
This is a lively and expanding area of party politics, and new suggestions
for party types is a regular feature of the literature.
Chapter 4 will look at the issue of ideology. One of the key ideas of
what parties are is that they are vessels for an ideology. Parties are there
to promote and defend a particular view of society, a view derived from
underlying and long-standing divisions in society. Most of the literature
on ideology focuses on Europe, and we will consider the extent to which
‘traditional’ views of ideology also hold outside Europe.
The focus of Chapter 5 is party members. A number of questions will
be asked and answers sought, including why party leaders would want to
encourage the building or maintenance of a mass membership organisa-
tion; and why anyone would want to join a party. The chapter will also
look at the issue of what is happening to party membership numbers.
Chapter 6 will look at how candidates for public office are selected.
One issue the chapter will pay particular attention to is whether the
gender imbalance in most parliaments in the democratic world can be
explained by parties not selecting enough women candidates. Selection is
an important issue in party life since virtually no one gets into a legisla-
tive assembly without party endorsement. Parties therefore act as key
gatekeepers to power.
In Chapter 7 the book will look at policy making. The chapter will
cover issues such as: who are involved in policy making; who should,
normatively, be involved; and who wants to be involved.
Following that, Chapter 8 will examine the work that parties do in
election campaigning. The chapter will examine some of the arguments
about the ways in which parties’ campaigning activities have changed and
how those changes have affected party organisations.
The focus of Chapter 9 is the struggle of parties to get into govern-
ment. The chapter will look at whether all parties desire government
office and what might explain the decision either way. The chapter will
also look at the challenges a party faces once the decision to try to enter
government has been made.
Chapter 10 will look at the internationalisation of party politics. This
is a topic that has a long history, but is also an area which has delivered
less than some people would have liked. The chapter will look at the
party Internationals, at Europarties and at the internationalisations of
campaign consultancy.
Finally, Chapter 11 will consider the future of party politics and
whether parties are in crisis.
Chapter 2
Party Systems
One of the main concerns of political parties in their day-to-day activities

in democratic countries is other political parties. The vast majority of a
party’s efforts are devoted to competing with other parties for power and
influence. However, depending on the details of the political system of a
country other political parties can also be potential, usually short term,
partners against common opponents. In short, one of the key issues we
need to understand when looking at political parties is what patterns
exist in the interaction between them. The politics of a country is clearly
going to be very different if there are only two parties competing, com-
pared to there being for example a dozen roughly equal parties struggling
for power. One of the main tools used to analyse how parties interact is
the concept of party systems.
When we talk about the party systems we can mean one of two things.
One is simply the fact that any country where parties are a key part of
the political system can be said to have a party system. In short, saying a
country has a party system could simply mean that parties exist.
However, in a more systematic sense the study of party systems is the
effort to come up with concepts that will help to structure what is in
practice a very complex reality. Each and every democratic country can
be said to have a party system that is unique to it. After all, there are a
near infinite number of permutations when we combine the possible
number of parties (from one to 100s) and their relative size. Hence,
‘party system’ also refers to specific categories of party systems which can
be used to analyse how party politics works and changes across different
countries. What these categories are will be explored later in the chapter.
The importance of understanding party systems
The idea of party systems is a key element in understanding a country’s

politics. One of the central elements in a party’s environment, and one of
the main factors determining its behaviour, is other parties in the political
system – how many there are, how strong they are and what (or whom)
they represent. Understanding party systems is, therefore, an important
part of understanding political parties. The number and strength of com-
21
petitors will affect the behaviour of a party considerably. If there are only
two parties in a system there is only one ‘stream of interaction’ to con-
sider. This is the case in the United States where the only two noteworthy
competitors are the Republicans and the Democrats (see Figure 2.1). In a
slightly simplified view of the UK party system there are three streams of
interaction. If we look at the case of Denmark, which after the 2011 elec-
tion had eight parties in parliament, the tactical situation faced by each
party is far more complicated. Each party has to consider seven rivals
when deciding on what action to take. In short, the more parties there are
the more complicated is the political battlefield a party has to cope with.
The number of parties present in a parliament will also have a signifi-
cant impact on government formation. In a two-party system one party
will by definition get a majority of the seats. However, in a party system
with more than two parties this is no-longer automatically the case. The
larger the number of parties in a parliament, the bigger the chance that
no party will manage to achieve an overall majority of the seats. This
means that a coalition government may be necessary. This will be
explored further in Chapter 9.
In addition, the extent to which seats are spread out between several
parties will have an impact on how many parties will have to be involved
in a coalition. Compare the situation in the United Kingdom and
Denmark (Figure 2.2). In Denmark after the 2011 election no single party
came anywhere near an overall majority. The two biggest parties, the
Liberals and the Social Democrats, had no interest in forming a govern-
ment together. Indeed, no party from either the left-of-centre or the right-
of-centre ‘blocks’ would join in a government with parties from the other
side. Collectively the parties in the left-of-centre block had a majority.
However, the Red Green Alliance is regarded as being too far to the left
to be a realistic government partner. The result therefore was a minority
coalition of the Socialists, the Social Democrats and the Social Liberals –
with the Red Green Alliance providing support from outside the govern-
ment. By contrast the 2010 election in the United Kingdom was far
simpler. While a very unusual result for the United Kingdom in that no
party achieved an overall majority, the Conservative Party was close and
only needed the support of the Liberal Democrats to be able to form a
majority coalition.
This indicates another reason why party systems are important:
understanding a party system will help us understand how fragmented
power is in a country. In a two-party system power will be far more con-
centrated (in whichever party has a majority) than in a multi-party
system. This will have a major impact on the nature of politics in a given
country. In a country where politics is more or less dominated by two
parties (such as the United States and more often than not the United
Party Systems 23
USA
Republicans Democrats
UK
Labour Party Conservative

Party
Liberal
Democrats
Denmark
Red-Green Socialist
Alliance People’s
Danish Party
People’s
Party
Social
Democrats
Liberal Party
Social
Conservative Liberals
People’s
Party
Liberal
Alliance
Figure 2.1 Streams of interaction

Denmark (2011)
Red Block Blue Block
30
⎫
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎬
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎭
⎫
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎬
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎪
⎭
25
Percentage of seats
20
15
10
0
Red Green Socialist Social Social Liberal Conservative Liberal Danish
Alliance People’s Democrats Liberals Alliance Party Party People’s
Party Party
UK (2010)
50
45
40
Percentage of seats
35
30
25
20
15
10
5
0
Labour Party Liberal Democrats Conservatives Others
Figure 2.2 Election results in Denmark (2011) and the UK (2010)
Sources: Statistics Denmark (2011) www.dst.dk/da/Statistik/emner/valg/folketingsvalg.aspx BBC

(2010) http://news.bbc.co.uk/1/shared/election2010/results/.
Kingdom) politics will be marked by confrontation and a lack of com-

promise. If a single governing party has an overall majority there is no
need to compromise with any other party in the system. By contrast, in a
situation where many parties have a reasonably chance of either entering
government or influencing the legislative programme of a minority gov-
ernment, as is the case on most of the countries on the European conti-
nent, compromise and cooperation will, by necessity, be far more
common. The minority government created in Denmark after 2011
clearly needs to be able to compromise in a way that a single-party
majority government would not.
Party Systems 25
Finally, and as an extension of what was said above, in comparative

politics the concept of party systems helps us simplify the complexities of
party politics. Comparing and understanding party politics in several
countries is difficult without some kind of ordering principle. Using the
idea of party systems will immediately tell us something about the differ-
ences in the politics of whatever countries we are interested in. The mere
fact that we can assign countries to different party system categories will
help give us a starting point for understanding what the similarities and
differences between these political systems are.
Early approaches to party systems
The time has now come to look at some of the most important
approaches to classifying party systems.
Duverger was the first scholar to seriously attempt to classify party
systems. His categorisation schema was simple and also somewhat crude.
Duverger limited himself to three categories: ‘the single-party, the Anglo-
Saxon two-party, and the multi-party system’ (Duverger, 1964, p. 203).
While this is certainly a start it is also reasonably obvious that such a
simple schema is inadequate for a number of reasons. The single-party
system is perhaps defensible in that Duverger sees this kind of system as a
non-competitive system. In other words, some form of authoritarian
regime where all other parties are simply banned. It is, therefore, fairly
easy to identify. The idea of a two-party system seems fairly simple on the
surface. However, the exact dividing line between a two-party system and
a multi-party system is not entirely clear. So, while the United States is
very clearly a two-party system, the United Kingdom has often been
regarded as being a two-party system, but there are several significant
players in the system beyond the main two parties, a factor that became
especially evident after the 2010 election when the third largest party, the
Liberal Democrats, joined the Conservative Party in government. Indeed,
Duverger himself acknowledged this problem.
However, perhaps the biggest problem with Duverger’s schema is how
he deals with the multiparty system. Duverger virtually abandons any
attempt to say much about ‘more-than-two’ party systems. All he has to
say is that they are somewhat complicated to deal with: ‘The typology of
the multi-party system is difficult to establish: innumerable varieties can
be imagined ranging from three parties to infinity, and within each
variety innumerable patterns and shades of difference are possible’
(Duverger, 1964, p. 229). He is left with merely describing a number of
specific examples. It therefore seems that while Duverger’s approach to
party systems has been called ‘pioneering’ (Blau, 2008, p. 169) this is
mainly because he was the first to deal with the issue of party systems in
a reasonably systematic way. Nevertheless, while Duverger’s schema of
party systems may have been somewhat underdeveloped there are still
two issues in his work on this that are worth following up on: his view of
the duality of politics and the impact of electoral systems on party
systems.
According to Duverger, politics is usually about a duality of choices –
or put differently, about a choice between two options. There may not
always be a ‘duality of parties’, but there is always a ‘duality of tenden-
cies’ (Duverger, 1964, p. 215). This means that for Duverger ‘the centre
does not exist in politics. The term “centre” is applied to the geometrical
spot where the moderates of opposing tendencies meet: the moderates of
the Right and the moderates of the Left’ (Duverger, 1964, p. 215). Based
on that Duverger argues that two-party systems are superior to multi-
party systems since the two-party system ‘seems to correspond to the
nature of things’ (Duverger, 1964, p. 215). However, if two-party’ism is
natural and corresponds to the nature of politics, that does somewhat
leave the questions of why multi-party systems exist and indeed why they
are so prolific. Duverger (1964, pp. 230–3) suggests a number of reasons,
but the most significant factor he proposed in the creation of multi-party
systems is the nature of the electoral system. This issue will be considered
later in this chapter.
Others have tried to overcome the inadequacies of Duverger’s categori-
sation. Most notably Jean Blondel (1968) who investigates the voting
pattern in 19 democratic countries and identifies six types of party
systems.
1 The two-party system such as that found in the United States.

2 A two-and-a-half party system with two major parties on the left and
the right of centre, and a smaller centre party.
3 A two-and-a-half party system where the smaller party is on the left.
4 A multi-party system where no single party gets more than about 25
per cent of the vote.
5 A multi-party system where a dominant left-wing party is faced by a
divided right-wing.
6 A multi-party system where a dominant right-wing party is faced by a
divided left-wing.
Blondel’s approach is clearly much more fine grained that Duverger’s. His
idea of a two-and-a-half party system appears useful in dealing with the
problem of countries where there are two major parties, so not a multi-
party system, but where there is also a clear third ‘presence’. The British
Liberal Democrats is a good example of this ‘half’ party.
Party Systems 27
However, the problem with Blondel’s is that it is based very specifically

on the voting patterns of 19 countries during the period 1945–66. This is
therefore not necessarily a general categorisation which can be easily
applied outside of the specific context of those countries at that time.
Party system typologies
Both Duverger and Blondel are important models to mention because

they provided the groundwork for what has become the most widely
accepted schema for categorising party systems – that devised by Sartori.
One of Sartori’s key contributions was creating a system for counting
‘relevant’ parties.
The issue of counting may seem obvious. After all, Duverger and
Blondel had already talked about one, two and multi-party systems which
suggests some form of counting being used. However, according to
Sartori: ‘we are even incapable of deciding when one is one and when two
is two’ (2005, p. 106). Counting all parties that exist in a country does not
make sense. Several hundred parties are registered with the UK Electoral
Commission, but only three of these have been of national significance,
with about half-a-dozen having some regional importance. Similarly, there
are more than two parties in the United States, but only really two that
‘count’. The problem then is to determine which parties ‘count’.
Looking at electoral strength (which was Blondel’s approach) and
deciding some cut-off point is not only arbitrary, but also inadequate
since electoral strength alone does not always tell us very much about the
‘relevance’ of the party. The British Liberal Democrats have been
achieving electoral support of between 15 and 20 per cent, and the
Danish Social Liberals half that. However, the Danish Social Liberals
have experienced government participation, and legislative influence
when out of government, beyond anything achieved by the Liberal
Democrats after 1945 and before 2010. The German Free Democratic
Party has had much the same experience as the Danish Social Liberals:
modest electoral support, but significant government power. Between
1946 and 1992 The Italian Communist Party obtained between 22 and
34 per cent of the vote in general elections, but never achieved govern-
ment power. Clearly some other criterion than electoral support is needed
to understand which parties are worthwhile looking at. Sartori’s solution
is a twofold criterion for relevance: to be relevant, a party must have
either ‘coalition potential’ or ‘blackmail potential’.
Coalition potential is related to a party’s ability to influence who is in
government. A party should generally be seen as irrelevant if it remains
superfluous in that it is not needed to determine any feasible government
or coalition majority. On the other hand, a party, regardless of how

small, must be counted if it is in a position to participate in a coalition or
is the kingmaker between two competing options.
Blackmail potential is related to the fact that there are some parties
which, while big enough to secure a majority for one possible govern-
ment, are never asked to participate. Examples would include the Italian
Communist Party which for years achieved significant electoral success,
but was nevertheless excluded by the other parties. Such a party may still
count as relevant if ‘its existence, or appearance, affects the tactics of
party competition and particularly when it alters the direction of the
competition [...] of the government oriented parties’ (Sartori, 2005, p.
108). In short, Party A must be seen as relevant if parties B and C, which
normally compete with each other, change their behaviour to exclude
Party A from power. An example of this would be the 2002 presidential
elections in France. In France the president is elected over two rounds. In
the first round the full list of candidates are presented to the voters. The
two top candidates in the first round go through to the second round. In
the second round the candidate who gets the most votes will win. In 2002
the Front National leader Le Pen managed to get into second place ahead
of the Socialist Party candidate in the first round. In the second round the
voters then had a choice between a far right candidate and a centre-right
candidate. This meant that the normal competition between the main
leftwing block and the main rightwing block was suspended in order to
ensure that Le Pen was comprehensively beaten in the second round. The
Front National had no members of the parliament, and thus very little
real influence, but their presence changed the normal rivalry between the
centre-left and the centre-right.
It should be pointed out here that Sartori insists that these two criteria
for relevance or irrelevance are ‘postdictive’ rather than predictive
(Sartori, 2005, p. 108). That is, a party is only regarded as relevant if it
currently has or has recently had either coalition or blackmail potential.
A party which is in a position where it may in the future have either kind
of potential is not counted.
Based purely on the number of relevant parties there are seven classes
of party systems (Sartori, 2005, p. 110).
1 One party
2 Hegemonic party
3 Predominant party
4 Two-party
5 Limited pluralism
6 Extreme pluralism
7 Atomised
Party Systems 29
Segmentation
Party Party Party Party Party

A B C D E
Left Right
Polarization
Party Party Party Party Party

A B C D E
Left Right
Figure 2.3 Segmentation and polarisation
However, focusing purely on the number of parties in a system is not

enough in the case of limited and extreme pluralism. Sartori defines the
border between the two as being five to six parties. This is, he admits,
rather arbitrary and not necessarily very useful. What we need in addi-
tion to numbers is to include the issue of ideology. Sartori argues that ide-
ological distance is important, in particular whether the party system is
segmented or polarised (see Figure 2.3). A segmented party system is one
where ideologically the parties sit next to one another like pearls on a
string. Most modern democratic countries would be regarded as ideologi-
cally segmented. In contrast a polarised system is one where the parties
cluster around mutually exclusive and hostile ideological positions. One
key example of a polarised party system is the dying years of the German
Weimar Republic. Here there were a number of moderated centrist
parties, flanked on either side by Communists and National Socialists.
Based on the numerical criterion and the issue of ideology Sartori
modifies limited and extreme pluralism. If a multi-party system is domi-
nated by ideological segmentation it is termed moderate pluralism. If on
other hand it is dominated by ideological polarisation it is termed
polarised pluralism. In the final typology limited and extreme pluralism is
replaced by moderate and polarised pluralism. The party systems in
Sartori’s scheme are then as follows:
• One party
• Hegemonic party
• Predominant party
• Two-party
• Moderate pluralism
• Polarised pluralism
• Atomised
A ‘one party system’ is characterised by all power being concentrated in

one party. There is only one party as no other parties are allowed to
exist. This matches Duverger’s view of one-party systems as being iden-
tical with authoritarian regimes. Clear current examples are the
Communist regimes in the People’s Republic of China, North Korea and
Cuba. Historical examples include all the former Communist regimes in
Eastern Europe as well as Nazi Germany 1933–45 and Fascist Italy
1922–43.
The ‘hegemonic party system’ is similar to the one party system in that
one party dominates and that this dominance is maintained through non-
democratic means. However, unlike ‘pure’ one-partyism other parties are
not banned outright. At the same time they are only allowed to exist as
long as the power of the hegemonic party is not overly challenged.
Examples include Zimbabwe under ZANU PF. Other parties exist, most
notably the Movement for Democratic Change, but when they look like
they may threaten the dominance of the ZANU PF they come under
intense and often violent pressure. Arguable this pressure has lessened
somewhat since the aftermath of the 2008 elections, but Zimbabwe prob-
ably still falls into this category. Another example is Singapore. The
People’s Action Party (PAP) has been very firmly in control of
Singaporean politics since independence and although other parties are
not banned, they face strong PAP pressure in the form of libel suits and
censorship. The pressure the other parties face is not in the same category
as that experienced in Zimbabwe, but nor is the political system suffi-
ciently free to fall into the more democratic ‘predominant’ category (see
next paragraph).
In a ‘predominant party system’ one party consistently wins an overall
majority of the vote. Elections are generally free and fair, but for what-
ever reasons, one party dominates the electoral arena. From achieving
independence in 1947 until the late 1960s the Indian political system fell
squarely into this category. The Indian National Congress so dominated
Indian politics that it became known as the Congress System. Since the
1960s the Congress System has steadily disappeared, to be replaced by a
volatile multiparty system. For most of the post-war era until the early
1990s Japan also fell into this category with the Liberal Democratic Party
dominating the political scene. Several African countries, having gone
through democratisation in the early 1990s and avoided single or hege-
monic party rule, have also experienced predominance by one party. One
prime example of this is Botswana, one of Africa’s most successful coun-
tries both economically and politically. It is also the only country to have
Party Systems 31
held open, free and fair elections continuously since independence in

1966. However, despite the open and free elections the Botswana
Democratic Party (BDP) has held power since independence. Finally, it
could be argued that some parts of the United States have at various
times fallen into this category. As an example can be mentioned the City
of Baltimore which has had Democratic mayors uninterrupted since 1967
and the City Council is also dominated by the Democratic Party.
Typically all 14 council members and the council president are
Democrats.
According to Sartori the ‘two-party system’ poses ‘no problem in as
much as their power configuration is straightforward: Two parties
compete for an absolute majority that is within reach of either’ (Sartori,
2005, p. 112). The United States is a clear example of this system. The
Democrats and the Republicans dominate the political system at all levels
and regularly exchange power at the national level in both the legislative
and executive arenas. Other parties do exist, but their impact is close to
non-existent. The United Kingdom is also commonly placed in this cate-
gory. Government power has since 1945 been exchanged between the
Labour Party and the Conservative Party, with no other party having
come close to challenging this dual domination until 2010. However, if
we go beyond the national governmental level, this picture becomes sig-
nificantly murkier. However, as far as government power goes, the United
Kingdom was very much a two-party system from 1945 until 2010.
Both pluralism categories share the characteristics that no single party
is able to achieve an absolute majority. Where they differ is in their ideo-
logical makeup. Moderate pluralism is marked by ideological segmenta-
tion with each party filling a particular ideological space, but without any
parties that reject the legitimacy of the entire political system. All parties
are therefore pro-systemic and work to gain influence within the system
rather than change it completely. Several countries can be seen to fall into
this category including all the Scandinavian countries, Germany, Spain,
Israel and Ghana (one of the few relatively successful multi-party democ-
racies in Africa). It is true that large parties dominate in for example
Germany, France and Sweden, but their dominance is not usually suffi-
ciently great that they can marshal an overall majority on their own.
The ‘polarised pluralism’ party system represents what one might call
the down-side of multi-party systems. Eight key features characterise
polarised pluralism:
1 The existence of relevant anti-system parties, that is, parties that

dispute the legitimacy of the existing system of government, not just
the present government, and seek to change the whole structure
rather than merely take control of it.
2 Bi-lateral opposites, meaning the presence of extremist parties which

oppose the existing political system and each other with uncompro-
mising passion.
3 A triangular interaction between anti-systemic opposites at two
corners and pro-systemic parties at the third corner.
4 A polarisation of ideological beliefs, which in turn leads to:
5 Centrifugal competition pulling away from the centre.
6 An increased focus on ideological grandstanding and emphasis on
‘true believers’ over pragmatists.
7 Irresponsible opposition, exemplified by the fact that anti-systemic
parties do not have to present an alternative government programme
but merely oppose any and all initiatives from the pro-systemic
parties.
8 The politics of outbidding. Parties that have no intention of pre-
senting a government programme that is realisable within current
government structures can afford to ‘over-promise’ in their pro-
grammes. This in turn leads other parties to attempt to match the
unrealistic promises of the anti-systemic parties.
Not all these features need be present to make a system polarised, but the
more that are present the more polarised a system is. A quintessential
example of polarised pluralism is Weimar Germany (1919–33). It had
most of the characteristics of a polarised system, particularly the presence
of strong anti-systemic parties – Communists and Nazis – which created
a centrifugal pull away from the centre. The Nazis and Communists
attacked both each other and the pro-Weimar parties in a triangular com-
petition and were not concerned with obtaining control of the system,
but with replacing it completely, something the Nazis were eventually
successful in doing. Another example is France during the Fourth
Republic (1946–58). Again there were anti-systemic parties in the form of
the Communists and the Gaullists. Again one side was successful in over-
turning the entire government structure when the Gaullists ushered in the
Fifth Republic.
It is important to note that extreme pluralism (many parties) does not
necessarily mean polarised pluralism (intense ideological confrontation).
The number of parties is not the distinguishing feature. For example,
Denmark, Israel, the Netherlands and Switzerland all have enough parties
to have ‘extreme pluralism’, but are not ideologically polarised and
therefore remain in the ‘moderate pluralism’ category (Sartori, 2005, pp.
130–7). This is an important distinction to make. The number of parties
can change quite quickly, certainly from one election to another. Hence,
relying on the number of parties would mean that a country could shift
from one category to another relatively quickly, making our analysis of
Party Systems 33
the party politics of that country more complicated. By contrast, it

usually takes a fair bit of time for a country to change between polarised
and moderate ideological competition. Hence, using ideology in addition
to numbers will create a system of categorising the party systems of coun-
tries that will remain relatively stable over time.
The final category, the ‘atomised party system’, is a situation where
there are so many parties that counting becomes unimportant. Beyond a
certain threshold the number of parties ceases to matter. Sartori describes
it thus: ‘The atomised party system can be defined in the same way that
atomistic competition in economics, that is, as a situation where no one
firm has any noticeable effect on any other firm’ (Sartori, 2005, p. 111).
Sartori calls this a ‘residual’ category, which seems to suggest that it is
not a type which one will see very often in reality, but which nevertheless
needs to be included in the model to complete it. One possible example of
Atomised Pluralism may be India in the last few decades, that is after the
decline of INC and the rise and decline of the BJP. The INC and the BJP
to a certain extent dominate the political landscape. However, the most
noticeable element on the Indian party system since 1989 is the explosion
of small, often short-lived but constantly appearing, parties based on very
local concerns. As many of these are very local affairs they often hardly
notice similar localised parties in different parts of the system, which
seems to fit with an atomised system.
The main innovations in this typology compared with Duverger, apart
from the issue of counting and the inclusion of ideology, is that both the
single-party category and the multiparty category have been split up.
Compared with Duverger’s typology this more detailed examination of
multi-party systems is particularly valuable. Where Duverger more or less
gave up saying anything sensible about multi-party systems, Sartori’s
typology allows for a more sophisticated approach to analysing party
systems with more than two parties.
The effective number of parties
Sartori’s approach is without a doubt the most widely cited way to cate-
gorise party systems. It has admittedly been subjected to a fair few criti-
cisms, not least the fact that it can be difficult to actually judge which
parties have blackmail potential.
However, another, perhaps more serious problem, is that in Sartori’s
scheme a party is either relevant (counts as 1) or irrelevant (counts as 0)
– there is no middle ground (Blau, 2008; Lijphart, 1999). This is
perhaps too black and white for the shades of grey world that is poli-
tics. Some parties will inevitable be more relevant than others. It may
therefore be worthwhile to consider other ways of understanding party

systems.
The most popular alternative to Sartori is to use ‘the effective number
of parties’. In Gallagher and Mitchell’s (2008, p. 598) words ‘this
measure was devised [...] as an attempt to summarise the degree of frag-
mentation of a party system.’ In other words the ‘effective number of
parties’ is not a way of calculating exactly how many parties there are,
but a way of understanding the extent to which votes and seats are con-
centrated in a few parties or spread out among a larger number. The
effective number of parties is typically calculated for seats and votes. The
details of how the effective number of parties is calculated can be found
Gallagher and Marsh (2008, p. 598) and can briefly be described as
follows:
• Calculate each party’s proportion of the vote or seats depending on

which one you are interested in.
• Square each of these values.
• Add all of these values together to get a sum of the squares.
• Divide 1 by the sum of the squares.
The background of this process can be found in Laakso and Taagepera

(1979), but in essence, the higher the number is the more fragmented a
party system is, either in terms of how votes are distributed or how seats
are distributed. Finding the value of the effective number of parties can
tell us a number of things about a country.
Figure 2.4 shows the effective number of electoral parties (ENEP – that
is, based on how people have voted) and the effective number of parlia-
mentary parties (ENPP – that is, based on how seats have been distrib-
uted) in the United Kingdom and New Zealand. In the United Kingdom
we can see the effect of using an electoral system that is not designed to
be proportional. The voting pattern has clearly become far more frag-
mented since the 1940s as can be seen in the steady increase in ENEP.
However, because of the electoral system this has not been reflected in a
similar rise in ENPP. Hence, when looking at how people vote, the United
Kingdom has not been a two-party system for a long time. It is only the
electoral system which has in the past prevented this electoral multi-party
system from breaking through to the House of Commons.
Another interesting case is New Zealand. Until 1994 New Zealand
used the same disproportional electoral system as for the UK House of
Commons. In 1994 they opted for a more proportional electoral system
and the effect of this is immediately clear. A slowly fragmenting system at
the electoral level becomes even more fragmented. With a more propor-
tional electoral system in place, more parties had a chance of winning
Party Systems 35
UK
4
3.5
3
2.5
2 ENEP
1.5 ENPP
1
0.5
0
1940s 1950s 1960s 1970s 1980s 1990s 2000s
New Zealand
4
3.5
3
2.5
2 ENEP
1.5 ENPP
1
0.5
0
1940s 1950s 1960s 1970s 1980s 1990s 2000s
Figure 2.4 ENEP and ENPP in the UK and New Zealand
Source: Trinity College Dublin http://www.tcd.ie/Political_Science/staff/michael_gallagher/ElSystems/

index.php, (accessed 21 November 2013).
seats and therefore become more attractive options for voters. Before
1994 the parliamentary party system was stubbornly resistant to the
increasing fragmentation at the electoral level. However, after 1994 the
wider distribution of the vote was also reflected in a wider distribution of
seats. So, using the idea of the effective number of parties we are able to
calculate quantitatively what the impact of, for example, changing the
electoral system has been.
The effective number of parties approach is not without its flaws. For
example, a party which has only a very small impact on the effective
number of parties score may actually be very important politically. An
effective number of parties score may suggest that a country is dominated
by two parties, but gloss over the fact that a small centrist party is in the
position of ‘king maker’ and therefore wields enormous power. The effec-
tive number of parties can, therefore, not by itself tell us everything about
the distribution of power in a country. However, the approach does
provide useful insights into a country’s party system. The fact that we
have a quantifiable measure of party system fragmentation at several
levels of analysis (such as electoral and parliamentary) allows us to
explore changes within and across countries fairly easily.
Explaining party system variation
If we were to apply these different approaches to categorising party

systems, as indeed has been done frequently, we are left with the puzzle
of explaining why each country has the party system it does. This is par-
ticularly the case is we accept Duverger’s idea that duality or two sides,
and thus two parties, is the natural state in politics. What explains the
variation in party systems across countries?
Generally speaking there have been two broad approaches to
explaining the number of parties in a country. One is a sociological
explanation and is based on the number of ideological conflicts in a
society, what has been referred to as societal cleavages (see Chapter 4).
The other is technical in nature and focuses on political institutions; in
particular the type of electoral system in use. There is no doubt that the
second type of explanation, concentrating on the impact of the electoral
system, has been by far the most popular. Indeed. Riker (1982) argues
that this is a rare example of a true ‘scientific’ debate taking place in
social science – that is, a debate which is marked by the steady accumula-
tion of knowledge and where new advances are based on previous
advances. The investigation of the relationship between electoral systems
and party systems is one of the few examples in social science where
‘each one of the series of revisions is more general or more precise than
its predecessor’ (Riker, 1982, p. 753). It is, therefore, this area we will
focus on first before moving on to ideological factors in the creation of
party systems.
The concern with the impact of electoral systems is sometimes seen as
having originated with Duverger (1964). He makes two statements which
have since become the topic of intense and widespread debate. The first
statement, what Riker (1982) and indeed many others refer to as
Duverger’s law, is that: ‘the simple-majority single-ballot system favours
the two-party system’ (Duverger, 1964, p. 217). The second statement,
which according to Riker (1982, p. 754) has not been elevated to a law,
but merely a hypothesis, is that: ‘both the simple-majority system with
second ballot and proportional representation favour multi-partism’
Party Systems 37
(Duverger, 1964, p. 239). By way of explanation, the simple-majority

single-ballot system, also sometimes called first past the post or single
member plurality, uses single member seats, and whichever candidates
gets the most votes (not necessarily more than 50 per cent) wins. This is
the system used in the United Kingdom for election to the House of
Commons. The simple-majority system with second ballot usually works
by the two top candidates in the first round going through to a second
round, and the second round of voting determines the final winner. A
version of this system is used in France. Proportional representation (PR)
comes in a number of forms, but basically works to ensure that there is a
relatively strong relationship between the proportion of the vote a party
gets and the proportion of the seats it gets.
While this general idea, that single member plurality favours two-party
systems and two-round systems and proportional representation favour
multi-party systems, is most often associated with Duverger, he was
merely putting into formal and specific language an assumption with a
very long pedigree. The earliest documented explicit mention of this
effect comes from Henry Droop, the inventor of the Droop quota used in
the Single Transferable Vote electoral system (used for example in
Ireland), and is from 1869 – although it was also alluded to by John
Stuart Mill in 1861 (Riker, 1982, p. 756). Since then the idea that the
electoral system influenced the party system became increasingly com-
monplace. In 1901 it was part of the debate over Australia’s new elec-
toral system (Ashworth and Ashworth, 1901, pp. vii–viii; Riker, 1982, p.
756). Ramsey MacDonald, future leader of the UK Labour Party and the
party’s first Prime Minister, rejected proportional representation exactly
because he wanted to promote a two-party system where the Labour
Party would have the opportunity to form a single-party majority govern-
ment (MacDonald, 1909, p. 137).
The question then is what it is about the different electoral systems
that affect the party system of a given country. Broadly speaking the
causal factors are said to be either mechanical or psychological (Shugart,
2008, p. 30). The mechanical effects refer to the way votes are converted
into seats. In a system like the one used in the United Kingdom, larger
parties tend to get a proportion of the seats that is often significantly
higher than their proportion of the votes. Imagine a situation of three
major parties all competing across the country. Party A and Party B
usually get around 40 per cent of the vote each in every constituency. In
some constituencies Party A will beat Party B by a few percentage points,
and in some constituencies Party B will win. Party C gets the remaining
vote – that is around 20 per cent in each constituency. Its 20 per cent of
the vote will never beat the 40-ish per cent for the two other parties and
thus, despite winning 20 per cent of the vote it will get no seats. Party A
and B will get about 50 per cent of the seats each on 40 per cent of the
vote. This is in essence what happens in the United Kingdom, except that
the vote is much more unevenly distributed than in the above example
which allows smaller parties to concentrate enough votes in some con-
stituencies to still win some seats.
The psychological effects are the ways that voters and party leaders
react to these mechanical effects. The idea is that if voters expect their
vote to be wasted by voting for smaller parties they will be more inclined
to give their vote to a larger one.
Further, within any political party there will be extremists and moder-
ates – the ‘pacific and the fire-eaters’ (Duverger, 1964, p. 230). If these fail
to cooperate we may see a split which creates a new party. However, the
pacific and the fire eaters may have to work together when operating under
an electoral system where smaller parties are punished. By contrast, under
PR smaller parties find it easier to survive which may encourage splits
within existing parties (Duverger, 1964, p. 252) as minority groups strike
out on their own. In other words PR makes establishing broad umbrella
parties less appealing. Groups, which may under a plurality system have
joined together, will be tempted to form separate parties under PR. In a
country like the United Kingdom, 5 per cent of the vote across the country
is unlikely to get you any seats. In more proportional systems, such as in
Sweden or Germany, 5 per cent of the vote will yield something like 5 per
cent of the seats, enough to potentially make a difference in a multi-party
system. This may well explain the increase in electoral fragmentation
evident in New Zealand after 1994 as illustrated in Figure 2.4. With a
more proportional electoral system, voters have more incentive to consider
voting for smaller parties as these now have a far greater likelihood of
success; which in turn makes it more tempting for political activists to
strike out on their own, rather than join or remain in a larger party.
However, while the technical and psychological effects of the electoral
system certainly provide reasonable and believable arguments for how
different party systems came into being, there is a problem. This problem
could broadly be summarised as a chicken and egg problem. In other
words, what came first: the party system or the electoral system? It has
been argued that ‘it may be more accurate to conclude that [the electoral
system is] a result rather than a cause of the party system in a given
country’ (Grum, 1958, p. 375). Several studies have taken up this idea.
Taagepera (2003, p. 7) suggests a ‘causality flowing in the reverse direc-
tion, from the number of parties towards electoral rules’, which in turn
echoes an earlier claim by Lipson that ‘chronologically, as well as logi-
cally, the party system is prior to the electoral system’ (Lipson, 1964, p.
343). In a wide ranging analysis of 219 elections in 87 countries since the
nineteenth century Colomer (2005) concludes that:
Party Systems 39
Majoritarian rules induce the formation of two large parties, but two-
party configurations also maintain or choose majoritarian rules.
Proportional representation permits the development of multiple
parties, but multi-party systems also tend to establish or confirm pro-
portional representation rules. (p. 4, see also pp. 17–18)
The reason why it might be argued that the party system pre-dated the
electoral system in a given country is that the number of parties may in
fact be more easily explained by the number of societal cleavages or con-
flicts in society rather than the electoral system. Instead of seeing the elec-
toral system encouraging or restricting party formation, it could be
argued that the number of parties is influenced far more by underlying
societal conflicts. The idea is that parties grow out of and represent
groups in society sitting on opposing sides of social conflicts – such as the
classic class conflict between proletariat and bourgeois, or between
national and/or religious majorities and minorities. This would then
suggest that both party systems and electoral systems are affected by the
ways in which society is split politically in a given country. So, in a society
where there is only one political conflict, for example between proletariat
and bourgeois there will only be two major parties. This may then lead to
the use of an electoral system that reflects and maintains this duality.
Conversely, if in addition to the conflict between proletariat vs. bourgeois,
there is also a conflict between say different national groups this might
lead to the adoption of an electoral system that allows both of these
cross-cutting conflicts to be represented in the form of political parties.
This idea is represented in Figure 2.5. This shows an example of an
imaginary country with two societal cleavages. On the one hand there is
disagreement over whether the government should promote national
unity or autonomy for minority populations. Real life examples of such a
debate would be Scotland and Wales within the United Kingdom and
National unity
A B
Laissez faire Planning
C D
Minority autonomy
Figure 2.5 The making of multi-party systems through overlapping political divides
Catalonia and the Basque Country within Spain. The other disagreement
is over economic laissez faire vs. planning as in the classic class cleavage
found in many countries. In this example parties A and B can agree on
laissez faire, but part ways on national unity vs. autonomy. They, there-
fore, cannot exist as one party necessitating two parties. Similarly B and
D agree on the need for planning but also part ways over national unity.
A and B agree on national unity, but disagree over the need for national
planning and C and D agree on minority autonomy, but fall out over
planning. Hence, because of over-lapping and incompatible political
cleavages, this imaginary country has a multi-party system. So the under-
lying societal conflicts has created a multiparty system, which in turn has
led to the adoption of a PR system which ensures, or at least encourages,
the continuing existence of the multiparty system as illustrated. As
Taagepera and Grofman (1985, p. 350) argue: ‘it may be that countries
with many issue dimensions do purposely pick electoral systems which
enable more than two parties to survive’ (see Chapter 4 for a discussion
of societal conflicts and parties).
It is of course possible that an electoral system was adopted at a time
when there were only two main opposing groups in society leading to the
use of Single Member Plurality. However, later more conflicts started to
emerge, but the two main parties created by the original conflict saw no
need to change the electoral system. The traditional two-party system is
therefore ‘artificially’ maintained by the electoral system despite not fully
reflecting new political conflicts. Arguably this is what has been hap-
pening in the United Kingdom since the 1970s.
Hence, when trying to explain the variations of party systems across
countries, there seems to be a complex interaction between societal cleav-
ages, the electoral system and the number of parties in existence. Political
cleavages create political parties. However, different electoral systems
make it more difficult for small parties to survive, which could mean that
a single large party is able to absorb several political cleavages within it.
Then again, the electoral system of a country is chosen by political parties
– and any party would want an electoral system that suits its situation –
that is, small parties will be concerned with proportionality, larger parties
less so. The electoral system chosen will, therefore, depend on how many
parties exist at the time – which will to a large extent be explained by the
political cleavages in a country.
Which party system is best?
Quite regardless of which party systems exist and how they are formed,
there is the question of whether some systems are somehow ‘better’ than
Party Systems 41
others. It has been argued that ‘the literature on party systems is [...]
replete with normative statements which extol certain party systems but
damn others’ (Daalder, 1992, p. 274).
Indeed, there is a long standing argument that the two-party system is
better than the multi-party system. One example is Duverger and his view
that dualities are natural in politics. If two opposing camps are natural in
politics, so must be two-party systems. Linked to this is an idealisation of
the single-party majority government. Single-party majority governments
are extolled for their stability, and according to Amery’s (1947, p. 19)
view of ‘strong and stable government first’ there really is no viable alter-
native to the two-party system. A single party with a majority is far better
able to make swift decisions, without the need for debating and reaching
compromises that are said to hold back minority and coalition govern-
ments.
The strongest advocacy for the two-party system comes from the idea
of ‘responsible two-party government’ (American Political Science
Association Committee on Political Parties, 1950; Ranney, 1962). The
argument here is that with two parties competing to form single-party
majority governments it will be clear who is responsible for the actions of
governments. Parties can thus be very easily judged on whether they have
succeeded in carrying out their electoral promises.
The other side of this argument is a rejection of multi-party systems.
This rejection is often based on the problems experienced in the Weimar
Republic and Third Republic France before the Second World War and in
Italy and Fourth Republic France after 1945. An excessive number of
extreme parties made for unstable governments, and it is argued, ulti-
mately caused the breakdown of democracy in Germany in the 1930s and
the rise of Nazism. The same led to the breakdown of the Fourth
Republic in France to be replaced with the Fifth Republic. Italy after the
Second World War was (and arguably is) a basket case of too many
parties and weak governments.
However, from the 1950s onwards there has been a re-evaluation of
multi-party democracies. One moving force behind this re-evaluation was
the discovery that economically and socially there was no evidence that
two-party democracies did any better than multi-party democracies.
Indeed, Lijphart argues that consensus style democracies, with multi-
party systems at their heart, are better than majoritarian democracies and
their two-party systems. According to Lijphart (1999, pp. 275–6) con-
sensus democracies:
are more likely to be welfare states; they have a better record with
regard to the protection of the environment; they put fewer people in
prison and are less likely to use the death penalty; and consensus
democracies in the developed world are more generous with their eco-
nomic assistance to developing countries.
Ultimately, it is probably impossible to objectively say which type of

party system is the ‘best’ one. As the above have shown there is a consid-
erable debate about this and the final conclusion depends on one’s own
preferences. In essence, whether one prefers one type of party system over
another depends on: a) what one thinks the consequences of each type of
party system is; and b) whether one approves of those consequences or
not. This is exemplified by Lijphart’s argument for a multi-party con-
sensus democracy. If one approves of, for example, welfare states, not
putting people in prison, and aid to developing countries then one might
well also approve of a party system that encouraged such things.
However, there are many who do not approve of the things Lijphart asso-
ciates with multi-party consensus democracy and who hence would prob-
ably prefer a different kind of democracy. In short, the debate over which
party system is ‘best’ seems to be, at least partly, an ideological debate.
Conclusion
As we have seen, party systems are an important issue to consider when

examining politics. Different party systems will have very different effects
on the politics of a country, and may affect how well the politics of a
country functions. Looking at party systems will also help us compare
different countries. Knowing that the United States is dominated by two
parties, whereas many countries in Europe are dominated by multi-party
politics will tell us a lot about how power is distributed in each country.
However, how party systems are formed and whether one is better than
another is a source of considerable uncertainty. This only makes the
study of party systems that much more important – we know they criti-
cally affect the politics of a country, so the more we can lean about how
they emerge and change the better we can understand the changing poli-
tics of modern democracies.
Chapter 3
Theories of Party Development
Understanding the origins and early development of parties is a critical

aspect of the study of party organisations. The main reason for wanting
to understand the origins of political parties is that the way a party
emerged has a major impact on its organisation for decades afterwards:
a party’s organisational characteristics depend more upon its history

than upon any other factor. The characteristics of a party’s origins are
in fact capable of exerting a weight on its organizational structure even
decades later. Every organization bears the mark of its formation, of
the crucial political-administrative decisions made by its founders, the
decisions which ‘moulded’ the organization. (Panebianco, 1988, p. 50)
This is not to say that parties do not change as they clearly do. The point
is that once a party has become ‘institutionalised’, to use Panebianco’s
term, or set in its ways, it often takes a considerable amount of effort to
change it. Parties, as with most established organisations, are ‘small c
conservative’ in that they change only gradually and often reluctantly. To
understand the current shape of parties it is, therefore, necessary to also
understand how they originated and developed.
Some of the key theories of how parties emerged and evolved as organ-
isations will be the main focus of this chapter. There are two main theo-
ries of how parties first developed. The most well known comes from
Duverger. Duverger’s theory is fairly simply and also the one most com-
monly used to understand the origins of parties and the effects those
origins have on the subsequent life of parties. Panebianco starts from
Duverger’s theory, but critiques it and presents an alternative perspective.
In the following we will first look at Duverger’s theory followed by
Panebianco’s. After that will be a review of one of the central preoccupa-
tions of the party politics literature – the evolution of party types.
The ‘internal’ and ‘external’ origins of parties
The most popular way of analysing the origins of political parties is to

distinguish between parties of ‘internal’ and ‘external’ origins. Whether a
43
party is of internal or external origins is said to have had profound impli-

cations for how they develop as organisations. The internal/external dis-
tinction is related to whether a party started as a group of
parliamentarians who later added an extra-parliamentary organisation
(internal origins), or as an extra-parliamentary organisation who later
succeeded in winning parliamentary seats (external origins). The relation-
ship between the parliamentary group and the extra-parliamentary parts
of a political party is said to be permanently marked by its parliamentary
or extra-parliamentary origins.
The parties of parliamentary origins are mostly associated with early
forms of representative assemblies such as the British Parliament and the
French Constituent Assemblies of 1789 and 1848 (Duverger, 1964, pp.
xxiv–xxvi), but are still relevant in contemporary party politics. Parties of
parliamentary origins start life as groups of parliamentarians to which an
extra-parliamentary organisation is only later added. Traditionally the
impetus to add an extra-parliamentary organisation came from the exten-
sion of the franchise in many countries in the late 1800s and early 1900s.
Before that the electorate was very small and a large extra-parliamentary
organisation was simply not needed. For example, in the United Kingdom
where political parties first developed, less than 10 per cent of the male
adult population were allowed to vote before the Reform Act 1832, and
even that piece of legislation only increased it to 14 per cent. That
amounted to hundreds of thousands of voters, rather than the millions of
voters today. That meant that each constituency would have only a
couple of thousand voters. Each candidate and a few associates would be
enough to do the campaigning needed. However, as the electorate grew it
became necessary to have larger and larger numbers of volunteers to
campaign at election time. This necessitated the creation of an extra-par-
liamentary volunteer membership organisation. Classic examples of
parties of parliamentary origins include many European Liberal and
Conservative parties who can trace their origins back to the 1800s and
sometimes earlier (Duverger, 1964, p. 21).
The parliamentary origins of such parties have significant conse-
quences for their organisational culture. The party on the ground is
essentially there to ensure the survival of the party in public office: ‘the
winning of seats in political assemblies is the essence of the life of the
party, the very reason for its existence and the supreme purpose of it life’
(Duverger, 1964, p. xxxv). The extra-parliamentary organisation is seen
as the servant, or the ‘hand maiden’ as it was called in the British
Conservative Party, of the parliamentary party – there to be used as a
tool by the parliamentary party. Such parties, therefore, tend to have a
fairly oligarchic organisational ethos. The parliamentary group is the
source of power and the raison d’être of the party and the extra-parlia-
Theories of Party Development 45
mentary party is openly and overtly controlled from parliament. Intra-

party democracy is not much in evidence or even deemed particularly
necessary (Duverger, 1964, p. xxxv).
In addition, parties of parliamentary origin tend to be fairly de-cen-
tralised. In their early life the party ‘organisation’ is little more than local
election committees associated with an individual member of parliament
or candidate. The external elements of the party are at this stage not
really units in a formal organisation, but linked purely through the work
of the parliamentarians they have helped get elected (Duverger, 1964, p.
xxix). Only later is a proper national organisation created. Parties of par-
liamentary origins are, therefore, both highly top led, in that the parlia-
mentary group is in control, and tend to be quite decentralised in that the
main link between the different local elements of the party goes through
individual parliamentarians.
In sharp contrast to parties with parliamentary origins are the parties
with extra-parliamentary origins, the externally created parties. These
parties started life outside parliament with little or no parliamentary pres-
ence. Their main purpose was to mobilise voters to ‘break’ into parlia-
ment. This party strategy is usually regarded as mainly associated with
left-wing parties, particularly those based on the organised working class,
that is, trade unions. The creation of these parties was very much a con-
sequence of the extension of the franchise in several European states in
the late nineteenth century. Some of the main beneficiaries of this exten-
sion were the rapidly growing urban working class of the industrial revo-
lution. The first externally created parties were, therefore, driven by the
desire to mobilise these thousands and even millions of new voters to rep-
resent their interests in parliament. The new externally created parties
mobilised hundreds of thousands of voters, incorporating many of them
into their membership organisations, also a new invention. A mass elec-
torate required a mass membership organisation to campaign at election
time. The candidate and a few associates were no-longer enough in the
emerging mass-democracy. The issue of the mass-membership organisa-
tion will be explored later in this chapter and in Chapter 5.
The ‘old’ parliamentary parties, exemplified typically by conservatives
and liberals, did not successfully incorporate the interests of these new
voters in their programmes, leaving the field more or less open for new
parties to emerge. The result was working class social democratic and
socialist parties. The electoral success of these extra-parliamentary organ-
isations with their mass-membership organisations forced the parliamen-
tary parties of the right wing to follow suit. The idea of a membership
organisation outside parliament was, therefore, a left-wing idea which
spread to the right. Duverger (1964, p. xxvii) referred to this spread of
mass-membership organisations as a ‘contagion from the left’.
Parties of extra-parliamentary origins tend to be created around some

form of nucleus – either a particular leading individual, or a pre-existing
organisation – in the case of Social Democratic and Labour parties more
often than not trade unions. Because the party, on the ground existed
before the party in public office their formal roles tended to be the
reverse of the parties of parliamentary origin. The party in public office
was created through the, often hard and lengthy, work of the party on
the ground and it was, therefore, expected to work to defend the interests
of the party on the ground. Indeed, the parliamentary side of the party
was often regarded as only one element of its work: ‘the electoral and
parliamentary struggle remains very important, but it is only one element
in the general activity of the party, one of the means, among others, that
it uses to realize its political ends’ (Duverger, 1964, pp. xxxv–xxxvi). This
means that the parliamentary group does not have the same central place
in a party of extra-parliamentary origin. As Duverger writes: ‘in such
parties there is a certain more or less open mistrust of the parliamentary
group, and a more or less definite desire to subject it to the authority of
an independent controlling committee’ (Duverger, 1964, p. xxxv). The
reality of who is actually in control is often very different, but formally
speaking in parties of extra-parliamentary origins the party in public
office tend to be controlled by the party on the ground.
Also in contrast to the parties of parliamentary origin, the parties of
extra-parliamentary origin tend to be much more centralised. This is inti-
mately linked to how they were created. Parties of extra-parliamentary
origins can often identify a founding date and just as importantly a single
founding place. In the United Kingdom the Labour Representation
Committee, later known as the Labour Party, had its founding conference
on Tuesday 27 February to Wednesday 28 February 1900 in the
Memorial Hall, Farringdon Street in London. The ‘International in
Denmark’, later the Social Democratic Party, was founded at a meeting in
number 27 Adelgade in Copenhagen on Sunday 15 October 1871. The
‘General German Workers’ Association’, later the Social Democratic
Party of Germany, was created on 23 May 1863 in Leipzig. Few if any
parties with parliamentary origins can point to such founding dates and
places as they evolved slowly over time. Many parties of extra-parliamen-
tary origins therefore started on a specific date at a specific location.
Often a key element of a founding conference is to agree on an organisa-
tional structure and ratify a party rule book. This is an important point
to make because it gives them a very different organisational ethos and
structure than parties of parliamentary origins. Because many parties of
extra-parliamentary origin started from an original founding conference
or meeting they were from the beginning highly centralised organisations
with a single hierarchical structure. This is in contrast to the amalgama-
tions of disparate local groups typical of parties of parliamentary origins

(Duverger, 1964, pp. xxxiv–xxxv). The way a party originated can then
be seen to have had a crucial part to play in the organisational structure
of that party.
An alternative view of the emergence of political parties
While Duverger’s view of party emergence is the most popular, it has also
been criticised. Panebianco points out that it fails to explain how parties
of similar origins nevertheless end up having very different structures. His
claim is that we need a more complex model if we are to understand how
parties develop (Panebianco, 1988, p. 50).
Panebianco develops what he calls a ‘genetic model’ for party emer-
gence, which can then be applied to analyse and understand specific
parties (Panebianco, 1988, p. 50). The genetic model consists of two
parts. The first part focuses on how a party first emerged and the forces
involved in its creation. This in turn affects how the party ‘institution-
alised’ – that is, how the party settled into a regular pattern of organisa-
tional activity.
There are three issues we need to consider when investigating the
emergence of a political party (Panebianco, 1988, pp. 50–3). The first
issue is whether the party was created through territorial penetration or
territorial diffusion. Territorial penetration involves a pre-existing centre
controlling, stimulating or directing the creation of the ‘periphery – local
and regional party organisations. Territorial diffusion on the other hand
is when local elites create party organisations in their locality which are
then later merged into a national organisation. A combination of the two
can also take place. Local parties emerge in some areas and eventually
form a national organisation (diffusion). This national organisation then
goes on to create local party organisation where there were none before
(penetration). However, there will more often than not be a prevailing
tendency in that a party is created mainly by one or the other method.
Panebianco argues that liberal parties have tended to emerge through a
mixture of the two. Many socialist parties and religious parties have
tended to be formed through diffusion. Conservative and communist
parties have typically been created through penetration.
These two forms of origins will have an effect on how the party is
structured. A party created through penetration will by definition have a
sufficiently strong centre to stimulate the creation of local party organisa-
tions. Such parties therefore tend to be highly centralised. By contrast
parties created by diffusion will be based on the merger of many local ele-
ments with their own elites and power centres. These parties therefore
tend to have a more federal character and be marked by struggles

between local groups for control of the national party.
The second issue we need to consider when looking at a party’s origins
is whether there was an external ‘sponsor’ organisation (such as a trade
union or a religious organisation) that created the party as its ‘political
arm’. This will make a difference to the legitimacy of the leadership. The
leadership of a party with an external sponsor will base their power and
legitimacy as leaders on the support of the external organisation. The
external sponsor can, therefore, intervene in a party and shift its support
from one group to another. Where there is no external sponsor the leader-
ship base their power and legitimacy on the party itself. There is then a
difference between externally legitimated parties and internally legiti-
mated parties.
The third and final factor relates to the role of charisma in the forma-
tion of a party. Charismatic leaders more often than not have a big part
to play in the formation of most parties, but the ‘concern here is with
parties formed by one leader who imposed himself as the undisputed
founder, conceiver, and interpreter of a set of political symbols [...] which
became inseparable from his person’ (Panebianco, 1988, p. 52).
Examples of such parties would be List Pim Fortuyn in the Netherlands,
the short-lived Veritas in the United Kingdom (created by TV personality
Robert Kilroy-Silk), and Berlusconi’s Forza Italia. In addition to this
‘pure’ charisma, there is also ‘situational charisma’, where a particular
crisis creates a leader that people are willing to follow. ‘A party based on
pure charisma has no existence apart from its leader [...] a party based on
situational charisma is not simply the leader’s creation, but is a product
of many different impulses and thrusts’ (Panebianco, 1988, pp. 52–3). An
example of such ‘situational charisma’ might be the Progress Party in
Denmark which was created in 1973 by Mogens Glistrup on the back of
increasing reaction against raising taxes. The party was closely linked
with Glistrup, but still survived his forced absence from running the party
when he was convicted for refusing to pay tax.
These three key issues will combine to affect a party’s ‘institutionalisa-
tion’. Institutionalisation refers to: 1) the degree to which a party is able
to dominate its surroundings as opposed to being dominated by them;
and 2) the extent to which the various elements making up the party,
such as local branches, are independent of each other or tied together
into a coherent whole. A party which is highly autonomous in relation to
its surroundings and where the various units are highly interdependent of
each other is a party with a high degree of institutionalisation
(Panebianco, 1988, pp. 55–8). If a party is to survive it has to institution-
alise, but the extent of institutionalisation depends on how the party was
created.
Type 1 Created through territorial penetration; strongly

institutionalised; externally legitimatised.
Type 2 Created through territorial diffusion; weakly institutionalised;

externally legitimised.
Type 3 Created through territorial penetration; strongly

institutionalised; internally legitimised.
Type 4 Created through territorial diffusion; weakly institutionalised;

internally legitimised.
Figure 3.1 Party origins and institutionalisation
The issue of how a party has emerged and its degree of institutionalisa-
tion can be combined to analyse and understand how a party has devel-
oped, and the importance of its origins in that development (see Figure
3.1).
The parties that most resemble Type 1 are the communist parties of
the Cold War period. They were often created by territorial penetration,
and although many had an external sponsor, the Communist Party of the
Soviet Union, this sponsor was ‘extra-national’ and therefore had a some-
what indirect impact. They, therefore, tended to be highly institution-
alised. Type 2 is represented by labour and social democratic parties,
such as the British Labour Party and some religious parties. The fact that
these parties were created by territorial diffusion and were dependent on
external sponsoring organisations, in many cases trade unions, meant
that they were dependent on their external sponsor and based on local
semi-autonomous units. This tended to lead to low levels of institutionali-
sation.
Type 3 consists of parties created from a pre-existing centre, such as a
parliamentary group, and with no sponsoring organisation. They are
therefore highly institutionalised as they are controlled from the centre
and are much less dependent on their external environment, in contrast
to some social democratic parties’ dependency on the trade unions that
helped create them. One could mention the British Conservative Party as
an example of this type of party.
Finally there are Type 4 parties. While there is no external sponsoring
organisation these parties were still created by territorial diffusion. They
were, therefore, federations of existing groups, which slows institutionali-
sation. One example is the German CDU.
This leaves the issue of charisma, which Panebianco refers to as a
‘deviant case’ since it does not fit into the above schema (Panebianco,
1988, p. 65). Parties based on charisma can go one of two ways. Either
they simply melt away once the original leader leaves the stage – or they
manage to live on after his or her departure. If the party dies then institu-
tionalisation is not relevant. If the party continues then it will most likely
end up as a highly institutionalised party. A party created by a single
charismatic leader will be highly centralised as all decisions, ideology and
purpose flow from that person. If the party survives the departure of the
founding leader the organisation will more than likely retain that highly
centralised structure.
These two different ways of looking at the origins of parties are pow-
erful tools in understanding the current organisational state of a party.
Duverger’s approach has the benefit of simplicity, but is perhaps also a little
lacking in detail. Panebianco gives a far more detailed tool for analysing
parties, but one that is also more complicated to apply, exactly because of
the extra detail. However, both approaches show the importance of taking
into account a party’s origins when trying to understand its current state.
Understanding party types
In addition to analysing how parties have emerged and how this affects
their current situation, a lot of effort has gone into categorising different
ways that parties have organised themselves internally. Indeed, one of the
most active areas of party politics research is the development, opera-
tionalisation, critiquing and re-development of party types. There is not a
single commonly agreed definition of what a party type is – which is a
problem we will return to at the end of this section. However, the party
type debate has still dominated our understanding of how parties have
changed over time. Various authors have described party types as illustra-
tions of how parties currently behave, how they used to behave and how
they might behave in the future. Often the development of a new party
type comes as a response to parties, or a particular group of parties,
doing something in a new way. Each party type describes how parties
behave, their central characteristics and motivating forces. In the fol-
lowing we will examine the most important party types and show how
this approach to our understanding of parties has helped us to see how
parties have changed over time. The following will also deal with some
problems which have become apparent regarding the ‘party types’
approach to analysing party life.
Cadre parties to mass parties
The first two types of party organisations to be described were cadre

parties and mass parties (see Duverger, 1964, p. 63). Cadre parties
equated to the early parties of parliamentary origin, whereas the mass

party matched the organisations of extra-parliamentary origin.
The first parties to emerge in early legislative assemblies were cadre
parties (Duverger, 1964, p. 65). These parties were hardly recognisable as
parties in the modern sense. They had little formal structure and tended
to be based around more or less permanent agreements among parlia-
mentarians based on short- and long-term common interests. Cadre
parties existed at a time when the franchise was highly limited and there
was little need for an extra-parliamentary organisation to mobilise voters,
as described earlier. Cadre parties did not have card carrying and fee
paying members as most modern parties do. Membership of, or perhaps
rather association with, a cadre party was highly exclusive and based on
a person’s financial resources, personal contacts and influence or
expertise. Cadre parties are primarily associated with legislative assem-
blies in Europe in the late eighteenth and nineteenth centuries, periods
where membership of such assemblies was based on Royal favour, mem-
bership of the aristocracy or election by a small, rich and exclusive elec-
torate based on property ownership and wealth.
The cadre parties started to face serious competition in the late nine-
teenth century when the franchise was extended. As described earlier, the
much enlarged electorate led to extra-parliamentary organisations mobil-
ising new voters, particularly among the working class. These new
working-class parties did not rely on the significant resources of a few
individual members, but the regular, albeit small, contributions of many
members. In these ‘mass parties’ as Duverger called them, membership
was not exclusive and based on recommendations from existing
members, but ‘wide open’ to all who could support the basic goals of the
party. In the mass party the main focus was the representation of a par-
ticular group or class in society, hence the phrase ‘class-mass’ party
(Kirchheimer, 1966, p. 183). In the mass party the membership was a
crucial element of the party and central to its survival: ‘the members are
the very substance of the party, the stuff of its activity’ (Duverger, 1964,
p. 63) and indeed the concept of a fee paying member has been described
as a socialist mass party invention (Duverger, 1964, p. 63).
It is important to note that the main difference between a cadre party
and a mass party is not one of size. Even when at their most powerful,
the biggest West European communist parties, such as the French and
Italian, still did not rival the main Socialist, Labour and Social
Democratic parties in terms of membership numbers. However, commu-
nist parties are still to be regarded as mass parties, although of a particu-
larly centralised and tightly run variety. In short, ‘mass party’ refers to a
particular form of organisational principle, rather than the size of the
membership organisation.
The mass parties eventually proved themselves able to challenge the

old cadre parties in term of electoral success. As a result the latter started
to copy the mass party model: ‘cadre parties sometimes admit ordinary
members in imitation of mass parties. In fact, the practice is fairly wide-
spread: there are few purely cadre parties’ (Duverger, 1964, p. 64). In
other words, cadre parties ‘after the contagious pattern of the mass party’
(Duverger, 1964, p. 64) sometimes rival the largest mass parties in size.
However, in the ‘modernised’ cadre party a ‘mass membership’ exists
purely to ensure the survival of the parliamentary party and sometimes to
add some popular democratic legitimacy to the organisation. This is in
contrast to the mass party where the members are the base from which
all activity of the party flows.
It is clear that mass parties emerged as a result of the enfranchisement
of the working class. They were therefore ‘[a child] of democracy, of
mass franchise, of the necessity to woo and organise the masses’ (Weber,
1990, p. 35). This ‘contagious organisation’ (Duverger, 1964, p. 25)
spread from the left to the right of the political spectrum as the success of
the mass party model forced the old cadre parties to ‘counter mobilise’
(Lipset and Rokkan, 1967). By the 1950s the mass party model stood tri-
umphant and seemed to be the type of party best fitted to modern mass
democracy (Bille, 1997, p. 30).
Challenging the mass party model: the catch-all party
The idea that mass parties were the party type of the future, either in its
‘pure’ form in left-wing parties, or as imitations in cadre parties with extra-
parliamentary membership organisation add-ons, was quickly challenged
by Kirchheimer (1966). In an incomplete article published shortly after his
death, Kirchheimer argued that the evolved cadre parties were in fact better
suited to the post-war environment with its more diverse electorates and
reduced class antagonism than was the traditional mass party based on
class representation. The old cadre parties, having imitated some of the fea-
tures of the mass party, most notably the presence of fairly large member-
ship organisations, were becoming ‘catch-all people’s parties’ (Kirchheimer,
1966, p. 184). As the electorate grew, the cadre parties were not able to
rely for their electoral survival on the small societal elite they had previ-
ously been based on. They, therefore, needed to reach out beyond their tra-
ditional privileged audience and as a result turned more fully to the
electoral area. The mass parties were based on the mobilisations of a spe-
cific, albeit large, group of voters. The evolving cadre parties based their
new electoral strategy on attracting as many voters as they could from
whatever groups in society they were able to tap, hence the term ‘catch-all’.
As with the success of the mass party model in the late nineteenth
and early twentieth centuries, so the success of the catch-all party model
in the post-1945 era became a contagion which spread from the right-
wing of politics to the left (Epstein, 1967, p. 257). By the 1970s the
catch-all party, rather than the mass party, looked like the model of the
future.
The catch-all party has a number of key characteristics, including:
• ‘a drastic reduction of the party ideological baggage’;

• ‘further strengthening of the top leadership groups’;
• ‘downgrading of the traditional role of the individual party member, a
role considered a historical relic which may obscure the newly built-
up catch-all party image, and
• ‘de-emphasis of the classe gardée, specific social-class or denomina-
tional clientele, in favour of recruiting voters among the population at
large’. (Kirchheimer, 1966, pp. 190–1)
It is important to note that becoming a catch-all party does not mean that
a party is so ‘heterogeneous as to represent the whole social spectrum and
whose connection with its original classe gardée had completely disap-
peared’ (Panebianco, 1988, p. 263). Rather what the catch-all party is
trying to do is simply to open its doors to different social groups ‘whose
interests do not adamantly conflict’ (Kirchheimer, 1966, p. 186). A catch-
all party is, therefore, not a party which has no ideological or societal
connections what-so-ever. No party could survive long without any
coherence or predictability to its policy preferences or legislative actions.
The main difference between the mass party and the catch-all party
models is that the mass party focus was primarily on the representation
of the interests of one particular group, whereas the catch-all party is
trying to open up to new electoral markets that do not fundamentally
clash with its old position.
As we shall see later in the book, when it comes to party ideology; the
role of party members; the balance of power in parties; and campaigning,
Kirchheimer’s description of the catch-all party seems to fit with a lot of
what has been happening in modern political parties.
Organisational consequences of the catch-all party: the

electoral-professional party model
The catch-all party model focused very heavily on the ideological aspects
of political parties, especially the idea of not targeting one specific class,
but rather casting one’s net as wide as possible. Hence, one criticism of
the catch-all party model was its strong focus on the ideological changes
and moderation that went into becoming a catch-all party and how that
affected the catch-all party’s relations with other political actors such as
interest groups (Panebianco, 1988, p. 263). Panebianco argued that this
focus overlooked important organisational implications of the catch-all
party model. The solution was the idea of the electoral-professional
party.
The electoral-professional party type does not reject the idea of the
catch-all model. Rather, what it does is develop the catch-all type with a
focus on organisational issues and in particular the professionalization of
the party. In the classic mass party ‘a crucial role is played by the appa-
ratus, the party bureaucracy: the “representative bureaucracy” [...] is
used by the mass party leaders to maintain close ties with the members,
and, through the members, with the classe gardée’ (Panebianco, 1988, p.
264). By ‘representative bureaucracy’ is meant a system where local party
members (the party on the ground as described in Chapter 1) interact
with voters at the local level and thereby get an idea of the wishes and
demands of the population. Party members then elect representatives and
party officers to serve higher up in the party in local, regional and
national executives and as delegates to the party conference (the party in
central office). These party officers and delegates carry the wishes and
demands from the voters to the party’s parliamentarians (the party in
public office) to deal with in the form of legislation. The representative
bureaucracy, therefore, acts as an unbroken chain from people to parlia-
ment.
In the electoral-professional party this organising principle is replaced
with an increasingly professionalized approach to running the party. The
electoral-professional party makes far greater use of ‘so-called experts,
technicians with specialist knowledge’ (Panebianco, 1988, p. 264) such as
pollsters and what would now be called spin-doctors, instead of relying
on members as the main source of information about what voters are
demanding. Overall there are five key differences between mass parties
and electoral-professional parties:
1 Where the mass party relied on party bureaucrats to run the party
and foster links with the electorate, the electoral-professional party
relies on experts.
2 The mass party is a membership-based party with strong vertical ties
between members and leaders. The party appeals to a sense of
belonging (class or group association). In the electoral-professional
party the vertical link is weakened and the party appeals to the
‘opinion electorate’ rather than the ‘electorate of belonging’ – the
electoral professional party attracts voters by catering to their various

and shifting individual demands rather than their interests as a group.
3 Where the mass party was dominated by leaders of the party on the
ground, the electoral professional party is dominated by parliamentar-
ians and often by the leader of the party in public office.
4 The mass party was funded by membership fees and affiliation fees
from a narrow range of key interest organisations such as trade
unions. The electoral-professional party is funded by donations from
a range of interests groups and public funds.
5 The mass party was dominated by ideology and ‘believers’, whereas
the electoral professional party is dominated by shifting issues, and
pragmatic careerists.
It is important to note that few if any parties fit perfectly into either the
mass party or the electoral-professional party categories. They are what
we might call ‘ideal types’ which help us to analyse real-life party
organisations (Panebianco, 1988, pp. 264–5). What the two types do is
show that there is a tendency for parties to go from having primarily
mass-party characteristics to having primarily electoral-professional
characteristics.
The extent to which a party has moved from one type to another
depends on several factors, relating both to the internal life of a party and
its external environment. According to Kirchheimer only major parties
are likely to move in a catch-all direction. Large parties are by definition
parties that have managed to attract a fairly large vote share, something
which would presumably be easier to achieve and maintain with a catch-
all stance. Parties with a very specific and narrow focus are also by defi-
nition likely to be small and are more likely to stay away from a catch-all
position. Parties not expected to reach catch-all status included organisa-
tions such as the post-War German Refugee Party, the Danish anti-tax
Justice Party and the Swedish Agrarians (Kirchheimer, 1966, pp. 187–8)
as their purpose is too narrow or their clientele very small. In addition
even major parties in fairly small democracies, such as the Scandinavian
countries, are unlikely to become catch-all parties (Kirchheimer, 1966, p.
188). Parties well entrenched in small homogeneous democracies, such as
the Norwegian and Swedish Social Democrats have little incentive to
move in a catch-all direction (Kirchheimer, 1966, p. 188). In addition
Kirchheimer argues that ‘conversion to catch-all constitutes a competitive
phenomenon’ (Kirchhermer, 1966, p. 188). If one party has successfully
adopted a catch-all position other parties in the system are likely to
follow suit, either in the hope of electoral gains or out of fear of losing to
a competitor.
Two key factors are central to understanding ‘the speed and intensity
of transformation’: the level of institutionalisation of a party and the
party’s environment (Panebianco, 1988, p. 265). A move from mass
party to electoral-professional status is more likely to happen before a
party has institutionalised, or if the level of institutionalisation is very
low. A highly institutionalised party is more likely to resist pressure for
organisational reform, be it from internal or external forces. In addition
the argument that large parties are more like to undergo transforma-
tions than smaller ones suggests that we are more likely to see electoral-
professional parties emerge in less fragmented party systems than highly
fragmented ones (Panebianco, 1988, p. 265). If a political system is
dominated by a few large parties ‘change will take place sooner and
[take] place more quickly’ (Panebianco, 1988, p. 265). A political
system characterised by many small parties each with a very specific
electoral market is less likely to see change. It makes sense that in, for
example, the United Kingdom where the electoral system punishes small
parties, parties with a desire for power have an incentive to broaden
their appeal as much as possible. Conversely, in countries with very
proportional electoral system, such as in Scandinavia, Italy or Israel, the
incentive to widen one’s appeal and professionalise is much less intense.
Smaller parties can and do lead very productive lives, participating in
coalitions or supporting minority governments and, therefore, have less
reason to professionalise.
Finally Panebianco argues that changes in two major environmental
factors are important in affecting party change. The first factor is related
to changes to the socioeconomic structure of a country (Panebianco,
1988, p. 265). In other words the rise of a large middle class and the
simultaneous decline in working class voters will encourage party change.
Electorates have ‘become more socially and culturally heterogeneous’
(Panebianco, 1988, p. 266) meaning that focusing on just one (possibly
shrinking and certainly changing) social group is not electorally viable.
This will have a major impact on parties and spur them in a catch-
all/electoral professional direction.
Further, technological change will have an impact on the professionali-
sation of party organisations. As new ways of communicating develop,
the need for experts to tap these new communication channels increases.
In the context of technological change Panebianco writes about the rise
of television campaigning, but in the current communication environment
the rise of increasingly mobile and high-speed Internet connections is a
major factor as illustrated by Barak Obama’s skilled use of Internet net-
working in both his primary campaign and presidential victories. Clearly,
to tap the full potential of such communication channels professionals

are needed.
The cartel party
Where the cadre, mass, catch-all and electoral-professional party models

have focused on changing ideologies and internal professionalisation,
there has been a rising interest in the relationship between parties and the
state on the one hand and the electorate on the other. This is where the
idea of the cartel party comes in (Katz and Mair, 1995).
The cartel party model continues the move away from the grassroots
organisation started by the catch-all/electoral professional party type.
Having loosened their ties – in the form of the membership organisation
– with civil society, modern parties have become increasingly dependent
on the state.
First, as the income from membership fees has declined, so state
support for parties has increased. Second, MPs have gone from focusing
on their role as representatives of specific societal interests and groups to
focusing on their role as legislators and members of government. This
development has taken place over a number of stages. In the era of the
cadre party, civil society (effectively limited to the few wealthy and privi-
leged levels of society with the vote), political parties and the state over-
lapped. Indeed, civil society, parties and the state were more often than
not indistinguishable from each other (Katz and Mair, 1995, p. 10). The
mass party on the other hand served as a bridge between civil society –
increasingly seen as all citizens limited only by age – and the state.
Parties, anchored in civil society through their membership organisations,
represented civil society through their MPs who also served as govern-
ment ministers and therefore linked to the state. The cartel party has now
more or less severed its links with civil society and is becoming increas-
ingly merged with the state.
It is important to note that not all parties are necessarily becoming
cartel parties. By definition, not all parties can become part of the cartel.
The idea of the cartel party, in addition to the developments described
above, is that these parties use their privileged access to state resources to
exclude challengers. With the rise of the cartel party the state ‘becomes a
fount of resources through which these parties not only help to ensure
their own survival, but through which they can also enhance their
capacity to resist challengers from newly mobilized alternatives’ (Katz
and Mair, 1995, p. 16).
The idea of the cartel party is linked to the claim that parties are
becoming increasingly separated from the people they were originally
designed to represent. Becoming ever more dependent on state resources,
and thus independent of civil society resources, they lose contact with the
ideas, demands and lives of voters. The rise of the cartel party could
therefore be seen as a negative development.
Conclusion: Beyond the cartel – beyond party types?
Since Katz and Mair’s work on the cartel party in 1995, the work on
party types has continued apace. Examples of party types to emerge in
the literature since 1995 include the post-cartel party (Yishai, 2001) the
franchise party (Carty, 2004); the media and the communitarian party
types as a suggestion for two alternative paths of development (Richards,
2000); the cyber party (Margetts, 2001); the business firm party (Hopkin
and Paolucci, 1999); modern cadre parties and leader-centred parties
(Wolinetz, 2002); catch-all party plus (Puhle, 2002); and the grassroots
party.
This plethora of supposed new party organisational forms illustrates a
number of problems with the party type approach to the study of polit-
ical parties. While the cadre to catch-all party evolution is more or less
accepted, there seems to be no consensus on what has happened since.
The competition to be Kirchheimer’s successor is certainly fierce and has
not necessarily served to help us get a clear picture of what the overall
trends in party organisational change actually are, apart from suggesting
that change is indeed happening.
The reasons for this proliferation of new types since the mid-1990s
compared to the relatively straight forward cadre to catch-all develop-
ment are not entirely clear. One option is that the study of parties has
experienced an increase in popularity in recent decades. Hence, what we
are seeing is not a sudden proliferation of new types compared to the
late 1800s and early 1900s, but rather an awareness of the complexity
of party organisational change. A complexity perhaps hidden by the rel-
atively narrow field of scholars previously engaged in party organisa-
tional research, leading to an equally narrow field of theories related to
how parties are changing. In other words the relatively simple cadre to
catch-all theory may in fact hide a picture as complex as our current
one.
On the other hand it is also possible that democracy and, therefore,
party politics may have become more complicated than was previously
the case, although most generations seem to view their own age as more
developed or complex than the previous age. Whichever is the case, what
is clear is that we are faced with a very confusing picture of party organi-
sational development.
This would suggest that there is a need to go beyond the party types
literature, important as it is, to get a clearer understanding of what the
overall trends in party organisational development are. The following
chapters will do just that.
Chapter4
Ideology
One of the most important and influential ideas of what political parties
are or should be is that they are the organisational embodiment of an ide-
ology. It is a common assumption that the whole point of a party is that
it has a vision of ‘the good society’ and its main purpose is to make that
vision a reality. Indeed, a common criticism of modern political parties is
that many of them, especially those that manage to get into government,
have lost their ideological foundations and are interested in power and
government perks not to change things, but for the personal benefits that
such power brings. In other words, having an ideology is assumed to be
at the core of what a party should be, and the absence of ideology is a
key criticism of (some) political parties.
This chapter will deal with the issue of ideology, and what will in some
places take the place of ideology as the guiding force of party action. The
chapter will first explore the meaning of ideology in the context of party
politics. It will then look at what drives ideology – that is political cleav-
ages. Having done that the chapter will present the key ways in which
parties are grouped when it comes to their ideological position that is
into ‘party families’. The chapter will then look at party ideology in
Western Europe, the United States, India and Africa.
On ideology and party politics
One of the main illustrations of the importance of ideology in the lives of

political parties can be found in the names that many parties have
adopted. Frequently, although far from always, ideology will be at the
centre of a party’s name: the Conservative Party in the United Kingdom;
the Sozialdemokratische Partei Deutschlands (German Social Democratic
Party), Partido Socialista Obrero Español (Spanish Socialist Workers’
Party), The Communist Party of India, The Liberal Party of Canada also
known as Parti libéral du Canada, the Liberal Party of Australia, and
Partido Conservador Colombiano (The Columbian Conservative Party)
to mention but a few. It is true that the name of a political party is not
always a reliable indicator of a party’s ideology and frequently the name
of a party is not related to an ideology in the first place. The Democratic
60
Ideology 61
and Republican parties of the Unites States are not named after ideologies
and indeed the Democratic Party started out as the Republican Party,
then became the Democratic-Republican Party and eventually The
Democratic Party (Witcover, 2003). Nevertheless, the fact that so many
parties name themselves after an ideology is an indication of the formal
importance of ideology in the life of political parties.
The most prominent version of what party ideology is based on is the
classical left–right spectrum inherited from the French Revolution. In the
National Assembly created after the revolution in 1789, supporters of the
monarchy would congregate on the right-hand side of the chamber.
Opponents of the monarchy and the old order would congregate on the
left-hand side of the chamber. This division continued through several
changes and reforms of the French legislative assembly – to the point
where ‘left’ became associated with the desire for change and opposition
to the old and traditional; and ‘right’ became associated with tradition
and maintaining existing structures. On this simple scale social demo-
cratic, socialist and communist parties on the left are opposed by liberal,
conservative and fascist parties on the right. Indeed, the idea of a left–
right spectrum is so ingrained in discourse on political parties that
Laponce (1981) referred to it as ‘political esperanto’, a simple and uni-
versal language which helps structure a very complex world of overlap-
ping policies and options. In short, the left–right spectrum helps
politicians, the media and voters to understand where different parties
and candidates are located both in relation to each other and in relation
to the position of the observer (Freire, 2008).
It is worth noting that the ‘left/right’ seating of the French National
Assembly of 1789 is continued in the European Parliament. The cross-
national party groups in the European parliament are seated such that
members of the European United Left/Nordic Green Left, commonly seen
as the most radical socialist group in the European Parliament, sits on the
far left of the chamber. To their right is the left-of-centre Alliance of
Socialists and Democrats. Broadly in the centre is the Alliance of Liberals
and Democrats and to their right the right-of-centre European People’s
Party which is broadly conservative and Christian democrat in character.
On their right are several smaller right wing groups. Clearly, the idea of a
left–right spectrum is very much alive today.
The traditional view of the left–right spectrum is based on socioeco-
nomic issues principally centred on the role of the state in the operation
of the free market and the extent of support offered by the state to those
who have fallen on hard times. The common assumption is that ‘left
wing’ means advocating state intervention in the free-market and gen-
erous support for those unable to work, be it due to disability or lack of
available work. The right wing tends to be associated with small state,
free market economics and individual self-sufficiency. There is clearly

much more to it than that, but this is a simplified view of the basic left-
right divide.
It is also assumed that voters will be distributed across this spectrum,
with the highest concentration of voters around the centre, and rela-
tively few at the fringes. This distribution of voters is said to heavily
influence the ideological position of parties. According to Downs
(1957), parties will position themselves ideologically to achieve
maximum voter support. Downs assumes a single left-right ideological
dimension, a bell-shaped voter distribution along that dimension (see
Figure 4.1a) and the existence of two parties in that system. According
to Downs these two parties, if they wanted to win, would approach
each other in the middle. Downs assumes that voters would choose the
party closest to their own position. Voter X might not be at all happy
with the position of Party A, but it is still better than Party B. The same
is the case for voter Y, but preferring party B over party A. In essence
each party controls everything on their flank. Under a different distribu-
tion of votes (see Figure 4.1b) the parties would position themselves to
ensure maximum voter support. In the case of Party B, more votes
would be available to the left of centre, but no party could make such a
radical shift in their ideological position and still remain credible. In
short, the ideological position of electorally minded parties will be
strongly affected by voter preferences.
There are many criticisms of Downs, not least the argument that
parties do not just react to voter choice. They also shape and indeed limit
them (Dunleavy, 1991; Ware, 1979). Parties, especially if they are in gov-
ernment, can to a certain extent shape the ideological preferences of the
electorate. In addition voters can only vote for the parties that exist. It is
often very difficult for new parties to break into a political system to
provide new voter choices. Hence, party leaders do not always have to
follow voter demands. They can to a certain extent decide what they
want to offer (see also Chapter 7).
Further, there is much evidence to suggest that a single left–right spec-
trum is not sufficient to understand the position of individual parties and
their ideological relationship with other parties. Two examples may serve
to illustrate this (see also Chapter 2). The first is that parties that pur-
portedly exist on the same side of the left–right spectrum may differ quite
significantly from each other. One prime example of this is liberal and
conservative parties. Both are traditionally seen as being right-of-centre,
but they can show quite significant variation from each other. This varia-
tion is not just about being more or less right wing – a matter of centre-
right/right/far-right – but about differences which are not easily explained
purely by the traditional view of a socioeconomic left-right spectrum.
Ideology 63
A B
Voter X Voter Y
Left Centre Right
A
B
Left Centre Right
Figure 4.1 Downs’ voter distribution and party placement
Source: Downs (1957).
Benoit and Laver (2006, p. 13) point out that while many liberals and
conservatives may agree, more or less, on a fairly minimal role of the
state in the economy, there can be a great difference between liberals and
conservatives when it comes to the role of the individual in society. At
one end we find conservative ideas of loyalty to the nation, and the need
for conformity to certain cultural and social values often founded on reli-
gious beliefs. At the other end we find liberal ideas associated with the
freedom of the individual to adhere to their own cultural and social
values independent from what ‘the nation’, the state, organised religion
or other institutions might think. So, while conservatives and liberals may
agree on promoting a free market economy, they might well disagree
strongly over issues such as abortion, gay rights and censorship. There
then seems to be a second ideological dimension stretching between

social-liberalism and social-conservatism.
Another example is the ideological position of the extreme right. While
parties such as the British National Party, the Front National in France
and the Danish People’s Party might agree with their more moderate con-
servative neighbours on the issue of moral values, they often disagree on
the role of the state. Many extreme right parties often support extensive
intervention in the economy by the state and support for the less well off
in society (provided these people belong to some definition of ‘the nation’
– in other words, are not ‘foreign’). They, therefore, can have very right-
wing policies on cultural norms and values, but at the same time very
left-wing policies (although with a nationalistic and sometimes racist
twist) on economic issues.
To add to the confusion there are issues that do not fit into either eco-
nomic or social values – one main example being debates surrounding
national self-determination and regional autonomy. A simple left-right
scale combined with a social liberal/conservative scale would not be able
to capture the full ideological stance of say the Scottish National Party,
the Partit dels Socialistes de Catalunya (Socialists’ Party of Catalonia), or
the Italian Lega Nord per l’Indipendenza della Padania (North League
for the Independence of Padania).
This has led Benoit and Laver to conclude that: ‘a single “left-right
dimension” [...] is not always enough to convey even the big picture’
(2006, p. 13). This suggests that ‘multidimensional policy spaces are [...]
a reasonable assumption’ (Warwick, 2002, p. 102).
It is, therefore, argued that parties are not simply expressions of left–
right ideological differences but rather are to be found somewhere within
more than one dimension where parties traditionally seen as opposites on
the simple ‘left–right’ spectrum can sit side by side on some other spec-
trum. It is clear that to understand party ideology we need to go beyond
the simple left–right ideological spectrum based on socioeconomic differ-
ences.
Political cleavages
The question then is what these other dimensions are that parties relate
to. The most well known and most cited approach to dealing with this
issue was developed by Lipset and Rokkan (1967). They argue that
parties are driven by and are expressions of underlying conflicts and
divides, or political cleavages, in society. Each party will represent the
people on one side of some major political conflict. Their main argument
is that the ideology of parties is shaped by two key societal revolutions:
Ideology 65
the national and the industrial. These two revolutions have been the
source of four political cleavages or conflicts.
The national revolution refers to the creation of the modern nation
state. The idea of the nation state is, relatively speaking, a recent phe-
nomenon, emerging from the treaties that ended the 30 Years War in
1648. Before this time the loyalty of an individual might be to the Pope in
Rome on religious matters and to a local aristocrat on political matters,
rather than to a central authority (such as a monarch). From around the
middle of the seventeenth century the idea that each country would have
a single political centre to which all the population owed loyalty on all
matters started to emerge. In short the national revolution saw a process
of unification of all power, both spiritual and political, in a single
national centre. In the study of the nation state and international rela-
tions this is referred to as the rise of the Westphalian System, named after
‘The Peace of Westphalia’ – that is, the treaties that ended the 30 Years
War. This gradual unification of power in modern states created two
counter-trends which eventually led to the emergence of political organi-
sations, including parties.
The first counter-trend was territorial and was driven by local and
regional opposition to nation building, what has been referred to as the
‘centre-periphery’ cleavage. As modern nation states consolidated their
power, ‘subject peoples’ or minorities within the emerging states
mobilised to try to retain or regain political independence. In other
words, there were those who did not want to be part of the new nation
states. This problem of incomplete ‘nation building’ is often associated
with former European colonies, especially in Africa (Clapham, 1996).
When former colonies gained their independence their borders were
based on the old colonial borders. These in turn were based on agree-
ments between European powers. They thus ignored traditional tribal
and national feelings that colonial populations might have. This meant
that people on either side of a border between two former colonies might
have more in common with each other than with people in the state they
found themselves in. This then led to internal conflict as groups within a
state sought independence. Sometimes the struggle for independence was
done though political means (that is, a political party), but often via mili-
tary means when political means failed.
However, while often associated with former colonies, this issue has
also affected European states. There are numerous examples of how this
‘centre vs. periphery’ cleavage has affected the creation and ideological
stance of political parties: The Scottish National Party, Plaid Cymru and
Sinn Fein in the United Kingdom; both the Basque Country and
Catalonia in Spain have left and right wing nationalist parties arguing for
further autonomy or even independence; the Belgian party system is
divided by language with Flanders and Wallonia spawning separate

parties and Francophone parties in Quebec resist rule by Ottawa. In addi-
tion one could argue that Euro-sceptic parties such as Junibevægelsen
(June Movement) in Denmark and the United Kingdom Independence
Party are evidence of a new form of EU centre-periphery cleavage.
Brussels as the ‘capital’ of the EU is the new political centre, and the pop-
ulation of (some) member states become the periphery resisting their
inclusion in the EU. This has also affected more traditional parties, many
of whom have taken a strong anti-integration (‘anti-Brussels’) stance.
The second political cleavage created by the national revolution is
between church and state. As the state consolidated its power it not only
acquired control over temporal matters within its territory, but also over
issues which the church felt it had a stake in, such as school curriculums,
the role of the church in politics, and moral issues such as abortion,
pornography, and gay rights. Again, parties mobilised on religious
grounds, most notable in the form of Christian Democratic parties found
in, for example, Germany, but also, sometimes to a lesser extent, in much
of the rest of Europe. Christian Democratic parties are also found in
Chile and Mexico. It could be argued that the rise of the Christian right
in the United States and its increased strength in the Republican Party is
an example of this cleavage at work. The Christian right in the United
States is very different from Christian Democracy in Europe but is
equally driven by the debate over the role of the state and the church in
political, social and moral life. Further examples of parties based on reli-
gion are the Israeli Shas (Sfarad’s Guards of the Torah) and United Torah
Judaism, both based on religious conservatism.
The industrial revolution likewise created two political cleavages. The
most well know is the class cleavage which is the basis of the traditional
left–right ideological spectrum. Almost every European state is domi-
nated by the traditional socialist/social democratic/ labour party vs.
liberal/conservative party struggle. A partial exception is Ireland, where
reactions to independence are still a major organising factor. The two
main parties find their origins in support (Fine Gael) and opposition
(Fianna Fail) to the treaty that granted The Republic of Ireland (minus
the six counties making up Northern Ireland or Ulster) independence
from the United Kingdom. For that reason The Republic of Ireland is
seen as a somewhat ‘deviant case’ compared to other European countries
because its party system is lacking the traditional left–right class cleavage
(Gallagher, 1985). Many countries outside Europe, but with strong links
to Europe, are similarly affected by the class cleavage in some form,
including Australia and New Zealand. There are several exceptions,
including the United States and Africa, both of which we will look at
later.
Ideology 67
The second cleavage created by the industrial revolution was the

struggle between urban and rural concerns. As the industrial revolution
swept across Europe from the late seventeenth century onwards the agri-
cultural sector lost much of its former predominance to the new indus-
trial urban centres. Rural populations saw themselves sidelined and
ignored as urban concerns took centre stage. As a result people
mobilised, which sometimes lead to agrarian parties being formed, partic-
ularly in Scandinavia and Eastern Europe, where this aspect of political
life still plays a significant role in party politics.
A final element of Lipset and Rokkan’s ideas on how political cleav-
ages have structured the ideological makeup of political parties, is that, at
least when they wrote on this in the 1960s, these political cleavages were
seen as being more or less permanent. Most of the cleavages and the asso-
ciated parties were in place by the 1920s and very little had changed since
then – the cleavage structures had frozen in place. In other words, party
ideology can to a large extent be explained by the existence and overlap
of these four political cleavages. Not all countries will have all, some may
only have one, but in essence all parties are shaped by one or more of
these cleavages. Whether this still holds early in the twenty-first century
will be explored later.
Party families
The concern with ideologies has led to the development of one of the key
comparative tools in party politics: the idea of party families. Party fami-
lies are one of the most important ways in which parties are categorised
and compared both within countries and cross-nationally. Like the idea
of party types, the party family framework allows us to categorise parties
and thus helps us provide structure to the often fractured and confusing
party politics of democratic countries. However, a party family is dif-
ferent from a party type. Where a party type focuses mainly on organisa-
tional concerns, a party family is based primarily on ideology.
The party family is a way of grouping parties, principally according to
their links with one or more societal conflicts or cleavages. So, most
Christian Democratic parties are based on both religion and ‘anti-
socialist’ conservatism – that is, they are shaped by both the class-
cleavage and the church-state cleavage. Regional and separatist parties
such as those found in Scotland and Catalonia are affected by both the
centre-periphery and the class cleavages.
One of the main contributors to the party family approach is von
Beyme (1985). While von Beyme does not agree entirely with Lipset and
Rokkan’s view of political cleavages, the idea of party ideology being
shaped by societal conflict is still central to his exploration of party fami-

lies (von Beyme, 1985, p. 23). Von Beyme identifies nine party families as
being common to Western Europe and, according to some authors,
beyond (see Mair and Mudde, 1998).
The first party family to emerge was the Liberals. Liberal Parties, in
the cadre-party sense, evolved slowly over time beginning from the late
seventeenth century. However, the term ‘liberal’ as a political label did
not come into use in the English language until 1815 (von Beyme, 1985,
p. 31), but many of the ideas associated with ‘liberalism’, such a greater
individual freedom, were to be found among the so-called English
‘Whigs’ as far back as 1688. Liberal parties were originally dedicated to
promoting the interests of the emerging middle class against aristocratic
privilege and entrenched political power. They were often ranged against
supporter of the monarchy, and the privileges bestowed by the monarch
on loyal servants – privileges which included uncontested seats in parlia-
ment. Liberals have generally been seen as defenders of liberal democracy
and individual freedom.
The next party family to emerge was the conservatives. Early conser-
vatism is often associated with protecting royal and aristocratic privilege,
and has deep historical roots. However, as political organisations
Conservative parties emerged as a reaction to the rise of liberal parties.
Von Beyme writes that ‘Conservatism is a movement to defend positions
which are threatened or – as in the French Revolution – already lost’ (von
Beyme, 1985, p. 46). Hence, the reason for conservative parliamentarians
starting to organise in more formal structures was to defend the status
quo (or Royal privilege) against liberal demands for the enfranchisement
of the middle class and a reduction in the power of the monarchy and the
aristocracy. Nevertheless, conservative parties tended to end up accepting
most of the liberal party family’s demands for democratic equality.
Eventually the main struggle between these two families would not be
primarily about the differences between them. Instead it became about
which one of them would be the main defender of what they held in
common against socialist and social democratic parties. Hence, most
countries will have either a conservative or a liberal party dominating the
right wing of the political spectrum. Seldom will there be both types of
equal size.
The next party family to develop in opposition to both liberals and
conservatives, were socialist and social democratic parties. These parties
were created, often with trade union help, to defend the interests of the
working class against the property owning economic elite (in other
words, the interest groups behind liberal and conservative parties). In
most European countries socialist and social democratic parties form the
main governing alternative to liberal and conservative parties.
Ideology 69
While these three party families emerged in rough chronological order

(liberal, conservative, socialist/social democratic) the remaining party
families do not necessarily follow such a straightforward line.
Christian democratic parties form a party family often based on
catholic opposition to liberal secularism. Most have long since taken a
non-denominational (that is, largely ignoring Protestant/Catholic divi-
sions) route and are no-longer associated with any specific branch of
Christianity, but rather based on religious conservatism. They have
tended to form as an alternative to a strong conservative party in their
respective country. Where there is a strong Christian democratic party
there tends not to be a strong conservative party (for example Germany)
and vice versa (for example the United Kingdom and Scandinavia). In
short, Christian democrats and conservatives both fulfil the role as
centre-right opposition to the main centre left socialist/social democratic
party.
The Communist party family was created in the wake of the Russian
Revolution of 1917. Their main declared purpose was to defend the
working class against two main ‘enemies’. The first opponent was the
liberal and conservative party families – these, being as they were, often
associated with property (especially factory) owners and inherited wealth
and privilege. The second opponent targeted by the communists were the
moderate centre-left socialist and social democratic parties. The socialist
and social democratic parties were seen as having been co-opted by the
right-wing parties into the state and were, therefore, no longer effective in
defending the ‘true’ interests of the working class. Most communist
parties were originally formed with support from the Communist Party
of the Soviet Union (CPSU). Indeed the communist parties in many coun-
tries did not see themselves as separate parties but as ‘local’ branches of
the ‘mother party’ in the Soviet Union (this will be explored further in
Chapter 10). Later many broke with the CPSU and became ‘Euro-com-
munists’ – still communists, but no-longer beholden to the CPSU. Key
examples of this development include the Italian and French communist
parties.
A further party family is made up of agrarian parties created to defend
the interest of farming communities against the alleged neglect of the
urban elite – as outlined in the above discussion of the rural–urban polit-
ical cleavage created by the ‘industrial revolution’. In a number of coun-
tries agrarian parties had a brief time of success between the First and
Second World Wars, but failed to make much headway. This lack of
success was driven by the fact that their concerns were either too mar-
ginal (with the decline of agriculture and rise of industry) or were co-
opted by conservative parties (as happened in the United Kingdom). They
have had some success in Scandinavia, although here they have often
grown beyond the agrarian origins into broad centre-right parties. Very
few specific agrarian parties survive, with a few exceptions in Eastern
Europe.
A party family closely linked to the national revolution is made up of
regional and ethnic parties based very much on Lipset and Rokkan’s
centre–periphery cleavage. These emerged to defend the rights of minori-
ties against the national political centre in places such as Scotland, Wales,
the Basque Country and Catalonia. These parties are to be found on both
sides of the traditional left–right spectrum, albeit perhaps more often on
the left than the right. Wanting to loosen the ties of one part of a country
from central control, or even create a new country out of one part of an
exciting one, is a serious challenge to the status quo. Such a challenge
lends it self more easily to the left wing than the right wing.
One party family which has seen considerable fluctuations in support
are right wing extremist parties. These rose to particular (and cata-
strophic) prominence in the lead up to the Second World War, especially
in Germany, Italy and several East European states. This party family has
seen a re-emergence since the 1970s, mainly as protest parties against the
main-stream governing parties of the centre-right and centre-left. In
Scandinavia some emerged as a protest against what was seen as exces-
sively high taxes, which then morphed into protest against immigration.
The anti-immigration stance is common to many organisations in the far-
right party family.
Finally there is what von Beyme refers to as the ecology movement (or
Green Parties). These parties mobilised in protest against the growth-ori-
ented society, what they saw as excessive use of natural resources and the
environmental damage caused by modern industrial societies. At the time
when von Beyme was writing global warming and climate change were
not yet on the agenda, but have since become a central concern for this
party family.
Ideology in contemporary European party politics
The next big question is the extent to which the traditional views of ide-
ology, most notable the idea of fast frozen political cleavages, are still rel-
evant for the context, that is Europe politics, they were originally based
on. Can these four ideological divides still adequately explain party ide-
ology?
One of the key areas of debate in the context of European party ide-
ology is whether and to what extent we are seeing the rise of a ‘new’,
‘post-material’ or ‘post-industrial’ politics. This has been an ongoing
debate since at least 1977 when Inglehart published his Silent Revolution:
Ideology 71
Changing Values and Political Styles among Western Politics. He argues

that a number of key factors are changing the underlying values of voters,
something which will eventually feed through to political parties and
cause elites to change their ideological and policy positions. Basically
people tend to worry most about the needs that are most difficult to meet
due to a lack of supply in that particular area. They are much less
worried about needs that are easy to satisfy. For example, if food is plen-
tiful, but the weather arctic the need for shelter is going to be more of a
concern than the need to eat. In addition, there is a hierarchy of needs. In
other words, certain needs will take precedence over others. In this hier-
archy physiological needs tend to come first: first we need to eat; then
ensure our physical safety; then find shelter and so on. Once those basic
needs have been secured people will start to turn to satisfying other
needs, such as ‘love, belonging and esteem’ and ‘intellectual and aesthetic
satisfaction’. In short, the desire to live in a house with nice fittings and
beautiful carpets is going to be lower down the order of priorities if one
is starving to death.
In the context of modern party ideology the important thing to note is
that since the Second World War, Western Europe has experienced
unprecedented levels of prosperity and an absence of the wars between
great powers that have been a regular occurrence in the continent’s
history. This is not to say that there is not poverty and sometimes hunger,
but at least in Western Europe this is no longer on a mass scale. In short,
basic needs are more or less taken care off. ‘Old style’ left wing concerns
such as avoiding starvation in case of joblessness and the provision of
free healthcare have been largely dealt with.
As a result, so-called ‘post-materialist’ issues have come to the fore,
which, it is argued, are having a significant effect on the ideological
stances of parties. The old ideological positions based on materialist
needs no longer reflect the current demands of voters, which means that
if parties want to survive electorally they need to adjust their ideological
position. According to Inglehart (1977, pp. 40–50) these post-materialist
issues include wanting increased influence on political decisions, better
work-life balance and civil liberties such a freedom of speech.
The question now is whether, how and to what extent these changes
have affected political parties. If indeed the old political cleavages were
losing their power and are being replaced or at least joined by new con-
cerns a number of things might happen. One possible symptom of new
political cleavages coming for the fore would be the rise of new political
parties carrying the banner of ‘new politics’. This is after all how the tra-
ditional political cleavages found expression: with the rise of social demo-
cratic and labour parties representing one side of the class cleavage being
the prime example. Any rise in ‘new politics’ parties would be matched
by an equal decline in the support for parties based on traditional cleav-

ages. We now need to consider whether this has in fact happened.
Figure 4.2 shows the combined electoral support for the main party
families in 16 West European states starting in the 1950s and ending in
the mid-2000s. As we can see, two of the older traditional party families,
social democrats and Christian Democrats have indeed seen a decline in
voter support in the post-Second World War era. This would suggest that
some of the political cleavages that motivated them have declined in
power draining off voters to other parties. However, having said that,
despite their decline they still dominate the party systems of Western
Europe. The only two party families that are new on the scene are the
Greens and the New Left.
The Greens are definitely a reflection of a fairly recent concern with
environmental issues and could be seen as evidence of a rise in new poli-
tics not directly related to the traditional political cleavages. The new left
can also be seen as a reflection of new politics and certainly post-materi-
alist and post-industrial concerns. Many new left parties have found their
main drive in issues such as gender equality, civic liberties, anti-racism,
anti-homophobia and green issues, all key post-materialist/industrial
issues. However, these two groups taken together, while certainly man-
aging to make an impact, have never come even close to challenging the
parties based on traditional cleavages in a major way. There has been a
rise in the far right, and as we shall see this may in fact be a reflection of
a new political cleavage. However, again as with the greens and the new
left, their rise has been notable, but not enough to unseat the traditional
parties from their dominant position.
Hence, while there is some, albeit limited, evidence of post-
material/industrial concerns finding a place in West European party poli-
tics in the form of new parties, there is also the possibility that new
concerns are being absorbed into the ideological stance of traditional
parties. In other words, although they were created as vessels for tradi-
tional political cleavages, the older parties are working to stay relevant by
positioning themselves in relation to new political cleavages. It has been
argued that the so-called new-politics of the post-war years was relatively
smoothly incorporated into existing cleavage structures (Kriesi et al.,
2006). According to this argument, the four traditional cleavages in
Western Europe (centre–periphery; rural–urban; state–church; and class)
essentially boiled down to only two: a cultural cleavage related mainly to
the role of religion; and a socioeconomic one based on social class (Kriesi
et al., 2006, p. 923). These two cleavages lost much of their potency with
the rise in both prosperity and secularization in post-war Europe, but
were revitalised by the rise of ‘new politics’. New politics added concerns
over social justice to the class cleavage and transformed the cultural
Ideology 73
40
35
30
Social Democrats
Communists
25 New Left
Percent
Greens
20 Christian Democrats
Conservatives
15 Liberals
Agrarian
10 Far Right
Other
0
1950s 1960s 1970s 1980s 1990s 2000s
Figure 4.2 The electoral support for party families
Source: Gallagher, Laver and Mair (2011).
dimension’s focus on religion to a broader debate between social liber-

alism and social conservatism generally. Hence, ‘the new social move-
ments did not add any fundamentally new dimensions to the political
space, but transformed the meaning of the two already existing ones’
(Kriesi et al., 2006, p. 924). What ‘new politics’ added to party politics
was, therefore, less the rise of new parties with ideologies driven by new
conflicts, but more of a transformation of existing parties and their ide-
ologies.
Globalisation has also had an effect on the ideological stances of polit-
ical parties. Globalisation has created new conflicts between what has
been referred to as the winners and the losers from the globalisation
process (Kriesi et al., 2006). The winners are those working in sectors
that thrive from to international competition and thus welcome globalisa-
tion – banking being an obvious example. The losers are people who
strongly associate themselves with traditional national values; those
working in sectors traditionally protected from international competition,
but increasingly forced to open up; and all unskilled workers. However,
as with new politics, globalisation has been incorporated into the existing
cleavages. In this context the class cleavage will now be dominated by a
debate over whether to open up the national economy to international
competition, or use the state to protect it from world markets. The cul-
tural dimension will take on a distinct national and ethnic character: the
protection of ‘national values’ against multiculturalism brought on by,
among other things, immigration.
In essence, it has been argued that existing parties have a very large
‘adaptive capacity’ (Kriesi et al., 2006, p. 924). It is suggested that main-
stream parties of both the left and the right will adopt moderate ‘global
integration’ positions (Kriesi et al., 2006, p. 926). This also suggests that
globalisation has created a powerful new demand from the ‘losers’ for
protection against the effects of globalisation. Those parties that are able
to meet that demand best will be a driving force for change in existing
party systems. So, just as the demands of the ‘losers’ of industrialisation –
the working class – created a long-standing struggle between left-wing
and right-wing parties, so the ‘losers’ from globalisation will create a new
dynamic between ‘integrators’ and ‘demarcators’ – that is, between those
who want to embrace globalisation and those who want to impose limits
on it. What we are seeing at the moment is that the far right has been
most efficient in putting together a ‘highly attractive ideological package’
for the ‘losers’ (Kriesi et al., 2006, p. 929).
So, the new politics emerging from the 1960s onwards forced existing
parties to change their ideological makeup to cope with new versions of
old conflicts. Globalisation created the need to adjust ideological offer-
ings to appeal to the ‘losers’ from the globalisation process. While the
‘new politics’ wave meant principally positioning oneself on issues of
morality (gay rights, abortion and minority rights), so globalisation has
meant dealing with the clash between integration and demarcation. So far
the extreme right has taken the lead in a demarcation direction, forcing
both the moderate left and right to take on demarcation positions to
answer that challenge, often in the form of moderate anti-immigration
policies.
Kriesi’s idea that the mainstream centre right and centre left parties are
adopting a moderate integrationist approach hints at another popular
argument when it comes to the ideological development of parties, or at
least the main moderate centrist parties: that of convergence on a neo-
liberal free market agenda. Some suggest that the traditional cleavages
have not just thawed from their previously frozen state to be replaced by
new conflicts, but that by and large we have seen ‘the successful resolu-
tion of the societal conflicts which had been embodied in the traditional
cleavages’ (Kriesi, 1998, p. 165). That is, the old conflicts have been
solved, or perhaps rather won by the supporters of the free market, and
now all that is left is a disagreement over who is best able to manage the
day-to-day running of the state. This is also reflected in the idea of the
‘third way’ taken by some social democratic parties, most notably ‘New’
Ideology 75
Labour under Tony Blair and Gordon Brown, and to a lesser extent the
SDP under Gerhard Schroder (Giddens, 1998, 2000; Gould, 1998). The
third way is seen by its supporters as achieving traditional social demo-
cratic ends, but to a greater or lesser extent using the free market to do so
– and by its critics simply as selling out to the neoliberal dogma. At the
heart of this view of convergence is the popular refrain that ‘they are all
the same now.’
While this point of view is popular there are many who argue that the
convergence theory is in fact somewhat exaggerated. Some (for example,
Haupt, 2010) argue that the main parties of the centre left and the centre
right are in fact not converging but are keeping in tandem and main-
taining their ideological distance – if one party moves to the middle it is
usually because the other has vacated that political space. Others, such as
Kriesi (et al., 2006, see above; Kriesi, 1998) have argued that while the
societal conflicts that spawned parties may have been modified by suc-
ceeding waves of societal change, they are still more or less intact, if in
somewhat changed forms. Finally, there are those who argue that the
level of working class socialism on the centre-left has long been exagger-
ated and that the ‘third way’ is in fact not so radically different from tra-
ditional social democratic positions (see, for example, Hinnfors, 2006).
A final possibility when it comes to how parties have reacted to soci-
etal change is that they have to a large extent ignored it. Williams (2009)
in her study of the catch-all phenomenon in France found that while the
main parties had indeed taken up a catch-all position, including moder-
ating their ideology as expected from the catch-all approach (see Chapter
3) this had not in fact been driven by ideological changes in society.
Rather, the institutional structure of French politics has pushed parties in
a catch-all direction. The main parties need to attract a wide range of
voters to win a presidential election and gain enough seats to control a
majority in the National Assembly. This is driven by a very dispropor-
tional electoral system and has forced the main parties to develop a polit-
ical stance designed to be as inoffensive as possible. The name of the
game is to avoid scaring anyone away – which means avoiding strong
ideological positions. This does not mean that voters do not want ideo-
logically committed parties, but for parties to win they have to attract
people from a wide range of ideological positions. To do that, parties
need to adopt moderate ideological positions. One of Williams’ conclu-
sions is, therefore, that in France ‘One implication of these findings may
be that for France ideology exists rather independently of the party
system. The importance placed on parties by Kirchheimer as the vessels
and perpetuators of ideology may have been misplaced in the French
case’ (Williams, 2009, p. 610). In other words, at least in the case of
France, the ideological position of parties seems not to be related to
‘demand side’ factors – what the voters want – but to the structure of the
political system. The demands of the electorate are, therefore, less of a
deciding factor in the ideological position of parties than might otherwise
have been expected.
That this may indeed be the case in a wider context beyond France has
been suggested before. Ware (1979) has argued that partly due to the lack
of internal democracy in most political parties (see Chapter 7) and the
fact that new parties find it difficult to survive, the leaders of older more
established parties are able to ignore new political divisions and keep to
tried and tested ideological positions. The problem is that these positions
‘by their very nature hark back to a vision of political reality that is prob-
ably obsolete’ (Ware, 1979, p. 51). Because demands from the electorate
are not filtered through to the party leaderships, and because new parties
are unable to effectively challenge old parties, leaders of established
parties are able to keep to ideological positions which may no longer
reflect the underlying tensions and conflicts in society. Party leaders are
able to both ignore internal demands for change and exclude most chal-
lenger parties from competing effectively. The result may be that parties
and the electorate are drifting further and further apart, leading to dissat-
isfaction with the political system.
Party ideology beyond Europe
A lot of the discussion of ideology presented above is heavily based on

European party politics. There is an open question whether the existing
debates about party ideology, based on the European context, is relevant
outside Europe. To consider that issue the following will look at three
important geographical areas: the United States as the most powerful
democracy in the world; India as the largest democracy in the world; and
Africa containing some of the youngest democracies in the world.
United States
First, we will look at the United States. The United States is one of the
oldest democracies in the world and is also the home to some of the
oldest continuously existing parties. However, the United States is in
many ways different from the traditional European heartland of party
politics, not least so when it comes to the matter of ideology. The first
thing to note is that the two parties, Democrats and Republicans tend to
be seen, at least from the European perspective, as remarkably similar. In
his analysis of party families across a range of countries Ware (1996, p.
Ideology 77
25) put both parties in the liberal party family. It is however, worth
remembering that ‘liberal’ has different connotations in the United States
than it does in most of the rest of the world. In the context of US politics
liberal is commonly seen as indicating left of centre. Very few
Republicans would identify themselves as ‘liberal’. Indeed, ‘liberal’ is
regarded by some parts of the political right in the United States as a term
of abuse.
Looking at the United States in the early post-war years Hofstadter
(1948, p. xxxvii) expressed a commonly held view when he wrote that
‘the fierceness of the political struggle has often been misleading for the
range of vision embraced has always been bounded by the horizons of
property and enterprise.’ This very much supports the idea that both the
Republicans and Democrats are in essence liberal parties. While Gerring
(1998) has argued that the Republicans are sometimes seen as conserva-
tive and the Democrats liberal, he also points out that it is not entirely
clear what this means in European terms. Ware flatly states that the
Republican Party simply cannot be seen as a conservative party. The
problem according to Ware is that the terms liberal and conservative have
taken on different meanings in the American context (Ware, 1996, p. 26).
He argues that Roosevelt ‘hi-jacked’ the term liberal as a label for his
‘New Deal’ which was in essence a massive intervention by the state in
the market to deal with the problems of the Great Depression in the
1930s. As a result of this usurpation of a term usually associated with
small state politics, those opposed to state intervention was left without a
label. Hence in the 1960s ‘conservative’ eventually became to be associ-
ated with the anti-state position taken by the Republican Party. In the
European context ‘conservative’ is most associated with an adherence to
traditional values. A small state position may also be found in many con-
servative parties, especially since Thatcher added a strong element of neo-
liberal laissez faire to the conservative position. However, traditionally in
Europe ‘conservatism’ is not synonymous with ‘small-state’ or unfettered
free market (see, for example, Gray, 1997).
In short, ideological terms do not have the same associations in the
American context as they do in the European. The ideological differences
between the Republicans and the Democrats are not as easy to under-
stand as is the case when looking at the traditional divide between
socialist/social democrats and conservatives/liberals in the European
party politics context. However, that is not to say that there are no differ-
ences between the two parties. There clearly are. In his comprehensive
overview of ideology in the American context Gerring (1998) argues that
the two parties have been based on long-standing differences and two rel-
atively continuous standpoints, even if those standpoints have been
expressed in very different ways over time.
According to Gerring the essence of the Republican ideological posi-

tion is one of national (if not always individual) prosperity, social order
and patriotism. While this position has been present throughout
Republican Party’s history it has meant different things in terms of policy
at different times (Gerring, 1998, pp. 14–19). In the period 1828–1924
the Republican position was highly statist. They took a mercantilist (that
is protecting domestic industries from international competition) position
to national prosperity and saw the state as having an important role to
play in protecting the prosperity of the nation from foreign competition.
In addition, the state was seen as having an important role in maintaining
good social order. Gerring argues that while national prosperity, social
order and patriotism remained the main Republican bedrock, by 1928
the means to their realisation had changed. From 1928 and to the present
day the Republicans have come to view the state with suspicion.
Increasingly the Republicans came to see the free market as the best route
to national prosperity. In addition civil society, rather than the state, was
the main source of social order. Hence the state should be kept out of the
affairs of individuals as far as possible. The exception, as Ware points
out, is when certain practices, seen as threatening traditional social
values, such as abortion and homosexuality, have to be banned or at least
limited. Hence the modern Republican position is one of economic free
market liberalism and social conservatism, the main role of the state
being to occasionally help out with maintaining the latter and largely
staying out of the former.
The Democrats by contrast have had a long standing concern with
equality (Gerring, 1998, pp. 19–20) and have gone through three dif-
ferent phases in terms of how equality was interpreted and promoted. In
the period 1828–92 equality was about personal liberty free from state
intervention, with the role of the state being limited to protecting prop-
erty rights. Liberty was in much of this period limited to the white popu-
lation – meaning it excluded African-American slaves.
The second phase was dominated by what might be called populism:
a defence of ‘the people’ against special interests, particular ‘big busi-
ness’ and monopolies, and in support of some limited redistribution of
wealth by the state. Hence, the state was now seen as a means by which
‘the people’ could be protected from special interests and to improve
demand-led economic growth through state spending on, for example,
big infrastructure projects. This is most clearly exemplified by the New
Deal.
The final phase, starting from 1945 onwards, focused on the extension
of civil rights to previously excluded groups. Hence the appeal was less
one of class (the people against big business) but rather one of margin-
alised groups against unjust exclusion from social and economic rights. In
Ideology 79
addition progressive social policies were to be achieved less through the

state and more through an active civil society.
Hence, the Democrats have according to Gerring had as their core
principle the protection of individual liberty, sometimes from the threat
of the state, sometimes from exploitation by privileged groups, and some-
times from racial and other forms of discrimination. The Republicans
have traditionally seen national prosperity, social order and patriotism as
their main and enduring focus. At various times the state has been an
enabler of or a hindrance to these principles, but the principles have
remained largely stable. The differences between the Republicans and
Democrats have therefore been real and enduring, but are not easily
translated into traditional categories of ‘liberal’ and ‘conservative’ as
understood in Europe. The two parties may both exist within a common
economic liberal framework of ‘the virtues of capitalist culture’
(Hofstadter, 1948, p. xxxvii) but there has been and remains a wide
scope for disagreement within that framework.
India
The second party system outside of Europe we will look at is India. India
is in many ways a remarkable case and is notable for, among many other
things, being the world’s largest democracy. That alone makes it worthy
of attention. However, it is also worthy of attention because, unlike so
many other post-colonial countries, it has managed to maintain a rela-
tively well functioning democracy – no mean feat considering its sheer
size and diversity. The question is then how the European notions of ide-
ology fit within the Indian party political system – and the answer is, as
with the United States, not all that well.
As mentioned in Chapter 2, for a considerable time after it gained its
independence in 1947, India was dominated by the Indian National
Congress (INC) in a predominant party system. The INC won elections
by massive majorities and controlled both national and local government
until the 1960s. To understand the role of ideology in the INC it is neces-
sary to keep in mind this dominance. No party achieves such dominance
in a democratic system, relatively free of fraud and corruption, without
sacrificing a good deal of ideological coherence. As Krishna (1967, p. 29)
writes, the INC engaged in ‘promiscuous accommodation of divergent
elements’ in order to maintain its dominance.
This means that thinking in terms of ideology does not necessarily
make much sense when looking at the INC as a whole. It is true that the
INC did adopt socialism as its official ideology in 1955 (Hardgrave and
Kochanek, 2008, p. 271). However, this was, at most, the official line of
the national leadership and certainly a very poor reflection of the party’s
actual stance, especially the further towards local politics one got. The
INC was for a significant length of time successful at politicising and
integrating large sections of the electorate which had previously not had a
great involvement in politics. However, while these newly politicised ele-
ments of society might have been integrated into the party organisation,
their integration into the party’s official stance of socialism, economic
development and democracy was less successful. Often, newly integrated
segments of the electorate were ‘caste conscious and parochial in orienta-
tion’ (Hardgrave and Kochanek, 2008, p. 273). In short, as the electorate
was politicised by the INC, so the INC was traditionalised by the elec-
torate (Brass, 1966, p. 2; Hardgrave and Kochanek, 2008, p. 273).
As the INC’s dominance started to fade in the 1960s so new parties
emerged – often based on issues not directly related to the traditional
left–right spectrum. Especially religious and regional concerns started to
dominate rather than traditional socioeconomic positions in the tradi-
tional European sense. According to Verney (2004) there are five types of
parties in the Indian party system each with different approaches to ide-
ology (taken in a very broad sense).
One group is made up of various communist parties. While these
parties are not based on caste or religion, their support tends to be con-
centrated in a few regions, especially West Bengal. After the end of the
Cold War these parties have suffered from the general decline in support
and ideological confusion common to many communist parties
throughout the world. Most of them have moved away from revolu-
tionary rhetoric in a more reformist direction.
Another group is the so-called transitional parties which rarely survive
from one election to the next. They tend to be focused on a very specific
and often local issues or the personal appeal of an individual politician.
Then there are parties that are based on fairly narrow religious, caste
or tribe appeal. These parties tend to be geographically concentrated,
neither able nor willing to expand their appeal beyond a fairly specific
group.
The fourth group is similar to the third, but instead of religious, caste,
or tribal foundations they are based on ethnic and regional identities such
as the Tamil population in the south east.
The final group is made up of parties which try to be or actually are
‘national’ parties. In other words, they approach the catch-all model by
trying to have a presence throughout the country and attract a wide
range of voters. The only two parties to really achieve this position are
the INC and the BJP. The BJP is very much based on Hindu nationalism
and as described above the INC has been increasingly taken over by a
plethora of local caste, religious and tribal concerns.
Ideology 81
In short, while the INC formally adopted a socialist position shortly

after independence, ideology has increasingly been driven out of the
Indian party system by group identities based on religion, caste, locality
and tribe. To understand the position of Indian parties, European party
families appear to be of little use.
Africa
The final area we will look at is Africa. It is probably fair to say that suc-
cessful, stable multiparty democracies have not been a regular feature of
post-colonial African politics. Early attempts at multi-party democracy in
the years immediately after independence very quickly descended into
single-party-ism or military dictatorship. However, the 1990s, in what
has been called the ‘third wave of democratisation’, saw a ‘dramatic’
move away from single-party and military rule (Manning, 2005, p. 715).
Although party-based democracies are now becoming more common
on the African continent, the party systems in the new democracies have
tended not to be based on socioeconomic cleavages – in contrast to
European party politics (Burnell, 2001, p. 251; Manning, 2005, p. 715).
The key socioeconomic cleavages in Europe, in particular the class
cleavage, have been based on how the proceeds of economic growth,
especially growth created by the industrial revolution, should be distrib-
uted. This also implied that there were a range of social and economic
policies to choose from – that is, whether and to what extent the state
should be involved in the management of the economy and the extent of
the freedom of the market. As Manning (2005, p. 720) argues:
the presumption in the comparative party literature is that by pursuing

certain sets of economic policies rather than others, parties in
advanced industrial democracies will redistribute national income to
their constituents. Economic policies determine the degree to which
growth benefits labour [workers] relative to capital [owners], or the
upper income groups relative to the ‘working poor’, for example.
In Africa, parties have had a much narrower range of policies to choose

from, largely due to chronic economic crisis and consequent aid depend-
ence (Manning, 2005, pp. 716, 720). Many African states rely heavily on
foreign aid, which often comes with ties attached as to how it should be
spent. The government in question, therefore, does not have much flexi-
bility to choose one economic policy over another. Whitfield (2009)
shows that in the case of Ghana the two main parties, the National
Democratic Congress (NDC) and New Patriotic Party (NPP) self-identify
as social democratic and right of centre respectively. However, when it

comes to policies actually pursued while in government, the parties have
very little scope for divergence due to the country’s dependency on
foreign conditional aid (Manning, 2005, pp. 715–16). In other words,
donor countries are to a significant extent able to determine the economic
policies pursued – whichever party is in power. The differences between
African parties are, therefore, not based to any significant degree on
socioeconomic mobilisation since such mobilisation would require diver-
gent economic policies.
One traditional view of African party politics is that rather than being
based on socioeconomic mobilisation parties rely on politicising tribal
and ethnic cleavages. However, while tribal and ethnic ties do play a role
they should not be exaggerated (Basedau and Stroh, 2009; Burnell, 2001,
p. 250; Norris and Mattes, 2003). Several studies have found that per-
sonal ties, often based on the region a politician is from, are far more
important than ethnicity (see, for example, Basedau and Stroh, 2009). So,
while there may be an overlap between tribal and regional origins, they
are not identical and regional ties outweigh tribal ties. These regional ties
tend to be based on so-called neopatrimonal and clientilistic relationships
(Basedau and Stroh, 2009, p. 20; Burnell, 2001, p. 253; Manning, 2005,
p. 715). In essence, politicians gain the loyalty of voters by ensuring that
voters gain some direct, perhaps even monetary, benefit in return for their
vote.
The main difference between parties and politicians are, therefore, not
based on disagreements over socioeconomic policies, but based on who
should benefit from access to the resources of the state. A politician’s
primary goal is not to implement certain policies but to gain access to
state resources which can feed personal networks in their respective
regions. African party politics, therefore, seem to be based on a personal
‘scramble for the spoils of office’ (Burnell, 2001, p. 253) rather than on
ideological differences. This is well illustrated by the fact that in Zambian
politics, prominent politicians frequently change their party allegiance,
taking with them significant numbers of members and supporters. People
are literally following the money when they join a politician in moving
from one party to another (Burnell, 2001, p. 251). Again, as with the
United States and India, the ideological battles identified in European
party politics seem a poor guide to understanding what drives competi-
tion between parties in African democracies.
Conclusion
Based on the above, it seems that many of the common assumptions

Ideology 83
about ideology only really apply in the European context where they
emerged in the first place. Here it still makes sense to talk about tradi-
tional ideologies even if they have changed considerable from their early
origins. However, as soon as we move outside the European party
systems, ideology based on socioeconomic cleavages decreases signifi-
cantly as the basis for policy choices in political parties.
In short, it seems that party politics in the European context can still
be understood in ideological and party family terms. This is significantly
less the case in the other three cases we have looked at. In the United
States terms such as ‘conservative’ and ‘liberal’ certainly have real
meaning, but a meaning very different from the European context. In
India, tribal, religious and ethnic ties have been rapidly replacing ide-
ology as a mobilising force. In many African democracies personal ties
ensuring access to the resources of the state are more important than
either ideology or ethnicity in explaining party allegiance and mobilisa-
tion. Hence, when considering the issue of party ‘ideology’ in a given
region is seems advisable to focus on local issues, rather than using pre-
existing models and assumptions derived from different contexts. Indeed,
in many cases, such as India and African democracies, the very idea of
ideology is difficult to apply as other concerns are of greater importance
in determining party action.
Chapter 5
Party Members, Activists and

Supporters
The mass party as described by Duverger is often seen as the standard of

what a political party should be: an organisation based on a mass-mem-
bership organisation, open to anyone who is willing to support the goals
of the party. In the mass-party model, party members are the life and soul
of the party – not only campaigning on the streets to get their candidates
elected, but also keeping those candidates in touch with the views and
demands of the voters once elected. Clearly, from this perspective party
members are an important element of party politics and, therefore, some-
thing requiring our attention.
As we saw in Chapter 3 the existence of party members is historically
speaking a relatively new phenomenon – created, according to
Duverger, by the mass parties of the late 1800s. This chapter will con-
sider a number of crucial issues related to party members. The chapter
will first consider what is meant by terms such as party member, activist
and supporter. The second section will examine the development in
membership numbers in European political parties. The focus will be on
Europe as this is where membership numbers are easiest to come by,
especially for Western European parties. Parties in the new Democracies
in Eastern Europe have only recently started to settle down after a
period of very rapid change, with parties emerging and disappearing
again in short order. Parties in the United States ‘do not have any kind
of formal membership’ (Ware, 1996, p. 90) and cannot, therefore, be
analysed in those terms. Having looked at changes in party membership
numbers the chapter will then consider at length why people might
want to join a political party and what reasons party leaders may
have to either encourage or discourage membership recruitment. The
chapter will then explore some of the reasons behind the changes in
membership numbers and concludes by considering the consequences
for democracy of the apparent decline in party membership organisa-
tions.
84
Party Members, Activists and Supporters 85
Defining party member, activist and supporter
Broadly speaking, ‘party members’, ‘activists’ and ‘supporters’ make up

the ‘party on the ground’. However, what each category actually involves
is far from clear.
Defining exactly what is meant by ‘party member’ is not straightfor-
ward. Different parties have different ideas of what is behind the term,
and different ways of registering who they regard as party members. The
easily identifiable ‘type’ of party member is the card-carrying and dues-
paying member. This type of member will literally have a membership
card and will pay a regular and specific amount of money as a member-
ship fee. Being a card-carrying member will entitle the individual to
certain privileges, such as being able to stand and vote for positions in the
party. These positions will include being on local, regional and national
committees or being the party’s candidate for publically elected assem-
blies (such as local government councils or national parliaments).
However, the difficulty is that some parties afford such privileges to those
who have merely expressed some form of support for the party, but are
not dues-paying members holding a membership card. A key example is
the United States where the two main parties do not have formal card
carrying members, but registered supporters.
A further noticeable element of the US system is that in some US
states, voters do not register as supporters with the party. Instead, when a
citizen registers to vote many states allow citizens to also register their
support for one or another of the main parties. In other words, they
declare their alliance for a party with the state and not the party. This is
clearly different from being a paid up card-carrying member, but in many
states will still give the same kinds of privileges in terms of standing and
voting for internal posts.
The card-carrying dues-paying party member is often found in the
most well established parties in longstanding European democracies. This
type of membership tends to be associated with parties that adhere most
closely to the ‘mass party’ organisational model. However, a plethora of
more recent parties have far less formal membership structures, where
some loose form of registration is enough to count – but also grants very
little in the way of privileges. For example, when what is now the Liberal
Alliance in Denmark was first founded, members were simply those
people who had registered an interest in the party on their website. The
Italian Forza Italia was based on a network of very informal ‘clubs’ set
up through the sales force of Berlusconi’s commercial company Fininvest
(Hopkin and Paolucci, 1999).
Further, even within the more formal membership arrangements there
are different ways in which one can belong to a party. Heidar (2006, pp.
303–3) argues that there are three ways in which one can be a formal
member of a party: individual; auxiliary and collective. The individual
member is the person who has paid a membership fee directly to the
party, and may well have been issued with a membership card (although
not all parties issue them anymore). Auxiliary membership is gained
through organisations attached to the party, such as its youth organisa-
tion. A party’s youth organisation may formally speaking be a semi-inde-
pendent organisation and, therefore, be separate from the national party.
So, being a member of the youth organisation makes you an auxiliary
member, but not automatically a direct member of the party. However,
someone who is a member of, for example, the party’s youth organiza-
tion might also at the same time be an ‘individual’ member – that is, be
paying a membership fee to both the youth organisation and the national
party. This somewhat blurs the boundary between the two. Finally, col-
lective membership is through an organisation, such as a trade union,
which is formally associated with the party in question. The trade union’s
members are at the same time collectively members of the party. As with
auxiliary members, a fees-paying member of an affiliated trade union
may also be a direct member of the national party.
The most notable example of such ‘multiple memberships’ is the
British Labour Party. A person can be a direct fees paying member of the
national party; a member of, say, UNISON, a trade union affiliated to the
party; of Young Labour – the party’s youth organisation; of the Fabian
society – a think tank affiliated to the party; and of the Co-operative
Party – an organisation which is affiliated to the Labour Party and sup-
ports candidates for public office. This way one would be a member of
the party five times, but in slightly different ways. It would also give such
a person five votes in electing the party’s leader.
The exact definitional boundary between ‘member’ and ‘supporter’ is,
therefore, far from clear, and there is a considerable grey area between
the two categories. It also makes calculating the membership strength of
political parties very difficult, something which will be considered further
later.
As mentioned earlier, being a member gives certain privileges.
However, depending on the party’s culture, becoming a member may
demand meeting certain requirements. Especially in the early to mid-
twentieth century some socialist parties had very stringent membership
requirements. So, the Argentinean Socialist Party in the 1930s required a
new member to formally and in writing declare his or her support for the
party’s goals, rules and methods. In addition, the applicant needed the
support of two existing members with at least six months membership
(Heidar, 2006, p. 302). Up until the middle of the twentieth century it
was not unusual for British political parties to tell applicants that their
local branch was ‘full’ and that they would have to wait until a place
became available. In addition, many parties have requirements for how
long a person has to be a member before acquiring the full privileges of
membership, that is the right to stand and vote in internal elections. It is
probably fair to say that as many parties (although not all) have seen a
serious decline in their membership numbers the requirements for joining
and the time it would take to acquire the full privileges of membership
have declined significantly (Heidar, 2006, p. 301).
There are also considerable differences in the level of loyalty that
parties try to instil in their members. There was a longstanding tradition,
especially among some major European social democratic and socialist
parties, to encourage very strong links between party members and the
party organisation. This connection was so strong in some parties that
they were referred to as ‘cradle-to-grave’ parties, a term sometimes par-
ticularly associated with Austria. However, the idea is also to be found
elsewhere. In the context of Denmark, Pedersen (1989) describes how
membership of the Social Democrats could be an all-embracing experi-
ence. A party member might have his or her money with the Workers’
Bank; buy the workers’ co-operative’s beer, bread and milk and ‘when
one had popped one’s (workers’ co-operatives’) clogs, and could not be
resuscitated by the workers’ Samaritans, then the Workers’ Burial
Association would deal with the final arrangements’ (Pedersen, 1989, pp.
56–7). For some, the party was almost literally from cradle to grave. This
though is no longer the reality of membership for the vast majority of
members. For most members of contemporary parties, membership
means little more than paying their membership fee.
However, regardless of whether one is a member or a supporter, one
may or may not also be an activist. As with ‘member’ and ‘supporter’
how to define an ‘activist’ is not clear. Clearly, an activist is someone who
is active in the party. There is though no clear definition of how active
one has to be to be regarded as an activist. Some people spend a signifi-
cant proportion of their free time doing voluntary work for their party –
campaigning on the doorstep, sitting on committees, and attending con-
ferences. Others limit themselves to displaying a poster for the party in
their window at election time. However, regardless of how formal a
person’s attachment is to a party, it is very clear that many modern
parties are very keen on encouraging their members and supporters to
also be activists. As we shall see in Chapter 8 local volunteer activist-
based campaigning plays an important part in election campaigning.
What should be clear by now is that a very significant element of a
political party, the so called party on the ground, is a very amorphous
entity. The party on the ground is, in some form, a key element in the
vast majority of parties and carries out very important tasks. However, it
is clearly not an easy entity to define in precise terms. The changes that
the party on the ground is undergoing and the consequences of these
changes will be considered next.
The rise (and fall?) of party members
One of the key pillars of the argument that parties are in crisis is that
party membership is in decline. This then assumes that a mass-member-
ship organisation is something a party ought to have. Clearly political
parties have not always had membership organisations.
Duverger claimed that the mass-membership organisation was
invented by socialist and social democratic parties in the mid-to late
1800s. There is some evidence that this may not be entirely accurate. For
example, in the United Kingdom both the Conservative Party and the
Liberal Party had already started organising their supporters well before
the time the Labour Party was set up in 1900 (Scarrow, 1996, p. 5). Also,
some religious parties had mass-membership organisations by the time
the left wing and trade union-based mass parties described by Duverger
emerged (Ware, 1996, p. 66). Whatever is the case, the main argument in
much of the literature is that the idea of mass-membership organisations
emerged in the late 1800s and by the mid-1950s had become the stan-
dard way of organising political parties. Indeed, as was described in
Chapter 3, the mass-membership party was briefly regarded as the party
of the future and was seen as a new step on the party evolutionary ladder,
superior to its predecessors (Scarrow, 1996, p. 5). However, this view did
not last long and was soon replaced by arguments surrounding the prob-
able and even inevitable decline of membership organisations.
In considering the accuracy of the claim that party membership organi-
sations are declining we are faced with several problems. As we saw
earlier, the nature of the party on the ground is somewhat amorphous,
which makes it difficult to establish how many members a party has.
First, and as we saw earlier, it is not always clear what counts as a ‘party
member’. It is usual to separate a ‘member’ from a mere ‘supporter’ –
that is, someone who votes for the party on a regular basis, and may even
sometimes volunteer during elections, although the distinction is not
always very clear (Duverger, 1964, p. 62). Perhaps, the easiest way to dis-
tinguish between a member and a supporter is that a member has signed
a membership form and agreed to pay regular fees for their membership
(Duverger, 1964, p. 71). However, particularly in the United States, this
kind of fees-paying, card-carrying member is not the norm. By Duverger’s
definition US parties are largely made up of supporters rather than
members in this sense.
To complicate the picture further, there is also the concept of indirect

member mentioned above. This is particularly important in the case of
parties set up by trade unions and has been a distinguishing feature of the
British Labour Party. The British Labour Party was originally set up by a
group of trade unions and socialist societies, such as the Fabian Society.
For the first 18 years of its existence it did not have any direct individual
members. Its membership consisted exclusively of people who were
members of the trade unions and societies who had affiliated to the
Labour Party. The members of the affiliated organisations where then
also ‘indirect’ members of the Labour Party. Only in 1918 was it possible
to join the Labour Party directly. This means that while the Labour Party
has typically had a direct membership numbering in the hundreds of
thousands, it has had an indirect membership numbering in the millions
(Laybourn, 2001, p. 177). So, when the Labour Party’s membership
organisation reached a peak of 1,014,524 direct individual members in
1952 its indirect membership was 3,071,935 – that is, more than three
times as large.
Second, even once we have decided how to define ‘party member’,
getting hold of membership numbers is notoriously difficult. Some parties
are reluctant to share that kind of information with outsiders. This may
be because a particular party is generally secretive about their internal life
– a feature of many parties on the ideological fringes. Some parties may
be reluctant to share membership numbers because they fear that it will
make them look weak compared with other parties in the country.
However, it is just as likely that a party may simply not know how
many members it has. It may sound strange that a party is unaware of its
own membership strength, but looking at the work of Duverger and
Panebianco (outlined in Chapter 3) it actually makes sense. The problem
arises because some parties are highly decentralised. If a party is very
decentralised, the centre may have little knowledge of, or indeed interest
in, what happens at the local level. Hence, each local branch may know
its own membership strength, but this information is not passed on to the
centre. In the case of such a party anyone wanting to establish the total
number of members would have to contact each local branch, of which
there may be several hundred.
The difficulty in examining the development in membership numbers
is illustrated by two important journal articles on the issue (van Biezen,
Mair and Poguntke, 2011; and Whiteley, 2011). While, as we shall see,
both articles reach the same overall conclusion as to the movement in
membership numbers, there are some considerable discrepancies in the
data for individual countries. Van Biezen et al. rely on data gathered
from country specialists. Whiteley uses data from the World Value
Surveys. In addition to using different sources the two papers use
slightly different ways of presenting the data. This makes direct compar-
ison difficult.
However, some issues still stand out. Van Biezen et al.’s data shows
that party membership has fallen dramatically in Slovakia in the period
2000–7. Whiteley’s data shows a slight increase in 1989–99 to 1999–
2004 which are the periods during which the surveys he uses were carried
out. In France van Biezen et al. see an increase in membership in the same
period as above. Whiteley’s data shows a decline. Van Biezen et al. (2011,
p. 28) note that Austria is a country with continuing high levels of party
membership. Whiteley points to Austria as a West European country with
a high level of decline (2011, p. 25). There are several similar discrepan-
cies between the data in the two papers. This is not to criticise either of
the papers, but to illustrate the difficulties in analysing party membership
numbers.
However, both papers agree that party membership is generally in
decline. Looking at 23 European countries (East and West), van Biezen et
al. (2011, p. 32) find that in all but six of them, membership numbers
had declined, sometimes quite dramatically. They note that several
former Communist countries have seen very dramatic declines which they
ascribe to Communist Parties losing their dominant position in society.
Whiteley (2011, p. 25) makes a similar observation. Both articles also
note that several West European countries have seen major declines
including the United Kingdom, Germany, and Finland – something that
can clearly not be ascribed to the sudden change in the political system as
in the case of former Communist Eastern Europe. Whiteley also finds
that membership numbers have declined outside Europe. His data shows
decline in Chile, South Korea, Mexico and Canada. He does find a small
increase in Japan (Whiteley, 2011, p. 26). The overall conclusions are
fairly clear: Whiteley (2011, p. 36) writes that ‘party activism and mem-
bership have been declining across most of the democratic world.’ Van
Biezen et al. (2011, p. 42) likewise conclude that ‘while political parties
continue to play a major role in the elections and institutions of modern
European democracies, it seems that they have all but abandoned any
pretensions to being mass organisations.’
The argument that party membership organisations are in decline
raises the question of what they are declining from. Presumably, there
was a time when they were in a better condition than now, and it is from
this former state that they are declining. There is sometimes a suggestion
that there was some form of a ‘golden age’ of mass membership organisa-
tions – and that this golden age was located sometime in the middle of
the twentieth century. The golden-age argument is closely tied with the
mass-party model and the idea that parties ought to match that model of
party organisation – that is, large membership organisations that help
politicians stay in touch with the electorate via local branches as part of a
healthy democracy of politically active citizens. In other words, the mass-
membership organisation is not only what parties used to look like, but
also what they ought to look like.
However, the mass-membership ‘golden age’ may be more imagined
than real. Scarrow (2002, pp. 92–5) suggests that mass-membership
organisations have in fact never been the norm in democratic states. She
argues that the 1950s and 1960s constituted a fairly brief period of wide-
spread mass-membership mobilisation: ‘Before and after this period,
parties exhibited an uneven pattern of commitment to, and success in,
enlisting supporters in permanent organizations’ (Scarrow, 2002, p. 94).
This would suggest that parties based on mass membership should not
necessarily be seen as the default party organisational form. It would also
suggest that talking about ‘decline’ may not be entirely accurate. The talk
of decline suggests that parties used to be stable mass-membership organ-
isations, and what we are seeing is a change away from this norm.
Scarrow suggests that the high membership numbers in some parties in
the 1950s and 1960s were in fact somewhat unusual, rather than the
norm (Scarrow, 2002, p. 94). The changes we are seeing in party mem-
bership numbers should perhaps, therefore, not be seen as a change away
from a pre-existing ‘normal’ state of affairs. Rather, parties are ever
changing organisations that are in a process of continuous adaptation to
the dynamic political environment in which they operate. Hence, when
we analyse changes in political parties we should do so from the perspec-
tive that change itself is normal, and any pre-existing state of affairs that
parties are moving away from was itself inherently temporary. It could,
therefore, be argued that mass-membership parties were an organisa-
tional form that fitted with a fairly brief moment in the development of
modern democracies.
Why members?
Having considered the problems of defining exactly who make up the

party on the ground, and looked at what may be happening to it; there is
one key question that has not yet been addressed: why have a party on
the ground in the first place?
We have seen earlier that one of the key reasons for the rise of mem-
bership-based parties was the extension of the franchise from a small
propertied elite to the vast majority of the population. As the franchise
was extended it became necessary for political parties to build, often
large, membership organisations to mobilise enough voters to survive.
However, while this might explain why the leaders of existing parties and
the founders of new parties might want to add membership organisations

to ‘their’ parties or create parties with membership organisations, this
does not explain why citizens would want to become members of these
organisations. The following will explore two of the main attempts to
explain the existence of membership organisations: the concepts of
supply and demand.
Party membership incentives: supply
Considering the apparent decline in membership number it may be

worthwhile to consider why anyone would want to be a member of a
political party. Several different theories have been proposed to answer
this question. In Downs’ model of democracy he assumes that party
members act solely to attain power (Downs, 1957, p. 28). Consequently,
members are power-seeking individuals and presumably any party that
succeeds in obtaining power should see its membership rise. However, as
is evident from the declining membership of many parties, even those
with long histories of being in power such as the Scandinavian Social
Democrats, this does not seem to be an adequate explanation.
It is true that Downs makes his assumptions about the behaviour of
individuals primarily to create an elegant model and that ‘[t]heroretical
models should be tested primarily by the accuracy of their predictions
rather than the reality of their assumptions’ (Downs, 1957, p. 210).
However, as some government parties are losing members, the pursuit of
power does not seem to predict changes in membership numbers. There
must, therefore, be more to it than that. Nevertheless we should not
reject Downs entirely as there are good reasons to believe that some
degree of self-interest is involved and that for a party to survive it must
be able to satisfy the personal needs of the members (Eldersveld, 1964, p.
272).
As outlined above, at least some parties were created to represent the
interests of their members in parliament. For example, mass parties
started out as working-class parties devoted to the protection of working-
class interests. In that light it would seem sensible for people whose inter-
ests are being promoted to assist the party by joining it.
However, there is evidence to suggest that this commonsense conclu-
sion is wrong. The problem is that any general benefits that a party will
provide, say a minimum wage, employment laws and unemployment ben-
efits, will be enjoyed by members and non-members alike. These benefits
are collective and it would be difficult to somehow limit them to party
members – at least in a democratic society. Rational thinking people,
therefore, whilst having an interest in acquiring these common goods,
have an equal interest in avoiding paying for them through party mem-
bership (Olson, 1965, p. 21). Consequently, in order to explain party
membership we need some reward or incentive that is available only
through membership of a political party.
Clark and Wilson (1961) identify three groups of selective incentives –
that is, incentives that are available only to members. The first is material
incentives – tangible incentives of monetary value, or things that can
easily be translated into monetary value (Clark and Wilson, 1961, p.
134). In this category they include ‘the improvement in property values
for a neighbourhood redevelopment association or the increase in wages
and other tangible “fringe” benefits obtained by a labor [sic] union’
(Clark and Wilson, 1961, p. 134).
Their second group of incentives they call soldiery incentives. These
include the enjoyment derived from socialising with likeminded people,
and from being part of a group (Clark and Wilson, 1961, p. 134) as well
as any feelings of loyalty to the organisation itself (Clark and Wilson,
1961, p. 136 n8).
Finally there are purposive incentives. These are incentives that are tied
to the stated purposes of the organisation itself (Clark and Wilson, 1961,
pp. 135–6). That is, people are drawn to a party because they believe in
its goals.
Clarke and Wilson’s selective incentives as a solution to the problem of
free riding identified by Olsen has been generally accepted as a reason-
able explanation of party membership (Ware, 1996, pp. 67–8). However,
there are problems with their model, most notably in their view of mate-
rial incentives. Their examples of material rewards include, for example,
wage rises secured by a trade union and property value increases from
neighbourhood redevelopment. These examples are exactly the kind of
benefits that are subject to free-rider problems – a wage increase would
be enjoyed by all workers whether unionised or not. It, therefore, seems
necessary to develop Clark and Wilson’s model a little, which is exactly
what Seyd and Whiteley have done. Seyd and Whiteley explain party
membership through their general incentives model (Seyd and Whiteley,
2002, pp. 89–93). They argue that five issues influence the decision to
join a party:
1 Collective incentives, that is individuals want to promote certain

goals, such as defending a national health service or employment
rights, through the party.
2 Selective incentives which involve personal advancement through the
party, such as using the party to get elected to a legislative assembly.
3 Group incentives – that is the belief that the party as a group can and
is making a difference to the country.
4 Expressive incentives, by which they mean affection and loyalty

towards the party.
5 Social norms, the meaning of which seems a bit vague, but seem to
refer to the idea that people will be more likely to be active in a party
(and not leave) if they were asked to join by a ‘significant other’, espe-
cially a family member.
These, or a combination of these, incentives will explain why people join

political parties. Conversely, the deterioration of the same incentives may
explain why some parties lose members, something which will be
explored below.
Leadership incentives: demand
The other side of the coin is the extent to which party leaders want to
encourage people to join. This will depend on whether they see party
members as an asset or a liability.
There are a number of reasons why a large membership might be
regarded as an asset. First, having a large membership can be seen to add
legitimacy to the party. If a lot of people want to join, the party will
appear popular and well anchored in the community, with party
members acting as ‘ambassadors’ for the party spreading its message
(Seyd and Whiteley, 2004, p. 361). Further, if a party is able to claim that
it has a large membership that is actively involved in deliberation and
debate about the big issues facing society, the party will have a certain
level of democratic legitimacy (Scarrow, 1996, p. 42).
There will also be a direct electoral benefit from having party
members. While party members make up a fairly small, and shrinking,
part of the electorate they can usually be relied on to turn out to vote for
the party whatever the circumstances (Scarrow, 1996, p. 43). This could
potentially ‘provide a crucial margin of victory’ especially if turnout is
low or the race tight (Scarrow, 1996, p. 43). However, while party
members turning out to vote could help clinch victory for the party they
also perform an important role following defeats. In bad times having a
solid membership organisation can be essential for a party’s survival and
eventual recovery (Seyd and Whiteley, 2004, pp. 360–1).
In addition to being able to rely on party members to turn out on elec-
tion day, the leadership also gain important benefits from party members
as local campaigners. There is a tendency to see party members and
labour intensive campaigning at the local level as somewhat old fash-
ioned and less effective than modern high-tech capital intensive cam-
paigning. This is one of the central aspects in the rise of the electoral
professional party. However, there is extensive evidence to suggest that

local campaigning is in fact crucial for a party’s electoral success (this will
be explored further in Chapter 8). It would, therefore, be very unwise for
party leaders to discount the contribution that local campaigning by
party members makes to the electoral success of a party.
There is also more to the local work of parties than campaigning.
Having a strong local membership will help the party communicate with
voters in a personal way that cannot be replicated through mass media
communication or opinion polling. As Seyd and Whiteley write:
‘members can be political communicators, both upwards and down-
wards. Upwards, they are one among many means of informing the party
of voters’ opinions [...] Downwards, members are one of the means by
which parties can communicate their ideas and policies to a wider group
of people’ (Seyd and Whiteley, 2004, p. 362). Voters may get a great deal
of their political information from the media and party campaigns, but it
has also been argued that voters, through conversations with family and
friends, will validate and confirm what they hear (Popkin, 1991; Seyd
and Whiteley, 2004, p. 362). If the party has a large membership organi-
sation it will improve the chances that the party’s stance is validated
when party members interact with their social circles.
This indeed is a crucial element in Lawson’s concept of linkage. By
assisting the two-way process of communication between politicians and
voters, party members strengthen the link between rulers and the ruled.
Having a strong membership organisation also increases the pool of
talent the party is able to recruit from, both for internal office holders,
such as organisers, agents and managers, and candidates for public office.
Membership organisations have been called ‘greenhouses for new polit-
ical talent’ (Scarrow, 1996, p. 45). It is true that party leaders could look
outside the party for potential candidates, but there are advantages to
being able to recruit internally. Recruiting candidates for public office
from within the party is more likely to ensure a high level of loyalty to
the party, its policies and its leadership than people who are brought in
from outside. As Cowley points out, by the time many politicians reach
public office they have invested a considerable amount of time and
energy in the party:
By the time someone becomes an MP they will have been a member of

their party for years, usually decades. And not just a sitting-on-their-
arse-in-front-of-the-telly-having-paid-their-membership-fee-by-direct-
debit type of member either. Those who become MPs will have been
amongst the most active members of their party. By the time they
reach Westminster, they’ll have put in hours, days, weeks of unpaid
work on behalf of their party. They will have canvassed, donated
money, delivered leaflets, and attended meeting after godforsaken

meeting. (Cowley, 2005, p. 25)
Cowley argues that people like the ones described in this quote are much
less likely to rebel against the leadership than some fair weather friend
brought in from outside the party who has not proven their commitment
to the organisation. Indeed, as was mentioned above, many parties require
both candidates for internal positions and would be candidates for public
office to have been a paid up member of the party for a certain length of
time before they are allowed to put their name forward. In short, having a
substantial membership organisation allows parties to ‘reproduce’ them-
selves (Seyd and Whiteley, 2004, p. 361) at all levels including in parlia-
ment. A party with few members may find that difficult, especially in bad
times when outsiders with little commitment to the party will be less
willing to be recruited and more likely to withdraw their support.
A final reason why party leaders might want a membership organisa-
tion is the funding it provides. As mentioned above, the typical definition
of a party member is someone who has signed a membership form which
includes paying a regular fee to the party. Indeed, as Duverger points out
this was one of the main reasons why mass parties emerged in the first
place:
Instead of appealing to a few big private donors, industrialists, or

important merchants, for funds to meet campaign expenses – which
makes the candidates (and the person elected) dependent on them – the
party spreads the burden over the largest possible number of members,
each of whom contributes a modest sum. (Duverger, 1964, p. 63)
This lesson still holds today. Seyd and Whiteley (2004, p. 361) point out
that membership funding is ‘less tied to the whims, pressures and partic-
ular policy preferences of large financial donors’. Big private donations
also cause problems for parties in that questions will be raised about
what donors are getting in return for their money: ‘accusations that a
party is selling influence are much harder to make when its funds come
from a wide variety of voluntary actors, rather than from a small number
of rich supporters’ (Seyd and Whiteley, 2004, p. 361). An additional
problem with reliance on a few large donors is that the party and its can-
didates run the risk of losing touch with the daily concerns of the people
they are there to represent. Addressing the issue of the need for large
funds to run election campaigns Obama wrote:
The path of least resistance – of fund-raisers organized by special

interest groups, the corporate PACs [political action committee], and
the top lobbying shops – starts to look awfully tempting, and if the
opinions of these insiders don’t quite jibe with those you once held,
you learn to rationalise the change as a matter of realism, of compro-
mise, of learning the ropes. The problems of ordinary people, the
voices of the Rust Belt town of the dwindling heartland, become a
distant echo rather than a palpable reality, abstractions to be managed
rather than battles to be fought. (Obama, 2006, p. 115)
It is true that large donations do hold a very big place in the life of
modern politics and it seems the lessons of Duverger have been forgotten
or abandoned under the pressure of running capital hungry election cam-
paigns. However, for parties with mass memberships, membership fees
still play a large role. For example in 2011 out of a total budget of just
over £31 million the British Labour Party got over £5 million from mem-
bership fees. While other sources of funding were clearly very important,
membership fees still accounted for about 17 per cent of the Labour
Party’s resources – money it would have to find elsewhere in the absence
of a membership organisation.
The costs of a membership organisation
There are clearly many reasons why a membership organisation is a good

thing to have for a political party. These reasons also explain why there is
such worry about the decline of membership organisations (something
which will be examined further later. However, there are also a number
of costs associated with building and maintaining a membership organi-
sation.
It was argued above that having large membership organisations might
enable the party to engage in a dialogue with voters through the
members. However, involving members in the policy making of a party in
this way might equally limit the room for manoeuvre of party leaders. As
we saw in Chapter 3 one of the factors that caused the rise of the catch-
all party was the decline in class-based voting and the subsequent rise in
electoral volatility. The result was a need for increased leadership control
in order to be able to react quickly to changes in the ever more volatile
and unpredictable electoral market. Having a large membership organisa-
tion might limit the ability of leaders to react swiftly to such changes
since involving significant numbers of people in consultation on policy
takes time.
However, even if it was possible to overcome the practical problems of
involving the members in the policy-making process, there are arguments
for why it may be better to avoid this. One argument is that internal
party democracy is incompatible with parliamentary democracy, because

it undermines the independence of parliamentary representatives. There is
also a strong and long-standing view that party members, and especially
activists – the people most likely to be involved in intra-party activities
such as policy making – are more extreme than a party’s likely voters.
Even though this may not actually be true, the mere impression that they
are may be enough for an electorally driven leadership to worry about
the image that party members give the party. Hence, justified or not,
leaders may be motivated to avoid the troubles believed to be associated
with large membership organisations. Both of these arguments will be
explored further in Chapter 7 when the details of policy making will be
explored.
One solution to the problems of involving members may simply be to
ignore them and just use their services when needed at campaign time.
However, even this carries problems as not all members are likely to
accept being ignored by the party leadership. Surveys of party members
in Canada, Norway, Denmark and in the British Labour Party have
shown that influence on policy is important to a great many party
members (Pettitt, 2007, pp. 231–2). They are, therefore, not likely to take
kindly to being excluded from influence and may protest in very public
ways. This raises the problem that signs of internal disagreement are seen
as being damaging to a party’s electoral prospects. Internal debate tends
to be framed by the media as being damaging to a party’s electoral
prospects and testimony to its inability to govern (Stanyer, 2001, p. 85).
Hence membership rebellions are regarded as damaging to the party’s
electoral chances. In short, considering that both excluding party
members from influence and incorporating them may be damaging to a
party’s image, electoral prospects and indeed parliamentary democracy as
such, it could be argued that it is perhaps best to not recruit too many of
these troublesome people.
A final reason why party leaders may want to avoid having
too many members attached to the party is that membership organisa-
tions are expensive and time consuming things to run – time and
money party leaders may decide could be better invested elsewhere.
This is especially the case if party members are seen as a source of
trouble. In Scarrow’s (1996, p. 41) words membership organisations
carry ‘opportunity costs’ which the leadership may simply not be
willing to pay.
There are then several reasons why new party founders may avoid
trying to build a party organisation in the first place; or why leaders of
existing parties may be happy to let the membership organisation whither
and perhaps ultimately die.
Explaining and reversing decline
Despite the arguments against building or maintain a membership organi-

sation it is still generally seen as ‘the done thing’ for a party to have some
form of ‘party on the ground’. As will be explored at the end of the
chapter, there are strong normative arguments supporting the position
that healthy party-membership organisations are good for democracy.
The question then is how we explain and reverse the apparent decline in
membership numbers.
As we have seen the trends in membership development vary quite sig-
nificantly between different countries. However, we have also seen that
there are many countries which have seen quite dramatic declines in
membership numbers. The question then is if this trend can be reversed in
the countries where membership numbers have declined.
According to Seyd and Whiteley (2004, p. 357) the answer to this
question depends on why the decline has happened in the first place.
There are two main suggested explanations for membership decline.
One is that the decline is due to structural factors – major socioeco-
nomic changes in society. Traditionally, parties attracted voters based on
key issues such as religion and class. However, as such social and cultural
identities have become weaker, people’s attachment to parties based on
group identities has also weakened. If voting and party support is now
based on short-term cost-benefit analysis – that is, deciding at each elec-
tion which party will give me the best ‘deal’ – then making a semi-perma-
nent commitment to a party by joining it makes little sense.
In addition, Seyd and Whiteley (2004, pp. 357–8) suggest that a
related structural reason for membership decline may be that the political
market place has become much more competitive than before. Social and
technological changes have provided a plethora of new ways of being
politically active. Citizens no-longer define themselves politically prima-
rily according to their employment, religion or place of birth. Instead, a
multitude of factors come together to create someone’s political stand-
point, aided by the ease of access to information and social groups via the
Internet. This has led to a rise of ad-hoc forms of political participation
such as demonstrations, consumer boycotts, blogging and petitions and
of single-issue pressure groups. These have provided alternative and, one
might argue, less demanding ways of being engaged in politics than party
membership.
If these major changes in society do explain the decline in party mem-
bership numbers, then this decline, or indeed any potential future
increase, in membership numbers is basically beyond the scope of polit-
ical parties to change. The rise and fall in membership numbers follow
societal trends and parties simply have to cope as best they can.
The second possible explanation is that changes in membership

numbers are based on the incentives provided by the parties for member-
ship. In other words if parties provide more of the incentives for member-
ship outlined previously in this chapter they could attract more members
should they want to.
Based on extensive questionnaire survey evidence, Whiteley and Seyd
(2002) argue that incentives-based explanations for changes in party
membership numbers are more convincing than structural explanations,
although structural factors still matter. In other words ‘decline of member-
ship can be turned around with the right incentives’ (Seyd and Whiteley,
2004, p. 357). Indeed, there are several examples of how an increase in
membership incentives has reversed a decline in members. The Labour
Party saw an (albeit short lived) increase in members in the mid-1990s
because the leadership decided it was worth growing the membership
organisation. The leadership invested extensive resources in the recruit-
ment and retention of party members. However, in the years after 1997
when the party entered government, the leadership became increasingly
divorced from membership concerns, leading to a major decline in mem-
bership numbers. The Danish Socialist People’s Party, the Danish Social
Democrats, the British Labour Party and the French Socialist Party all saw
an increase in membership numbers in connection with leadership elec-
tions in the first decade of the 2000s – elections which gave members the
power to choose the leader, which provided a significant incentive to join.
This would then suggest that membership decline is not inevitable, but can
be reversed should the leaders want it. Whether they do or not depends on
how they weight the costs and benefits of members outlined above.
The effects of membership decline
The final question this chapter will deal with is whether there are any
wider reasons why parties should be encouraged to have members. As we
saw above, whether party leaders want members or not seems to depend
on a cost–benefit analysis based on what priorities the leaders have.
However, there are more fundamental reasons why parties ought to have
members.
Seyd and Whiteley (2004, p. 364) argue very strongly that a general
decline in membership organisations would have the effect of under-
mining democracy. Without members parties cease to be linked to civil
society. This is a theme also found in the cartel party model (see Chapter
3). Further, it is view backed up by Mair who argues that what we have
seen in recent years is a mutual withdrawal by both citizens and politi-
cians into separate spheres of activity:
Citizens retreat into private life or more specialized and often ad hoc
forms of representation, while party leaderships retreat into institu-
tions, drawing their terms of reference ever more readily from their
roles as governors or public-office holders. The traditional world of
party democracy – as a zone of engagement in which citizens interacted
with their political leaders – is being evacuated. (Mair, 2006, p. 33)
This is a key element of Lawson’s (1988) claim that parties are failing in
one of their chief roles – to provide ‘linkage’ between voters and politi-
cians. Indeed, this failure of parties to fulfil their representative function
makes Mair question the sustainability of our current conception of
democracy as involving some form of mass participation. Mair sees a
worrying trend in the arguments about democracy where the popular
(that is, voter-based) element is being stripped away, thereby ending up
with a democracy without a demos (Mair, 2006, p. 29). This does not
necessarily mean the end of democracy, but perhaps the rise of a much
less meaningful version of it where the role of voters is radically reduced.
However, as we have seen, while Seyd and Whiteley (2004, pp. 364–5)
acknowledge that recruitment has become more difficult it is far from
impossible provided the right incentives are offered. Indeed, it has been
suggested that there is an increasing desire among the public to participate
in decision making (Kittilson and Scarrow, 2003, p. 63) – hence part of the
reason why citizens are withdrawing from parties into alternative methods
of engagement may be because many, if not all, parties do not provide ade-
quate opportunities for engagement. As Russell (2005, pp. 47, 48) writes:
If political parties are to be seen as worth joining by those who are

passionate about political issues, they must provide a real site of nego-
tiation with political leaders [...] Thus a key part of ending the disap-
pointment [with politics] is a re-engagement not only between the
voters and the politicians, but between the politicians and the parties
that sustain them [...] The potential prize – a better engagement
between the parties and the voters – is great.
This theme will be picked up again in Chapter 7 where the issue of policy
making and in particular membership influence on this process will be
addressed.
Conclusion
It should by now be clear that party members are a central, but also
much debated part of party political life. Despite disagreement and uncer-
tainty regarding who exactly count as party members, supporters or

activists; how many of them there are; and whether they are a good thing
or not, there are few if any parties that have managed to survive and
thrive without some form of party organisation outside of parliament.
Party members provide legitimacy, crucial links with voters, and funds
and personnel for campaigning. Many party leaders may view party
members as a source of trouble as much as an asset, but the decline in
their numbers has led many to worry about the health of modern democ-
racies. It would then be fair to say that despite any doubts, party
members continue to be a crucial element in the life and thus academic
study of political parties.
Chapter 6
Candidate Selection
The last few weeks before an election usually receive inordinate levels of
attention from the media. However, in many respects this part of the elec-
toral process is to some extent the least important in terms of deciding
who gets to run the country. It is widely recognised that the ‘long cam-
paign’ – basically, the period of time from the morning after an election
until the next one – is what decides elections, not the last few weeks that
make up the ‘short campaign’. The problem is that even the long cam-
paign is only half the story when it comes to deciding which specific indi-
viduals get to run the country. The other half is the process of selecting
the candidates for election.
When compared to the excitement of an election campaign or the
passion of ideological debate, the importance of candidate selection is
often under appreciated. Candidate selection is indeed something often
carried out by small groups of dedicated party activists, behind closed
doors, out of view of the general public. It is often a private process
carried out more or less in secret. This is problematic because selecting
candidates for public office is arguably one of the most important ele-
ments of a party’s life. If the definition of a party is an organisation
fielding candidates for public office (see Chapter 1), then the selection of
those candidates clearly lies at the very heart of what a political party is.
Hence, the first part of this chapter will explore the wider importance
of candidate selection before moving on to examining key elements in the
selection process itself. The chapter will look at the impact that different
processes have on those finally selected; it will explore two key areas of
variation between parties in the selection process: territorial (de-)centrali-
sation and the in/exclusiveness of the selectorate; it will finish by looking
at two explanations for the under-representation of women in parlia-
ments: the supply of and demand for candidates.
The importance of candidate selection
Regular, free and fair elections are seen as one of the foundations of a
functioning democracy and are, rightly, given an enormous amount of
attention in political debate and study. However, when it comes to deter-
103
mining and understanding exactly who ends up in elected assemblies,

elections are only the last step in a long and complex process. What
happens before the election – the selection of candidates by parties –
is arguably even more important to the social, ethnic and gendered
composition of parliaments: ‘a party’s candidates will help define its
characteristics – demographically, geographically and ideologically –
more than its organization or even its manifesto’ (Hazan and Rahat,
2010, p. 6).
It has also long been recognised that control over candidate selection is
a key factor in the balance of power between different actors in a polit-
ical party. At the beginning of the twentieth century Ostrogorski wrote
that while the day-to-day business of the local ‘caucus’, or branch, would
be left to ‘the small fry’ the ‘bigwigs’ took control when it came to the
selection of candidates. He further argued that control of the selection of
the candidate showed who ‘the real masters of the Organisation’ were
(Ostrogorski, 1902, p. 345). This is still seen to be the case today (Hazan
and Rahat, 2010, p. 6). Those who control the selection of candidates
will have a key influence on the ideological make up of the party’s parlia-
mentary group, which in turn will help determine the policy direction of
the party. Because of that, it could be argued that the struggle over candi-
date selection will be far more intense than the struggle over the content
of election manifestos (Gallagher, 1988). Members, it is argued, find it
almost impossible to influence policy making directly and, therefore,
direct their energies towards selecting candidates sympathetic to their
particular ideological standpoint.
In addition, in many democratic systems, especially parliamentarian
democracies, becoming a candidate for a legislative assembly is the first
step on the career ladder that potentially leads to the highest political
office in a country. A Prime Minister will often have served in other min-
isterial offices previously; becoming a minister frequently requires also
being a member of parliament; to have any chance of becoming a
member of parliament, one must first have been selected as a candidate.
Further, understanding the process of selecting candidates is important
since across the globe, democratically elected assemblies are typically
highly un-representative of the general electorate. This is especially so
when it comes to gender. Only about 19 per cent of parliamentarians
globally are women despite women making up more than 50 per cent of
the population. By and large this under-representation of women is not
because voters reject female candidates, but because parties fail to select
them as candidates in the first place. To understand why that is the case
one has to look at how candidates are selected.
Therefore, candidate selection is a key concern when trying to under-
stand the composition of parliament in terms of ideology, gender and
Candidate Selection 105
social characteristics, as well as when examining intra-party power strug-

gles and cohesion.
Selecting candidates
The key point to understand about candidate selection is that it is a fil-

tering mechanism (Norris, 1997, p. 3). The candidate selection process
eliminates 99.96 per cent of those who are eligible to stand (Gallagher,
1988, p. 2). It can be seen as a multi-stage funnel where more and more
people are eliminated at each stage (Krook, 2010a, p. 709; see Figure
6.1). The rest of this chapter is about how we go from often millions of
eligible people via the vastly fewer who aspire to stand to a few thousand
actual selected as candidates.
When considering candidate selection we need to consider both formal
and informal factors that influence who gets through the funnel. Formal
factors include national legislation on who are eligible to stand, usually
involving limitations based on citizenship and age. They also include
party rules determining both who are eligible to stand and how the
process is run. It is also important to understand the impact of the polit-
ical system of a country – such as whether it is federal or unitary and
which electoral system is in use. On the informal side we have the impact
of career paths (some types of jobs are perceived as being more conducive
to learning the skill necessary for a successful political career); overt and
covert discrimination by those who select candidates; and the ambitions
and resources of would be candidates. These factors will have to be con-
sidered in the context of selection processes that are both enormously
varied and also very often hidden from public view. Norris (1997, pp. 1,
2) boils all of these issues down to four key areas of particular concern:
Eligible
Aspire
Selected
Elected
Figure 6.1 Funnel of candidate selection
Source: Norris and Lovenduski (1993).

Political system
Selection process
Supply Demand Candidate
Figure 6.2 The candidate selection context
Source: Norris (1997).
the political system; the formal recruitment process; the supply of candi-
dates (those who aspire for office) and the demands of the people with
the power to select candidates (see Figure 6.2). The rest of the chapter
will examine each of these areas in turn.
The political system
That a country’s political system will have an impact on candidate

recruitment should hardly be surprising. Countries such as the United
Kingdom, the United States, India, Canada and France hold elections for
their national legislative assemblies using single seat constituencies. Each
party will need to select only one candidate for each seat. By contrast
countries such as Spain, all of Scandinavia and Israel use List
Proportional Representation. The process used for selecting a single
person per constituency clearly has to be different from selecting and
rank ordering candidates on a list. It will make a difference whether the
lists in List Proportional Representation systems are open (voters can
vote for individuals within each party’s list) or closed (voters merely vote
for the whole list and the party decides who is at the top of the list and,
therefore, more likely to get elected). In short, each electoral system will
need a different selection process – and electoral systems are nearly as
varied as are political parties.
In addition the attractiveness of a seat in a particular assembly – be it
national, regional or local – will differ depending on factors such as: the
extent to which power is centralised in the national capital or devolved
downward to local and regional government; whether a particular legisla-
tive assembly is unicameral or bicameral; and the party system (Norris,
1997, pp. 11–12). The issue of (de-)centralisation is particularly impor-
tant. In federations such as Germany and the United States, regional and
local legislative assemblies have far more power than in more centralised
countries such as the United Kingdom. Hence getting a seat in a sub-
national assembly will be more attractive, and therefore more highly
sought after, in the United States and Germany than in the United
Kingdom.
It is also argued that the political system of a given country will affect
how centralised the selection process of a party is likely to be. First, the
electoral system may affect the degree of centralisation in the candidate
selection process. In essence, the argument runs that local election means
local selection. Hence, in systems with relatively small, single member
constituencies, candidate selection is more likely to be decentralised than
in a List Proportional Representation system with much larger multi-
member constituencies (Gallagher, 1988, p. 9; Lundell, 2004, p. 26).
Second, it has been argued out that a party’s organization tends to
mirror the political structure of the country it is operating in (Epstein,
1980). In other words, in a country where most political decisions are
taken at the centre, parties are also more likely to be highly centralised.
Duverger (1964, p. 398) did argue that the effect is the other way around
in that the ‘internal structure of parties [...] exercises a fundamental influ-
ence on the degree of separation or concentration of powers’ in a
country. However, most writers see the effect running from wider polit-
ical structure to party structure (Gallagher, 1988, p. 8). So, the argument
is that candidate selection in federal countries tends to be de-centralised,
where as in unitary countries it is more centralised (Lundell, 2004, p. 26).
However, the empirical evidence for the connection between the polit-
ical system in a country and the internal selection processes of a party is
mixed. It has been suggested that a more important issue is whether a
party cares that much about elections, or is more focused on ideological
purity. In radical parties focused on ideology rather than electoral
success, candidate selection is more likely to be centralised than in elec-
tion-oriented parties (Duverger 1964, p. 60). In highly ideological parties,
the leadership will want to ensure that all candidates meet a strict
measure of ideological purity. At the same time, if no-one in the party is
particularly interested in winning elections then there will be less resist-
ance to the leadership controlling candidate selections. On the other hand
if elections are seen as important – and winning seats a route to power
and influence – then there will be many in the party with an interest in
influencing the process. As a result, local elements of the party will want
to have a say.
Further, in his study of West European countries Lundell (2004) finds
no statistical connection between the degree of centralisation in candidate
selection processes and either the electoral system of a given country or
whether the country is federal or unitary. Interestingly, what he does find

is a regional effect: the candidate selection processes in Scandinavian
parties tend to be more decentralised than parties in Southern Europe.
This is consistent with Gallagher’s (1988, p. 10) suggestion that ‘political
culture [...] could help explain the form that selection takes.’ He further
argues that ‘political culture [...] is naturally influenced strongly by his-
torical experience’ (Gallagher, 1988, p. 10). This is the line that Lundell
(2004, pp. 39–40) takes when trying to explain the regional variation
that he finds in his data. He argues that the Scandinavian countries are
strongly influenced by their long histories of parliamentary democracy.
There is a long tradition of popular participation in politics and strong
local government. By contrast, Southern Europe has had a much more
tortuous route to democracy. Italy did not become democratic until after
1945, and Greece, Portugal and Spain had to wait until the 1970s. As a
consequence ‘politics in Southern Europe is characterized by strong and
charismatic leadership, centralization and authority’ (Lundell, 2004, p.
40), which is reflected in the more centralised candidate selection
processes of Southern European parties. It could, therefore, be argued
that when trying to explain variation between parties in how centralised
their candidate selection procedures are, looking at the political structure
of the country they operate in may not be very fruitful. Instead we may
be better off looking at the characteristics of each party (Lundell, 2004,
p. 41). It, therefore, seems that while ‘macro-political’ factors (the polit-
ical system of a country) may have an impact on the selection process –
the selection of individual candidates for single member seats will be dif-
ferent from constructing lists in List Proportional Representation – it
matters just as much how individual parties respond to these macro
factors. To understand how parties respond we may, therefore, need to
look inside each party. Making generalisations about how parties select
their candidates is, therefore, not as easy as simply looking at the political
system of a country.
National legislation
Much more indisputable is the impact that national legislation has on the
selection of candidates. It is true that candidate selection is generally
speaking relatively unencumbered by national legislation (Hazan and
Rahat, 2010, pp. 3–4, 7). Parties are by and large left to do their own
thing when it comes to selecting candidates for public office.
Nevertheless, it would be wrong to ignore national legislation completely.
Particularly in the United States, state level legislation has a huge impact.
For the vast majority of legislative bodies, that is, the Senate, House of
Representatives, governorships and state legislatures, state law deter-

mines how candidates are to be selected, not party rules (Hazan and
Rahat, 2010, p. 19; Ware, 1996, p. 259). That this can have a significant
impact on who wins the selection is exemplified by Plouffe’s (2009)
description of the 2007 Democratic Primaries. States allocate delegates
for the Convention that selects the presidential candidate using very dif-
ferent methods. Some are proportional to the candidate’s vote; others use
a ‘winner takes all’ approach. Plouffe argues that by understanding these
differences better than the Clinton campaign, the Obama campaign was
able to allocate their resources better, which helped them secure victory
(Plouffe, 2009, pp. 167–73).
However, even outside the United States, national legislation affecting
candidate selection still exists. Often legislation is less about how candi-
dates are to be selected and more about who are eligible to stand for elec-
tion (see Table 6.1 for legal requirements in selected democratic countries
around the world). As can seen from Table 6.1 age and citizenship are
pretty much universal requirements for would-be candidates for the
lower chamber in these examples. Indeed, if we were to look at all the
208 countries and territories in the world – regardless of their actual level
of democracy – 93.2 per cent have age requirements and 94.7 per cent
have citizen requirements (see the ACE Electoral Knowledge Network
website http://aceproject.org). In addition, in many countries would-be
candidates have to register with the authorities and a number of countries
require candidates to be resident in the electoral district they are targeting
(which presumably would prevent a candidate from standing for more
than one constituency at the same level of representation). None of this is
likely to be very controversial or exclude very many people (in the case of
the age requirement the criteria would be met automatically in due
course).
However, a number of countries have more stringent requirements.
Some demand that candidates must have been born in the country in
question, which would permanently exclude some people (although
perhaps not that many except in countries with large numbers of recent
immigrants). There are though some requirements which would be a
serious hindrance for potentially large parts of the population. This
includes the demands for a minimum level of literacy and education,
something particularly prevalent in African countries. In poor countries,
such as much of Sub-Sahara Africa, which lack universal education of a
certain length and quality this does have the potential to exclude signifi-
cant sections of the population from even seeking to become candidates.
It is largely the countries with low levels of literacy and education that
have such requirements. In short, in a number of countries, national legis-
lation is preventing a notable proportion of the population from the pos-
Table 6.1 Illustration of registration requirements for candidates for the lower chamber

Country Age Citizenship Country Residence Minimum level Minimum level Registration Other
of birth of education of literacy
Europe
Austria • • •
Denmark • •
Estonia • •
France • •
Germany • • •
Greece • • • •
Hungary • •
Italy • •
Poland • •
Romania • • •
Spain • •
Americas
Argentina • • • •
Brazil • • • • • • •
Canada • • •
Chile • • • • • •
Ecuador • • • • •
Mexico • • • • •
Paraguay • • • •
Suriname • • •
Uruguay • • •
USA • • •
Country Age Citizenship Country Residence Minimum level Minimum level Registration Other
of birth of education of literacy
Africa
Botswana • • • • •
Cameroon • • • •
Chad • • • • •
Ghana • •
Kenya • • • • •
Madagascar • • • •
Mozambique • • •
Nigeria • • • •
Tanzania • • • • •
South Africa • • •
Asia and Oceania

Australia • •
Candidate Selection
Bangladesh • •
India • • •
Indonesia • • • • •
Japan • •
Korea, Republic of • • •
Malaysia • • • •
Nepal • •
New Zealand • • •
Thailand • • • • •
111
sibility of even trying to obtain positions of political power. This shows

the importance of considering the issue of candidate selection. In most
democratic countries there is a lack of representatives from the poorer
sections of the population. However, when there are such stringent
requirements to even become a candidate, such under-representation will
only be exacerbated.
In addition to the legal requirements that must be met, the imposition
of quotas for certain types of candidates can have a major impact on who
is selected. The most common quota is for gender, specifically for female
candidates, which makes sense since women are the most notable
example of under-representation across the world. According to Murray
(2010) at least 100 countries around the world use some form of gender
quota, with France being the first to bring in a parity quota (50 : 50).
Quotas come in many forms, but have the common purpose of alleviating
the massive under-representation of women in parliaments. The extent to
which parties actually stick to these quotas varies (Murray, 2010), but
they still have a significant impact on candidate selection procedures.
The candidate selection process: impact of different

methods
In most countries the actual process of candidate selection in a given

party is rarely if ever known to the general public. Indeed, a ground-
breaking study of candidate selection was entitled ‘the secret garden of
politics’ (Gallagher and Marsh, 1988). The obvious exception is the
United States where the process tends to be specified in state legislation.
However, the rules, regulations and informal byways of this area of party
life are vitally important for understanding who gets the chance of repre-
senting their party in parliament. It has been argued that rules on who
are involved in selecting a party’s candidates (commonly referred to as
the ‘selectorate’) will have major consequences for the balance of power
in a party. It will also have an impact on how easy it is for the leadership
to control MPs (Norris, 1997, p. 7).
To illustrate the impact that different selection methods can have on
MPs’ loyalty to the leadership, Hazan and Rahat (2010, pp. 1–3) explore
the behaviour of a particular (unnamed) candidate, based on real-life
experiences in Israel. This candidate is selected by a membership-wide
ballot, rather than a smaller committee as had been the case before. Once
selected by the mass membership and later elected to parliament the new
MP saw his loyalty as lying with the grassroots and not with the leader-
ship. He was, therefore, difficult to bring into line with the wishes of the
leadership – until, that is, the process was re-centralised under leadership
control and the MP went from rebel to loyalist. As Hazan and Rahat
point out this is not a one off example.
The British Conservative Party has been experiencing similar problems
after the 2010 election. In the aftermath of a major political scandal that
broke out in the United Kingdom in 2009 involving dozens of MPs from
all the major parties, the Conservative Party under David Cameron
attempted to show they wanted a new politics. One way of doing that
was by inviting people not associated with traditional political career
paths to apply to become candidates, and have them selected through
new methods, including open primaries. This certainly created a new
kind of MP – one not loyal to the party leadership, but to their con-
stituency and/or personal causes.
The problem of independently minded legislators is taken to the
extreme in the United States, where the party leadership have far less
control than in many other countries. Even the rules are not determined
by the parties themselves. This can be seen as part of the reason for the
notoriously undisciplined politicians of American parties. In short, candi-
date selection processes may often be hidden away from the general
public, but they are of vital importance to how political parties, and
therefore politics in general, function.
Territorial (de-)centralisation and in/exclusiveness of the

selectorate
What should be clear by now is that the variations in the way that candi-
dates are selected have important consequences for politics. It would,
therefore, be worth looking at some of the key ways in which selection
processes vary. The following will focus on two crucial areas where varia-
tions can occur: territorial (de-)centralisation and in/exclusiveness of the
selectorate.
Territorial (de-)centralisation is concerned with where the selection is
done. For example, is it the party’s national head office that is respon-
sible, or a regional head quarters, or local branches? In/exclusiveness is
concerned with how many people are involved. Is it a small group of
people in the party’s top leadership, or the wider membership? Arguably
there is a link between the two issues. There is tendency to see decentrali-
sation as also involving increased inclusivity – that is, if candidate selec-
tion is done locally it will involve more people than if it is done centrally:
‘Generally the more decentralized candidate selection is, the greater the
possibility for individual party members to play a role’ (Hazan and
Rahat, 2010, p. 55). However, it has also been pointed out that ‘this is an
inclination, not a rule’ (Hazan and Rahat, 2010, p. 55). If all candidates
were selected at a single big national conference made up of several dele-

gates from each constituency, that would be a very centralised way of
doing it. If selection was done by the chairperson of each local con-
stituency party, that would certainly be very decentralised, but also less
inclusive than the central national conference. So, there is a relationship
between the two factors, but it is not simply a matter of decentralisation
being equal to greater inclusiveness.
Figure 6.3 summarises the relationship between the two factors. Some
of the positions in Figure 6.3 face quite severe practical problems and
should perhaps be seen as illustrations of what is theoretically or logically
possible rather than an indication of what actually goes on in political
parties. For example, in all but the smallest of democratic countries the
top right and top left positions would be very difficult indeed from a
purely practical perspective. Even in unitary states it would be difficult
for a single leader to select all candidates. As an example, the French
National Assembly has 577 seats. For the leader of a major party to have
to select that number of candidates would be a huge task. Likewise, con-
ducting a ballot of the tens of millions of voters for 577 seats would be
both hugely expensive – and no voter is likely to want to make that many
choices. Even in smaller countries (for example Denmark 179 seats; New
Zealand 122 seats; Singapore 94 seats) the task would be considerable.
Hazan and Rahat (2010, p. 47) do give examples of where a single party
leader selects all candidates, although this does seem to be fairly rare.
Indeed, these practical considerations seem to have had the conse-
quence that candidate selection, as a rule, tends towards being fairly
decentralised. Even through the sheer diversity of candidate selection
processes makes generalisations difficult, Ranney, (1981, pp. 82–3; see
also Hazan and Rahat, 2010, p. 58) argues that in most parties selection
tends to take placed at the constituency level, although often with
national level supervision. In terms of whether there have been any
general changes Bille (2001) concludes in his study of West European
party rule books that the overall trend between 1960 and 1990 is
towards greater decentralisation.
However, the issue of national level supervision suggests that Figure
6.3 does not give the full picture. One of the things that serve to muddy
the waters is that often selection takes place in several stages, each stage
with different levels of centralisation and inclusiveness. As we can see
from Table 6.2 the national leadership often has some level of veto of
constituency decisions, or puts together a national ‘pool’ of aspirants
from which the constituencies have to choose. It would be fair to say that
being able to control access to a pre-approved pool of potential candi-
dates will give the leadership more control than having to reject local
choices. Limiting the number of available choices tends to cause fewer
Organisational exclusiveness
Most inclusive Most exclusive
Voters Party members Party delegates Party elite Leader
Most centralised
All the party’s All the party’s All the party’s All the party’s All the party’s
National
candidates candidates candidates candidates candidates
selected by all selected by all selected by a selected by party selected by the
voters. members. national delegate executive. party leader.
conference
Territorial centralisation
The party’s The party’s The party’s The party’s The party’s
Regional candidates in candidates in candidates in candidates in candidates in
each region each region each region each region each region
selected by all selected by all selected by a selected by the selected by the
voters in that members in that regional regional party regional party
region. region. delegate executive. leader.
conference.
The party’s The party’s The party’s The party’s The party’s
Most decentralised
candidates in candidates in candidates in candidates in candidates in

each each each each each
Candidate Selection
Local
constituency constituency constituency constituency constituency

selected by all selected by all selected by a selected by the selected by the
voters in that members in that constituency constituency constituency
constituency. constituency. delegate party executive. party leader.
conference.
Figure 6.3 Territorial (de-)centralisation and in/exclusiveness of the selectorate
Source: Hazan and Rahat (2010).
115
Table 6.2 Illustrations of candidate selection procedures

Party National Brief overview of process
electoral
system
Parti Socialist Two-round A call is issued for members to put themselves forward as candidates. In a small number of cases a
(France) single prominent personality may be asked to stand in a particular constituency. Members then vote on all
member aspirants within a constituency. National committee must validate choice and can reject a candidate.
seats Due to constitutional parity law some seats are designated as all-women shortlists.
Union pour un Two-round Party members select candidates within their constituency. Chosen candidates are scrutinised and validated
Mouvement single by a national selection committee. The national selection committee then reports to the Political Bureau for
Populaire member final approval, although the latter body is seen as too large and unwieldy, and meeting too infrequently to
(France) seats have significant input.
African National Closed list Local branches make nominations and select delegates for Provincial List Conference. Candidate
Congress (South PR, two tiers, nominations submitted to provincial level. Provincial List Committee, appointed by Provincial Executive
Africa) national and Committee, screen nominations and draft consolidated list. To be on consolidated list candidates most have
by province been nominated by at least five branches. Provincial List Conference votes on consolidated list and select
delegates to National List Conference. National List Committee drafts national list based on candidate
popularity. National List Committee vote on draft national and provincial lists. Provincial lists opened up for
appeal, decided on by an Appeals Committee. Nation List Committee finalizes lists.
The Single National party creates approved list (includes training of applicants for national list). National list regularly
Conservative Member replenished and contains more candidates than there are seats. Local branch members select candidate from
Party (UK) Plurality national lists. Some recent experiments to increase number of ethnic minority and women candidate by
requiring some constituencies to select candidate from ‘A-list’ containing higher proportion of desired
candidates. Also some experiments with open primaries.
Party National Brief overview of process
electoral
system
The Labour Single Would be candidates fill in application form for National Parliamentary Panel (NPP) of preapproved
Party (UK) Member aspirants (encouraged but optional). Regional Assessment teams interview candidates for NPP. Constituency
Plurality Parties advertise vacancies and potential candidate contact the constituency parties to declare an interest.
Local branches and affiliated organisations (e.g., trade union branches) within constituency nominate
individuals for constituency short list. Membership wide vote selects candidate from short list. Selected
candidate submitted to National Executive Committee (NEC) for endorsement. Candidate not on the NPP
must attend NEC endorsement interview. Some constituencies designated as all-women shortlists.
The New Single National Executive Committee (NEC) instructs constituency executive committee to invite applications from
Patriotic Party Member aspirants. National level officials vet applicants and NEC approves list of candidates for next stage.
(Ghana) Plurality Constituency delegate conference convened consisting of constituency executive and chair of each polling
district within constituency (typically about 100 delegates) and selects candidate from list by secret ballot.
Source: Murray (2010), Conservative Party (2012), Labour Party (2013), African National Congress (2012); New Patriotic Party (2009).
Candidate Selection
117
problems than rejecting decisions by constituencies once they are made.

As Norris (2006, pp. 93–4) points out, when analysing ‘who nominates’
one should also, and ideally first, consider ‘what choices are available’.
Excluding certain people from becoming candidates by ensuring they are
not available for selection, what Bacharatz and Baratz (1963) call a ‘non-
decision’, is much more efficient and less problematic than vetoing selec-
tions once made. Indeed, party leaders have sometimes used the
promotion of gender equality to make unwanted (for whatever reason)
male candidates unavailable for selection (see for example Cutts, Childs
and Fieldhouse (2005) for an example from the United Kingdom – in this
case it went horribly wrong as the unwanted aspirant stood as an inde-
pendent and won the seat). Hazan and Rahat (2010, p. 37) point out that
based on the experience of a number of countries when the national lead-
ership has a veto it is rarely exercised, which further supports the idea
that filtering out unwanted aspirants before selection bestows more effec-
tive control that a post-selection veto.
Another issue worth mentioning is the degree of democratic control
exercised by the membership, and the extent to which there have been
any changes in this area – that is, how inclusive the process is. As is so
often the case, the evidence is contradictory (Bille, 2001; Hopkin, 2001;
LeDuc, 2001). However, overall it appears that two conclusions can be
drawn: the first is that party members have acquired greater power to
choose candidates; the second is that pre-selection has increased, or put
differently the choice available to members has been restricted (Norris,
2006, p. 106). This again confirms that when analysing who controls
candidate selection one has to consider not only who selects, but also
who has the power to decide the range of choices. Clearly, being aware of
the impact that multi-stage selection processes have is important to get
the full picture. A highly decentralised and inclusive selection may be pre-
ceded by a centralised and exclusive long-listing stage, potentially under-
mining the democratic element of the second stage. Hence, candidate
selection processes may often be done away from the public eye, but
appreciating their intricate details is central to understanding who wins
the chance to become a member of a legislative assembly with all the con-
sequences for democratic representation that follows from that.
Supply and demand
The last issue this chapter will deal with is ‘supply’ and ‘demand’. In this
context supply is about who are willing to put themselves forward as
potential candidates. Demand is about is what the selectorate is looking
for in would-be candidates. The study of supply and demand springs
from trying to understand one of the most notable features of modern

elected ‘representative’ bodies: the massive under-representation of
women. Despite making up over 50 per cent of the population in the
democratic world, women still account for less than 20 per cent of parlia-
mentarians, with a high of just over 40 per cent in the Nordic countries
and a low of 12.5 per cent among Arab states (see the Inter-
Parliamentary Union website – IPU.org). If we go back to Norris’ ‘funnel’
(see above) this means that women make up more than half of those eli-
gible to stand, but somewhere along the line get filtered out and end up
amounting to less than 20 per cent of those elected. Most studies in this
field have found no evidence that voters generally discriminate against
female candidates (Krook, 2010b, p. 157; Murray, Krook and Opello,
2012; Norris, Vallence and Lovenduski, 1992; Studlar and McAllister,
1991). Hence, if the voters tend not to discriminate between male and
female candidates in their voting choices, the explanation for the lack of
women legislators must be found in those parts of the filtering process
that take place within political parties before the election.
Supply
One of the reasons given by party leaders for why there are so few female
candidates, and therefore MPs, is that women simply do not put them-
selves forward for political office – an idea reflected in several publica-
tions on women’s under-representation (for example ‘If only more
candidates came forward’ (Norris and Lovenduski, 1993) and ‘Men kvin-
derne vil jo ikke selv’ (‘But, the women themselves don’t want to’)
(Dahlerup, 2001)). In short, parties would like to select more women
candidates, but there is simply a lack of aspiring women putting them-
selves forward.
There are two key factors that influence supply: ambition and
resources. When it comes to ambition several studies have found that
education is a key predictor of political activism. Education increases
‘political knowledge, interest, confidence and skills’ (Norris and
Lovenduski, 1993, p. 388). This may not affect women unduly in much
of the Western world where women’s educational attainment often
matches and sometimes outdoes that of men. However, it is a factor in
many parts of the world, especially developing countries, where boys’
education is given priority over that of girls’.
In addition, research has found that gender itself has a major impact
on ambition, quite apart from any differences in educational attainment
(Lawless and Fox, 2005; 2010). According to Krook (2010b, p. 158) this
can be explained by ‘long-standing patterns of traditional socialisation
which associate men with the public realm and women with private’. Put
slightly bluntly, men are raised to be ambitious and encouraged to build
careers for themselves, whereas women are encouraged to think of them-
selves primarily as wives and mothers. This may not be overt and delib-
erate but is displayed in simple things such as buying girls a cookery set
and boys a home chemistry set.
This difference is also found in survey studies. Lawless and Fox (2005)
found that women were twice as likely as men to believe themselves not
to be qualified for public office. In addition they also routinely underplay
their chances of winning compared to men. Hence, this argument sug-
gests that ambition and confidence is about gendered socialisation which
drives women to underestimate their skills and chances and men to over-
estimate theirs, regardless of their actual skills and chances. As one senior
female business women said about the private sector ‘I believe that we
will only have true equality when we have as many incompetent women
in positions of power as incompetent men’ (http://www.bbc.co.uk/news/
business-12518277) a sentiment which could equally well be applied to
politics.
In addition to ambition, resources also play a role. Simply being
unable to pay to be a candidate can prevent people from standing. Even
where campaigns are not as capital intensive as for example in the United
States it can be expensive to even attempt to be selected as a candidate
(Norris and Lovenduski, 1993). With a high proportion of women being
concentrated in low paid and part time jobs compared to men, the lack of
financial resources is likely to affect women more than men. In many
countries once an aspirant has been selected as a candidate, the party will
step in and support their election campaign. However, there is rarely any
outside financial support available for becoming a candidate. This early
cost has to be borne by the aspirants themselves. Indeed, in the United
States the issue of financial barriers to women candidates has led to the
creation of several groups aimed at supporting and encouraging women
to put themselves forward as would be candidates (Krook, 2010b, p.
158). Two examples of this are the National Women’s Political Caucus
(www.nwpc.org) and The Wish List (www.thewishlist.org). In addition to
raising funding these organisations also provide training and encourage-
ment for women to enter politics.
Another resource issue is time. Becoming a candidate is a time con-
suming business and can take up a lot of free time as well as require
working time flexibility. Such flexibility is not available in many jobs.
The issue of job types is important beyond finding enough time to
become a candidate. Jacobs (1962) argues that what he calls ‘brokerage
occupations’, such barristers, teachers, trade-union officials, journalists,
and political researchers, often provide not only the necessary time and
flexibility, including the option of career breaks, to run for political

office, but also lead to politically oriented networks and useful skills and
expertise. In the case of the United Kingdom Norris and Loveduski
(1993, p. 386) write that the brokerage explanation:
helps illuminate not just the class disparity, but also why women and
ethnic minorities are under-represented in Parliament since they are
often concentrated in low-paying skilled and semi-skilled occupations,
or in family businesses, with inflexible schedules and long hours, in
sectors which do not provide traditional routes to political life.
Hence, parties may have a point when they say that they would like to
select more female (and other under-represented) candidates, but the
supply is just not great enough.
Demand
When it comes to demand there is both a formal and an informal side.

Many parties will stipulate some basic and more or less stringent qualities
that any aspirant must have if they are to have any hope of becoming a
candidate. These frequently involve a minimum length of party member-
ship, donations, party activism, commitment to the party’s goals, and
experience and expertise useful for being a candidate. According to
Hazan and Rahat (2010, p. 21) catch-all parties that are primarily con-
cerned with electoral success are likely to have fairly minimal formal
requirements. This is in order to have greater flexibility to select candi-
dates which are seen as being able to attract voters, even if they are not
long standing party ‘soldiers’. By contrast more ideologically focused
parties are likely to have far more stringing requirements in order to
ensure the ideological purity of the parliamentary group. Gallagher
(1988, p. 247) argues that variations in how stringent the requirements
are also affected by left/right positioning, with left-wing parties being
more demanding than right-wing parties. This fits with the discussion in
Chapter 2, which suggested that the catch-all party model first appeared
on the right of the political spectrum. By definition, a catch-all party is
more concerned with electoral success than ideological purity.
However, while formal requirements will have some impact on who
are able to get selected it is possible to argue that informal demands can
be as, or even more important. Sometimes the lack of supply of female
candidates is the function of subtly gendered demands – the criteria for
what counts as a ‘suitable’ candidate is more likely to be found in a man
than in a woman. This has been referred to as the ‘unconscious valorisa-
tion of masculine or male-dominated qualities’ (Murray, 2010, p. 58).

One example of this was the very stringent requirements of the Belgian
Socialist Party which includes ‘have his wife and children enrolled in the
appropriate women’s and youth organisation’ (Obler, 1974, p. 80). This
requirement does not state that a candidate has to be a man, but clearly
the reference to having his wife enrolled in the Women’s Section suggests
that men are regarded as the standard. Hence the unspoken assumption is
that anyone wanting to stand as a candidate for the party is a married
man.
The criteria for what counts as a good candidate may not be quiet so
blatantly gendered as in the above example, but rather the overall profile
of a candidate is one more likely to be found in man than a women
(Murray, 2010, p. 46). Hence discrimination by the selectors against
women may not be overt. Indeed, in their study of selectors in the British
Labour Party, Bochel and Denver (1983) did not find gender to be some-
thing selectors mention in either positive or negative terms. Further, selec-
tors may honestly believe that they are selecting on the basis of merit.
However, since few selectors actually know all the aspirants personally,
certain background characteristics, such as gender, race, education and
occupation, step in as proxies for merit. In short, electors are looking for
‘the right sort’ of candidate (Norris and Lovenduski, 1993, pp. 378–9).
In this context the ‘out-group effect’ (Niven, 1998) makes an impact.
Selectors will look for characteristics that are associated with successful
or at least ‘typical’ candidates in the past, one key characteristic being
that they are overwhelmingly male. The typical member of the political
elite is male and few women make it to the top. So, when looking for a
candidate that is likely to make it to the top, based on the status quo, the
choice will be a man (Krook, 2010a, p. 709).
However, while these subtle, hidden and often unconscious elements of
the discrimination against women will take their toll, there are also many
case studies that find overt gender discrimination and sexism. In her
study of women and politics in Chile, Franceschet (2005) found frequent
examples of patronising attitudes towards women at party meetings.
Even in the UK Liberal Democrats, which prides itself on its equality
agenda, Evans (2008, pp. 598–600) found many examples of female aspi-
rants having experienced overt sexism from selectors.
So, while there may be issues of supply when it comes to the under-
representation of women, there also seems to be a problem of demand,
even if it is sometimes covert and even unconscious, rather than overt and
deliberate. Indeed, problems with demand can have an effect on supply.
Norris (2006, p. 93) argues that evident lack of demand for certain types
of candidates, such as women, may lead to some aspirants being discour-
aged from even putting themselves forward. Further, when female aspi-
rants, or aspirants from other under-represented groups, do come

forward they do not fit the requirements for being the ‘right sort’. Hence,
those aspirants that do come forward are seen as lacking in merit because
they belong to ‘out-groups’ that do not match what a ‘typical’ candidate
‘ought’ to look like.
Conclusion
Candidate selection may not appear to have the same importance as ide-
ology or campaigning. However, as the above has illustrated it is a key
battle ground in intra-party power struggles, and has a critical role to
play in who gets the opportunity to run the country. Hence, candidate
selection processes have a major impact on how parties operate and who
make it to parliament. If elected assemblies do not look like the popula-
tions they are elected to represent, the most likely explanations are to
be found within parties. In short, if we are to understand who gets to
represent us in modern democracies we need to understand candidate
selection processes.
Chapter 7
Policy Making
Ideology is often seen as sitting at the heart of what a party is, or at least
should be – that is, a body of people united by the pursuit of a common
vision of the ideal society. As we saw in Chapter 1, this was Edmund
Burke’s definition of a political party. Chapter 4 also illustrated that ide-
ology is not necessarily the prime motivating force behind a political
party. However, regardless of what drives the people behind a political
party – an ideology, ethnic or religious loyalties or the desire to control
the resources of the state for the benefit of themselves and their followers
– this driving force will have to be turned into practical action. In a dem-
ocratic society such practical action will usually take the form of policy
proposals. Policy making is therefore an important part of what a party
does – it is the process through which the driving force of a party is
turned into practical real life action.
This chapter will explore four key issues related to policy making. The
first is related to the question of who should be involved in the making of
party policy. The chapter will consider both the normative and pragmatic
arguments for and against inclusive and even democratic policy making
processes. The chapter will then look at some of the practical obstacles to
open and inclusive policy making processes. Having looked at the back-
ground to this debate, the chapter will consider the extent to which there
is a demand to be involved from the party on the ground – that is, from
party members and supporters. Finally, the chapter will look at who are
actually involved in the making of party policy.
Who should be involved?
The issue of whether party members should be closely involved in the

making of party policy has been the subject of intense debate – in essence,
a debate over the desirability of intra-party democracy. This debate can
be split into two broad areas: a normative debate; and a pragmatic one.
The normative debate deals mainly with whether or not internally demo-
cratic parties are a help or a hindrance for democracy at the national
level. The pragmatic argument deals with whether, irrespective of the nor-
mative pros- and cons, a party leader would want to include party
124
Policy Making 125
members in the policy-making process. In short, whether or not one

thinks intra-party democracy is ‘A Good/Bad Thing’, what are the costs
and benefits for the party leadership of including/excluding party
members from the policy making process? The following will deal with
both sides of each of these debates.
The normative argument against intra-party democracy
The normative argument against intra-party democracy has a long

history, and, it could be argued, is somewhat old fashioned. There are
nevertheless several distinguished scholars who have argued against
intra-party democracy and their arguments should be noted. The main
thrust of the argument against intra-party democracy is that representa-
tive democracy as practiced in the modern world is about citizens
choosing their leaders, rather than themselves taking a very active role
in public affairs. Democracy is limited to citizens taking decisions on
only very broad national issues through electing representatives, rather
than making decisions on the details of the issues facing the country. In
democracy ‘the deciding of issues by the electorate [is] secondary to the
election of men [sic] who are to do the deciding’ (Schumpeter, 1976,
p. 269). Hence, democracy is a system ‘in which the people have a
choice among the alternatives created by competing political organiza-
tions and leaders’ (Schattschneider, 1960, p. 141). According to this
view democracy consists purely of the ability of voters to choose
between competing teams of elites. Indeed according to McKenzie
‘intra-party democracy [...] is incompatible with democratic govern-
ment’ (McKenzie, 1982, p. 195).
The reason for McKenzie making this stark assertion is that party
leaders, when making decisions on policy, have to act as leaders of the
entire political community and not just as leaders of a single party. The
wishes of the party can only be one of many sources that party leaders
take into account when making decisions (McKenzie, 1982, p. 195). If
the extra-parliamentary party did have the ability to control the parlia-
mentary party
then these party organs would supplant the legislature and the execu-
tive as the ultimate decision-making bodies of the polity. The organs of
party would transcend in importance the organs of government, as is
usually the case in totalitarian regimes. (McKenzie, 1982, p. 196)
This is echoed by Schattschneider (1960, p. 141) who argued that the

role of the extra-parliamentary party is to sustain competing teams of
parliamentary leaders that the electorate will be given the periodic oppor-
tunity to choose between by casting their vote in elections.
Common to all of these views is that the role of the people is not to do
the deciding themselves, but rather to chose those who are to do the
deciding. Parliamentary leaders need not be responsive to public opinion
in between elections, as long as they remain responsible to the electorate
by giving the voters the opportunity to pass judgement of their record in
regular elections (McKenzie, 1982, p. 200). It is though argued that this
system will most likely have the result that parliamentary leaders will
indeed be responsive to the voters in order to attract their vote
(McKenzie, 1982, p. 200).
A further central element of this view of democracy is that those
chosen to lead must be given time and independence to enact their pro-
gramme for the country:
democratic government based, in principle, at least, on delegation

from below can, no doubt, be made to work. But in order to do so, the
government, however chosen, must enjoy a real measure of independ-
ence and for a reasonably long period. (Amery, 1964, p. 19)
In the context of the United Kingdom, Amery further argued that: ‘[the
British] system is one of democracy, but of democracy by consent and not
by delegation, of government of the people, for the people, with, but not
by, the people’ (Amery, 1964, pp. 20–1). This is what has been defined as
‘Tory democracy’ (Beer, 1965, p. 91). The key concern of this approach
to democracy is:
to put those who are best able to identify and pursue the national
interest into office with sufficient independence to do so. The system is
democratic because the authority of the government rests on the
consent of the people, but it is not government by the people. (Katz,
1997, p. 30).
Once the best people to lead the country have been identified, they
should be left in peace to get on with doing their job until the next elec-
tion. Clearly, intra-party democracy would disrupt the work of the great
and the good we have selected to lead us.
The normative argument for intra-party democracy
There are a number of counter arguments to these objections to intra-

party democracy. For example, if the most important aspect of a democ-
Policy Making 127
racy is that parliamentary leaders open themselves up to the judgement of

the people through regular elections, then the power of the extra-parlia-
mentary organisation over the parliamentary party ought not to be a
problem. In essence, from a democratic perspective it should not matter
whether it is the parliamentary party or the extra-parliamentary party
that makes the decisions, as long as those decisions are open to be judged
by the voters in elections.
In addition, there is no particular reason to believe that intra-party
democracy will produce policies that do not take into account inputs
from a wide range of sources, as McKenzie correctly asserts is necessary.
In other words, why would decisions taken by a small group of party
leaders be more likely to take account of wider societal interests than
decisions made by a much larger number of party members through party
organs?
Further, it has been argued that elections are not good enough at com-
municating the wishes of the ‘people’ to the governors of a polity
(Gutmann and Thompson, 1996, p. 130). This is the essence of Ware’s
(1979) defence of intra-party democracy. Ware argues that ‘by itself, the
electoral mechanism does not provide the autonomy of citizens’ choice
that is necessary for the existence of a liberal society’ (Ware, 1979, p. 32).
The ‘electoral competition’ model of democracy, as represented by
Schumpeter and McKenzie, draws on an erroneous analogy with compe-
tition in free markets (Ware, 1979, p. 36). In other words, the argument
against intra-party democracy sees parties as companies working in a
market, engaged in a competition for votes. The party that fails to satisfy
voter demands will not gain enough votes to stay in business – acquiring
government power, or just staying in parliament. If at election time voters
are unhappy with the work done by a particular party they will simply
shift their votes to another party.
However, the analogy is erroneous because the party political market
is not perfectly competitive the way the electoral competition model of
democracy implicitly assumes (Ware, 1979, pp. 38, 40). Rather than
being perfectly competitive, party political markets are in fact often
highly oligopolistic, that is, controlled by a few players (Ware, 1979,
p. 38). This has serious consequences for the view of democracy as com-
petition between elites. In an oligopoly, parties have little interest in
redefining the political reality in which they operate (Ware, 1979, p. 51).
In other words, parties may compete on providing policies that will
attract voters, but only within a fairly limited range of issues which party
leaders deem to be relevant to contemporary society and which these
leaders feel comfortable competing on. Demands for policies that go
beyond the ‘norm’ are rejected by party leaders. Even if these new issues
are what voters want to hear about, they are still often excluded since
only the leaders of big parties can set the parameters for political compe-
tition and it is in their interest to stick to tried and tested issues.
Hence, even if the actual wants and needs of voters change, they
cannot articulate those wants and needs through the ballot box since
parties define which issues they compete on. In other words, if none of
the main parties presents policies reflecting new wants and needs then
voters are not able to express a desire for such policies since their only
choice is voting for existing policies. Because of the oligopolistic nature of
the electoral market, new parties with policies outside the accepted
parameter of normal politics find it difficult to successfully challenge the
existing parties. Indeed, if we are to believe Katz and Mair’s (1995) cartel
party thesis as explored in Chapter 3, this is an increasing problem as
existing parties actively cooperate to exclude challengers. This means that
if voters want a new set of policies to challenge ‘normal’ politics, they
will struggle to make that clear if their only option is the ballot box.
When parties lack internal democracy, the agenda is controlled by the
leadership, who more often than not prefer to compete on long-estab-
lished issues they are comfortable with, but which ‘by their very nature
[...] hark back to a vision of political reality that is probably obsolete’
(Ware, 1979, p. 51).
Ware’s solution is intra-party democracy: ‘the argument being pre-
sented here is that parties that provided for real control over policy goals
by their activists would come to have a leadership that took account of
new conflicts’ (Ware, 1979, p. 78). By having intra-party democracy,
parties will provide voters with a direct influence on policy to supplement
the often ineffective option of merely voting for another party. This will
deal with the problem that ‘the pressure to search for a new conception
of political reality [...] is most easily rejected when the party organisation
is least open to mass movements’ (Ware, 1979, p. 51). In parties with
intra-party democracy members will be able to bring onto the agenda
issues that the leadership might not otherwise have wanted to deal with.
Echoes of this argument are found elsewhere as well. John Stuart Mill
wrote that ‘a democratic constitution, not supported by democratic insti-
tutions in detail, but confined to the central government, not only is not
political freedom, but often creates a spirit precisely the reverse’ (Mill,
1994, p. 334). Sartori (2005, p. 28) argues that while parties have many
functions, their most important is that of ‘expression’. Expression
involves ‘the party as the agency which typically communicates the
demands of the society to the state, as the basic link or connector
between a society and its government’ and the ‘expressive function is a
continuous flow which outlasts electoral intervals’ (Sartori, 2005, p. 24).
Hence parties ought to work as conduits between citizens and govern-
ments not just at election time, but continuously. This is clearly linked to
Policy Making 129
Lawson’s idea of linkage. Arguably, in order for them to perform that

function best, membership influence on policy is necessary. This is sec-
onded by Miliband who writes:
Members of Parliament cannot nowadays realistically claim Burkian

independence nor is it in the least evident that democracy would be
better served if they could. To suggest that they should be free from
commitment on the great issues of the day until they have heard those
issues debated in Parliament is to take a purely mythical view of the
working of the political process. (Miliband, 1958, p. 173)
This argument also finds support from the ‘deliberative’ view of democ-
racy (Teorell, 1999). It has been argued that the view of democracy as a
competitive marketplace where voters ‘buy’ a party product, erroneously
assumes that voters have a set of well-established preferences. Instead, it
is better to think of voters as having an ‘empirical’ and a ‘hypothetical’
standpoint. The empirical standpoint is what they would come up with if
they were stopped on the street and asked to give an instant response to
an issue. The hypothetical standpoint is what people would come up with
if they had had time to consider all the facts and to deliberate with
others. The argument is that the hypothetical standpoint is a more con-
structive one than the empirical one, and that intra-party democracy is a
good way of getting to the ‘hypothetical popular will’.
The desirability of membership influence is further strengthened by the
concern over the breakdown of the representative function of parties. As
was outlined in Chapter 5, there is a strong argument that parties are
being separated from their voters because membership organisations are
shrinking. Bringing party members in to policymaking could help rebuild
those membership organisations because this increase in influence would
provide greater incentive for membership. In addition, strengthening the
role of members would help rebuild party linkage. As was argued by
Miliband ‘whatever contributes to the growth of [the non-professional
politically interested individual’s] significance also contributes to the
vitality of democratic politics’ (Miliband, 1958, p. 172).
In many ways, the pro intra-party democracy camp can be seen as
having won the argument – at least when it comes to the official position
taken by many parties and widely held views of with what counts as
‘good practice’ in terms of how to run a party organisation. Mill’s view
of the need for ‘democratic institutions in detail’ has been taken up by
projects dealing with advising emerging democracies on the best way to
build good democratic structures. So, The Netherlands Institute for
Multiparty Democracy (NIMD, 2004) writes that there are a number of
‘institutional guarantees’ parties need to meet in order to be able to play
their proper role in democratic societies – one of which is internal

democracy. In other words if parties are to be constructive players in
democratic societies they must practice internally what they preach
externally (Scarrow, 2005, p. 3). This was echoed by the conclusion
drawn by a group of Catholic bishops in Eastern Africa: ‘Our conviction
is that it is within the political party that democracy starts; it is also here
that it starts to fail’ (Svåsand, 2008, p. 6). Hence, while the academic lit-
erature is fairly evenly split on the issue, although most recent works are
pro, the ‘official good practice’ view seems to be fairly firmly in the pro-
camp.
Pragmatic arguments against intra-party democracy
However, even if there are good normative arguments why members

should be involved in policy making, there may also be equally good
pragmatic reasons for excluding them. One of the main pragmatic argu-
ments against involving party members in the making of policy is that
they are not the kind of people sensible vote-oriented party leaders would
want to decide the party’s platform. Party members, and especially
activists – those most likely to want to be involved in the making of
policy – have long been regarded as ideological extremists. Sidney Webb
said of members of the British Labour Party that they were ‘frequently
unrepresentative groups of nonentities dominated by fanatics, cranks and
extremists’ (see Seyd, 1987, p. 202 n1). The situation was hardly better in
the Conservative Party where local branches were described by the
Conservative politician Hugh Cecil as ‘knots of vehement, uncompro-
mising and unbalanced men’ (Cecil, 1912, p. 238; also May, 1973, p.
140). Indeed, the idea that activists hold extreme views can be traced
back to the very first study of party organisations:
It is an almost general fact that the Association is more Radical than

the mass of the party [...] It follows in the first place that the Caucus is
incapable of supplying a correct estimate of public opinion, of giving a
more or less accurate idea of its tendencies and aspirations. It has
proven this incapacity more than once. (Ostrogorski, 1902, p. 596)
This view rests on the idea that these ‘assiduous attenders of branch and
constituency meetings’ (Seyd, 1999, p. 388) are active not because of the
trappings of public office, but because of their commitment to a partic-
ular view of society often based on strong ideological foundations. They
are, therefore, more extreme than both the broad electorate and the elec-
torally conscious party elite.
Policy Making 131
High
Organisational status
Low
Left Centre Right
Policy position
Figure 7.1 Curvilinear disparity
Source: May (1973, p. 139).
This argument has found its most popular and elegant expression in
May’s so called ‘law of curvilinear disparity’ (May, 1973; see Figure 7.1).
The idea is that parties are stratified, with an upper, middle and a
lower layer (May, 1973, pp. 135–6). The main components of the upper
layer are a party’s members of publically elected assemblies and members
of executive bodies (ministers). The middle layer consists of a party’s
membership base: active and inactive members and those who express
strong loyalty for a party without necessarily being members. This also
includes US style registered supporters. The lower layer is made up of
lukewarm and occasional supporters – ‘ordinary’ voters with only a low
level of party political attachment (or indeed political interest). It is
argued that while each layer will be very diverse in its opinions, there will
still be an overall trend (May, 1973, p. 136). That trend will vary
markedly between the three layers. The most important variation is that
the middle layer will be relatively more extreme in their political stance
than the other two layers. The bottom layer will tend to be moderates,
and the top layer will be somewhere between the two (as indicated by the
vertical dashed lines in Figure 7.1), although closer to the lower layer
than the middle layer. Hence, there is a curvilinear disparity between the
ideological trends of the three groups.
This disparity is caused to a large extent by the different patterns of
recruitment to the three layers. Activists without leadership ambitions are
not required to follow anything other than their own ideological fervour,
which is likely to be strong. Indeed, the tangible rewards for party mem-
bership and activism are few and far between (May, 1973, p. 149).
Hence, the only reason why someone would want to be active in a party
is the fire of their own convictions. If as an activist you want to get
elected to internal positions in the party you will have to appeal to fellow
activists and their strong ideological views. In short, people in the middle
layer tend to be driven by strong ideological views, or required to appeal
to such views.
People in the top layer on the other hand do get substantial tangible
rewards for their position in the form of a salary and all the trappings of
power that come with reaching high political office. At the same time,
people in the top layer of a party are highly dependent on the bottom
layer (or voters) to remain in their position. Hence ‘legislators whose
opinions deviate from those of many of their constituents stand to gain
from discarding such opinions’ (May, 1973, p. 149). Even if politicians
have radical opinions that deviate from the ideological centre of gravity
of their voters, they are more likely to succeed if they moderate those
opinions.
All told this means that when it comes to determining a party’s policy
position, the top layer of the party is better off ignoring the middle layer
as much as possible.
Pragmatic argument for intra-party democracy
However, while this view of party member extremism has a long tradi-
tion, there is good evidence to suggest that things may not be as simple as
May’s model makes it look.
The first issue is that May’s law has been subjected to extensive
testing, and the results have been weakly supportive of the law at best,
outright against it at worst. One key criticism is that each layer in a party
is far too heterogeneous in terms of motivations and ideological positions
to make general statements about them Kitschelt (1989, p. 403). Others
have found general trends, but as opposed to May concluded that in fact
the top layer is more radical than the middle layer (Norris, 1995). Some
have found limited support for May’s law, but only under very specific
circumstances (Kennedy, Lyons and Fitzgerald, 2006, p. 802). In other
words, the law was not necessarily generally applicable.
Hence, while May’s law seems plausible and has the strength of sim-
plicity on its side, there is little evidence to show that it is actually an
accurate reflection of the opinion structure of political parties. Indeed, it
would seem that there is little in the way of structure in the opinions of
Policy Making 133
each layer due to the heterogeneous nature of political parties. Different

people become active in political parties and strive for high office for a
plethora of different reasons, and to say that one part of a party is gener-
ally driven by one or another concern appears not to be backed up by the
evidence.
Finally, in the context of declining membership numbers (see Chapter
5) it has been found that despite the fact that party members seem to be
an ever-shrinking minority there is no evidence that they are becoming
minorities in terms of other characteristics (Scarrow and Gezgor, 2010).
On a range of measures – income, union membership, religiosity, and
perhaps most importantly, ideology – the smaller membership organisa-
tions are becoming more, not less like society around them. It has, there-
fore, been suggested that modern membership organizations ‘still have
the potential to help link their parties to a wider electoral base’ (Scarrow
and Gezgor, 2010, p. 823).
Indeed, there is good evidence to suggest that letting party members
participate in policy making is not a danger in electoral terms, and may
also be a positive benefit. Local activist-based doorstep campaigning is
becoming increasingly important for the electoral success of parties,
something which will be explored further in Chapter 8. However, in
order for party supporters and members to be willing to invest time and
effort in campaigning they need to be motivated to do so. It has been
argued that one of the most effective and long-term sustainable means of
doing so is to involve party member in the making of the ‘product’ (poli-
cies) they are asked to go out and ‘sell’ on the doorstep (Pettitt, 2012a). If
party members feel a sense of ownership of the policies they are cam-
paigning for, they are more likely to invest the time and effort in cam-
paigning and to do so with conviction.
Who can be involved in policy making?
However, even if a party is committed to an internal policy making

process that is democratic, there may be practical obstacles to involving
very many people in this process. In order to survive, political parties
have to act like fighting machines and as such must ‘conform to the laws
of tactics’ (Michels, 1915, p. 41). By ‘laws of tactics’ is meant that wide-
spread consultation takes time, and in the day-to-day struggle of politics,
swift decisions have to be made: ‘democracy is utterly incompatible with
strategic promptness, and the forces of democracy do not lend themselves
to the rapid opening of a campaign’ (Michels, 1915, p. 42). So, for a
party to be able to react with sufficient speed to events and moves by
other parties ‘a certain degree of cæsarism’ is required (Michels, 1915,
p. 42) and hence in political parties, ‘democracy is not for home con-
sumption’ (Michels, 1915, p. 42). In other words, to be effective fighting
machines, parties have to be organised along oligarchic rather than dem-
ocratic lines. In making this argument Michels is supported by Duverger:
‘Democratic principles demand that leadership at all levels be elective,
that it be frequently renewed, collective in character, weak in authority.
Organized in this fashion, a party is not well armed for the struggles of
politics’ (Duverger, 1964, p. 134).
Michels and Duverger are not the only ones to make this assertion.
Explaining some of the reasons for the oligarchic nature of parties Bille
writes that:
a problem is presented in the morning and a solution is expected to

have been found by the evening or at the latest by the next day. The
demand for efficiency and authoritative action is great. That does not
leave much space for continuous consultation with very many people.
Centralisation is strengthened. The top must lead. (Bille, 1997, p. 229,
translated from Danish)
There are however reasons to believe that the ‘laws of tactics’ would not
be such a problem when it comes to the making of some aspect of a party’s
policy platform. While it may be true that politicians are often required to
take policy decisions very quickly – sometimes, as Bille argues, in a matter
of hours – this is not necessarily the case on all issues. When talking about
policy it is important to keep in mind that the policy concerns of a party
exist on several levels in terms of details and timeframes. Party policies
obviously have to cover a vast range of issues over the full spectrum of
government, from international to national to local. Party policy also has
to be dealt with in temporal terms, from grand long-term ideological foun-
dations of a party, to day-to-day issues. It is this latter consideration that
will be the topic of the following discussion. Temporally it is possible to
divide party policy into four levels (see Table 7.1):
1 The general and fundamental principles of a party; that is a party’s

basic ideology. This level tends to be very broad and unspecific in
nature and also tends to remain unchanged for long stretches of time.
2 The party’s plans and goals for each parliament/electoral period, that
is from election to election. These policies are what tend to be found
in election manifestos. By the nature of this category, the frequency of
change depends on how often elections are held.
3 A party’s plans for each parliamentary year.
4 The day-to-day work of a party negotiating with other parties in com-
mittees, whipping errant MPs and the other work that goes into
Policy Making 135
Table 7.1 Levels of policy and potential for consultation
Policy Level Consists of...: Potential for wide

consultation
1: Long term Basic principles, statements on ideology High

2: Medium term Electoral period, election manifestos High
3: Short term Annual, parliamentary year High
3: Day-to-day Daily negotiations internally and externally,
crisis management Very low
steering legislation through parliament. It also involves responding to

short-term issues and crises.
The question, then, is if the laws of tactics are equally relevant at all
levels. The problems of involving the wider party membership in the day-
to-day struggle of politics are obvious, as exemplified by the quote from
Bille above. However, when it comes to deciding on the party’s basic ide-
ological foundations, the contents of the election manifesto and even the
plans for a parliamentary year it is a lot less obvious why a need for
speed should automatically exclude widespread consultation. Especially
in countries with fixed-term parliaments there really should be little
problem in organising a process of consultation over broad themes for
the election and even more specific drafts of a manifestos.
It is true that there may be other costs than loss of speed associated
with a democratic policy making process. It has been argued that
‘a united party is more likely to do well at the ballot box’ (Katz, 1986,
p. 63) and while the connection between party unity and electoral success
is not entirely straightforward, party unity is certainly perceived to be of
importance by party leaders (Dominguez, 2005). This is supported by
Laffin, Shaw and Taylor (2005, p. 3) who write that ‘the assumption
among the party leadership is that a party must present an image of unity
to succeed electorally’. This is a problem when it comes to policy making.
Policy making is often seen as a very ‘messy’ process and one which is
better not done in public. As Spirer and Spirer (1991, p. 347) write:
‘Bismarck, the 19th century German chancellor, is supposed to have said
that there are two things you don’t want to watch being made, sausages
and policy’. Hence one of the reasons intra-party democracy is problem-
atic for party leaders is that it is associated with (seemingly electorally
damaging) disunity (Hoffman, 1961, p. 233).
However, as was argued above, there are also benefits in terms of the
incentives for activism generated through the sharing of ownership of the
policy-making process. Hence, party leaders will have to weigh up the

costs–benefits of apparent unity at the cost of lowered activism, or
increased activism at the cost of apparent unity. There not necessarily a
single right or wrong answer to this dilemma.
Who wants to be involved in policy making?
Even if intra-party democracy is regarded as something to be encouraged

and there are no insurmountable practical obstacles, there is still the
issue of whether or not anyone lower down the hierarchy actually wants
to be involved in the making of party policy. There is a long-standing
opinion that most party members are not that interested in policy
making and are generally quite happy to let the leadership get on with it.
Michels argued that the politically active part of the population, or those
who have ‘a lively interest in public affairs’, is very small (Michels,
1915, p. 49). The same pattern is observable in a political party: ‘In the
life of modern democratic parties we may observe signs of similar indif-
ference. It is only a minority which participates in party decisions, and
sometimes that minority is ludicrously small’ (Michels, 1915, p. 50).
Further, not only are party members generally quite inactive, they are
positively grateful to those who are willing to take on the reins of
power: ‘Though it grumbles occasionally, the majority is really delighted
to find persons who will take the trouble to look after its affairs’
(Michels, 1915, p. 53; see also pp. 60–8). On this issue Duverger is in
agreement with Michels:
leaders tend naturally to retain power and to increase it, because their
members scarcely hinder this tendency and on the contrary even
strengthen it by hero-worshipping the leaders: on all these points the
analysis of Roberto Michels continues to hold true. (Duverger, 1964,
p. 134)
However, there is evidence to suggest that members are not as acquies-

cent as Michels and Duverger claim. Surveys of party members in
Canada, Denmark and Norway show strong evidence that policy making
is of great importance to most. The notion that the leadership should
listen more to the voters than to the members was opposed by a majority
of members in 11 of the 16 parties in the Scandinavian surveys (Hansen
and Pedersen, 2003, p. 110; Heidar and Saglie, 2002, p. 195). In a survey
of Canadian Party members, very similar sentiments were found. When
asked about their opinions of the influence of various groups within the
party, the members identified ‘the perceived over-influence of pollsters’
Policy Making 137
and ‘under-influence of ordinary party members’ as the ‘greatest source

of discontent’ (Young and Cross, 2002, p. 682). The survey also docu-
mented strong support for the idea that members should have a greater
say in the making of the national policy platform. Similarly, in Norway a
majority in five of the seven parties in the survey felt that the party on the
ground should have the power to make the parliamentary group take a
specific standpoint on issues not dealt with in the election manifesto or in
congress resolutions (Heidar and Saglie, 2002, p. 236). Finally, in
Denmark there was little support for the suggestion that politics could be
too complicated for the members to understand (Pedersen and Hansen,
2003, p. 121).
Further, The German Greens have long been marked by membership
rebellions against the leadership’s move to the right and eventually into
government (O’Neill, 2000; Poguntke, 2001; Wolf, 2003). Also in
Germany both the Social Democrats and the Communist successor party
Die Linke have had problems with membership rebellions (Hough, Koß
and Olsen, 2007). Further, the former Israeli Prime Minister Ariel Sharon
suffered several defeats at the hands of Likud party members on the issue
of the government’s policies towards the occupied territories. Finally, the
Republican National Convention in 2012 saw repeated protests by dele-
gates objecting to what they saw as the leadership’s attempts at under-
mining their influence.
So, Michels’ and Duverger’s claims about the willingness of party
members to let their leaders take a lead in staking out the direction of the
policy of the party may be true in some cases, but certainly not all. There
are, therefore, good reasons to believe that not all party members are
happy merely to follow the leadership’s line.
Who are involved in policy making?
When it comes to who actually makes party policy one view has a very
wide following: some parties, especially the mass parties of the left used
to be internally democratic, but with the rise of the catch-all, electoral
professional and cartel party models, policy making is now firmly con-
centrated in the hands of the leadership. The changes wrought by the
catch-all party engendered conflict between the party in public office and
the party on the ground. In the mass party, the membership organisation
existed before the party in public office and therefore felt a sense of own-
ership of the party in public office. With the catch-all party came an
‘increased assertiveness of the party in public office’ which brought it into
conflict with the party on the ground and the traditional view of member-
ship supremacy (Katz and Mair, 2002, p. 121). A similar conflict
occurred in the old cadre parties. Here the membership organisation was
created to maintain the party in public office. However,
once recruited, party members start to make demands, abetted by the

principle first articulated as part of the ideology of the mass party that
the party in public office should be responsible to the party’s members.
The result is that, although the party in public office may be the domi-
nant face of the party, its dominance is constantly under challenge.
(Katz and Mair, 2002, p. 121)
However, it is argued that this period of conflict has come to an end and
we have witnessed the ‘victory’ of the party in public office over the party
on the ground:
In contemporary party organisations, however, these conflicts seem to

have been settled, in that what we now appear to witness is the ascen-
dancy of the party in public office, which assumes a more or less undis-
puted position of privilege within the party organization. In other words,
we suggest that the development of party organizations in Europe has
gone beyond the catch all period and has entered a new phase, in which
parties become increasingly dominated by, as well as most clearly epito-
mized by, the party in public office. (Katz and Mair, 2002, p. 122)
There is, then, a widely accepted view that party members have gone
from having a significant level of influence on the organisation, to being
virtually completely side-lined. As Carty (2004, p.13) writes, most
modern theories on party organisations ‘appear to agree that the impera-
tives of modern electoral competition have worked to consolidate control
of [decision-making on policy and programmatic issues] in the hands of
the party in public office, and often the party leadership more narrowly
defined’. So, it is widely believed that party members have seen a decline
in influence.
However, there is an alternative view – that members have never had,
nor can they ever have, influence on policy. This is because organisations
are inherently oligarchic:
It is indisputable that the oligarchical and bureaucratic tendencies of

party organisation is a matter of technical and practical necessity. It is
the inevitable product of the very principle of organisation. (Michels,
1915, p. 35)
Hence, rather than having seen a decline in intra-party democracy and an

exclusion of members from policy making, parties, and indeed all organi-
Policy Making 139
sations, have always been and will always remain oligarchies. Still, in
both views the result is the same: policy making is concentrated in the
hands of the leadership.
The extent to which this view of policy making is accurate is difficult
to ascertain with certainty. Policy making in political parties, and espe-
cially the issue of membership influence is not a very well researched
area. Research to date has served only to expose how little is known
about the role of party democracy in the making of policy (Scarrow,
Webb and Farrell, 2002, p. 144). Studies of membership influence have
not gone much beyond candidate and leadership selection. There is no
separate investigation of policy making in Bille’s study of Danish parties’
organisations, only a reference to the ‘regrettable lacuna’ in our knowl-
edge of the field (Bille, 1997, p. 45). Scarrow et al. (2002) do deal with
the role of the members in policy making but are forced to rely heavily on
feedback from subject specialists rather than ‘hard data’ (Scarrow et al.,
2002, p. 144). The issue is not central to Katz and Mair’s data set based
on party rule books (Katz and Mair, 1992; Scarrow et al., 2002: 144),
nor to the 1993 collection of essays based on those data (Katz and Mair,
1993). The one place where such rule book data have been used to
explore membership influence is as part of a study by Widfeldt of parties
as instruments of linkage (Widfeldt, 1997, pp. 41–51). Even here it only
merits a subsection of one chapter. This lack of attention to policy
making may be explained by the fact that it is generally assumed to be
absent.
There are however some exceptions and also some evidence that policy
making is becoming more widely studied. McKenzie’s (1963) work on the
Labour Party and the Conservatives in the United Kingdom is an excep-
tion to the traditional silence on policy making. He finds that regardless
of the very big differences in official rhetoric on the matter, Labour
claiming loudly and proudly to be more internally democratic than the
Conservatives (Attlee, 1937) both parties were in fact very similar in
being highly oligarchic.
In a study of the Swedish Social Democrats, a party that fits neatly
with the ‘mass party evolving into catch-all party’ idea, it was found that
membership influence on policy had not declined (Loxbo, 2013). Rather,
compared to the 1950s and 1960s membership influence had in fact
increased. This would then very much be a deviation from the decline
idea. Further, in a study of the rule books of Danish parties it was found
that, as opposed to what might have been expected from the decline
argument, Danish party organisations had in fact seen very little change
(Pedersen, 2010). There was certainly no sign that they were moving
towards an oligarchic convergence. Further it was found that Danish
parties afforded their members with very different levels of influence –
from firm oligarchies to much more open and even democratic organisa-
tions. This would then go against both the decline argument and Michels’
view of permanent and unavoidable oligarchy.
One the other hand, in a study of Malawian parties, it was found that
all were more or less firm oligarchies (Svåsand, 2008). There was some
variation, but only in the severity of the oligarchic practices. On the other
hand, considering the limits put on the policy options of many African
governments (see Chapter 4) having internally democratic policy making
processes, even in a governing or potentially governing party, may to a
certain extent be of limited use.
Whilt the idea of ‘party member’ is somewhat more complicated in the
United States than elsewhere, and the idea of intra-party democracy,
therefore, somewhat difficult to apply, researchers there have still found
variations in how open the processes are. Reinhardt and Victor’s (2008)
study of the making of the Democratic platform for the 2004 election
found a process which involved hearings in several cities and written and
well as oral submissions from both individuals and groups. By contrast,
Weinberg (1977) found the Republican process in 1976 to be very much
more secretive and limited to a relatively small elite.
This suggests that rather than general decline or longstanding oli-
gopoly, membership influence in facts varies from party to party
depending on a number of factors (Pettitt 2012b). It has been argued that
left-wing parties tend to have more rebellious members than parties of
the right. There is, therefore, more likely to be a demand for intra-party
democracy on the left. Further, the challenges of getting into and staying
in government may make it less likely for intra-party democracy to
flourish. Government-seeking parties have to be more internally disci-
plined than those without government ambition. This is what is suggested
by Michels’ idea of the ‘laws of tactics’. Finally, it has been suggested that
intra-party democracy has a greater chance of survival in consensus-style
democracies than majoritarian-style democracies. Consensus-style democ-
racies have a greater tradition of negotiation and compromise, and even
when a party is not in government it can often still have influence on leg-
islation. By contrast majoritarian democracies are dominated by con-
frontation, and any party which is not in government has very little, if
any, influence. This means that parties operating in a consensus democ-
racy can more easily afford to pay the costs of a democratic decision-
making process (it being slower and will potentially make the party look
divided). This is a lot less so in majoritarian democracies where the
margin between victory and defeat is tiny and both victory and defeat
tends to be total. Anything that impairs efficiency is less easily accepted
in majoritarian than consensus democracies. In short, the ability of
members to influence policy may not be in terminal decline or perpetually
Policy Making 141
absent. Rather, it will vary depending on the context that each party finds
itself in.
Conclusion
The making of policy in political parties is clearly an important issue.

Indeed, it seems to have a certain times been seen as an issue of such
importance that it is better kept out of the hands of the masses. There
does seem to have been a change in that view towards a greater aware-
ness of the desirability of intra-party democracy in the area of policy
making. However, as we have seen there are also obstacles to democratic
forms of policy making and many, indeed most, parties tend towards the
less democratic end of things in this area. However, these obstacles are
not insurmountable and some parties do appear to have managed to
build reasonably democratic organisations.
Chapter 8
Campaigning
As we have seen throughout this book, political parties are multifaceted

organisations. They contain a wide range of internal actors – leaders,
activists, inactive members, casual supporters, indirect members, and
affiliated organisations such as trade unions. These actors will have an
equally wide range of goals. Hence, parties can be geared towards any-
thing from the realisation of abstract goals to the cynical pursuit of
power and personal aggrandisement. However, the winning of seats in
parliament must be seen as one of the central activities of any serious
party organisation. Indeed, going by the definition of a party used in this
book (as discussed in Chapter 1), the act of fielding candidates for pub-
licly elected office is what distinguishes parties from other political organ-
isations. This definition of a political party has led Farrell and Webb
(2002, p. 102) to conclude that: ‘Given this symbiotic relationship
between parties and elections, it is important to have a clear idea of how
parties operate in elections and elections affect parties.’
This chapter will look at a number of issues connected to the relation-
ship between election campaigning and parties. It will first examine the
ways in which campaigning is said to have changed and some
critiques/modifications to that view. It will then look at how this change
has impacted on party organisations. Finally, the chapter will look at the
issue of party financing, a key element in modern capital-intensive elec-
tion campaigns.
How campaigning has changed
Election campaigning is seen by many as having gone through three eras

so far (summarised in Table 8.1). Different people take slightly different
views on the details of each era and use different names for each. So,
Farrell (1996) refers to the ‘premodern’, ‘television revolution’, and
‘telecommunications revolution’ eras; Norris (2002) calls them the ‘pre-
modern’, ‘modern’ and ‘postmodern’. Somewhat simpler Blumler and
Kavanagh (1999) talk of three ‘ages’ and Farrell and Webb (2002) talk of
three ‘stages’. In this chapter we will use three ‘eras’, partly because this
is simpler than Norris and Farrell’s versions, but also because it will
142
Table 8.1 Eras of campaigning
Eras Characteristics
‘Style’ of campaigning Personnel Voter feedback Voter segmentation Communication channels
First ‘Propaganda’ approach; De-centralised local Local canvassing Heavy emphasis of Party press, posters,
rallying the ‘masses’ campaigning by volunteer returns. specific class (class- billboards, pamphlets, word
behind a pre-existing activists. Impressions and mass party). of mouth via volunteers.
product. ‘gut instincts’. Little if any focus on press
management.
Second ‘Selling’ approach; research Highly centralised Extensive Catch-all approach; Heavy focus on national
to find out which voters campaign organised by nationwide attempting to attract media, especially television.
might be interested in pre- internal media and polling. voters from across Decreased emphasis on local
existing product and explore polling experts; heavy society irrespective channels. Rise of media
which elements of the focus on a leadership of social management; building
product might resonate driven national media background. relationship with journalists.
with the potential voters campaign.
at a given time.
Third ‘Marketing’ approach; less Highly centralised Increasingly Continued use of Targeted direct mail to
about attracting voters to a campaign making increased advanced polling catch-all approach; carefully selected sectors of
pre-existing product and use of external campaign and use of focus also increased use the electorate; return of
Campaigning 143
more about discovering professionals contracted in groups. of detailed voter volunteer driven word of
voter needs and designing a on an ad-hoc basis. Return segmentation mouth; increasingly intense
product accordingly. of the local campaign according to array media management
relying on volunteer of social and ‘spinning’ the message.
activists to sell the product demographic factors.
door to door, but local
campaign co-ordinated
from the centre.
Sources: Farrell (1996); Norris (1995); Blulmer and Kavanagh (1999); and Farrell and Webb (2002). Table amended from Pettitt (2012a).
inevitably one day be necessary to add a fourth era, and rather than
having to find a name for whatever is beyond the ‘postmodern’ stage,
being able to call it the ‘fourth era’ is considerably more straightforward.
Despite the differences in terminology, there is a great deal of overlap
between the different views of campaigning eras and the following will
seek to summarise the overall view of each era based on Farrell (1996),
Norris (2002), Blumler and Kavanagh (1999) and Farrell and Webb
(2002). One feature of many of these explorations of ‘eras of cam-
paigning’ is that they are not always very well defined temporally – or
put differently, it is not always very clear when each stage is supposed to
have started and finished. Nevertheless, this approach does help give
some structure to changes in campaigning. The following will look at the
general features of each era.
First era
The first era of campaigning was dominated by short-term campaigning.

A party’s election campaigning efforts would be focused on the so-called
‘short campaign’. By ‘short campaign’ is meant the period of time from
when the election is called to election day. There would be little prior
preparation for the campaign. The campaign would be overwhelmingly
local, based around the activities of the local party machine and local
party activists, with very little national coordination. Campaigning was
conducted door to door by local activists. When the national intruded on
the local it would be through ‘whistle stop’ tours of party grandees
coming to speak to mass rallies around the country.
Significantly, the first era saw no use of ‘external’ experts such as pro-
fessional pollsters, media managers, and campaign advisers. Any cam-
paign expertise was strictly ‘in house’. Indeed, there was little use of
opinion polls at all. Hence, feedback on the party’s performance was
largely through the ‘intuition’ and ‘feel’ of local party leaders and
activists (Farrell and Webb, 2002, pp. 104, 105). This lack of ‘objective’
quantifiable information on what was going on in the electorate could
lead to surprise results. One key example of this was the 1945 election in
the United Kingdom. Churchill, the charismatic war leader, and incum-
bent Prime Minister, was facing the efficient organiser of the home front,
but also modest and unassuming Labour Party leader, Clement Attlee.
The ‘popular’ feel was that having ‘saved’ his country from the clutches
of Nazism, Churchill would easily defeat Attlee. As Howard (2005, pp.
1–12) describes in his book on the 1945 election, the masses of opinions
polls and the close attention that parties, voters, and journalists pay to
them was just not a feature of elections at that time. Hence, there was
Campaigning 145
little sign of the electoral earthquake that was about to erupt, with the
Labour Party securing a firm majority for the first time in its history.
Indeed, the polls that were available were mistrusted and regarded as
inaccurate: ‘Support for Gallup Polls declined when they made an incor-
rect calculation about the 1944 American presidential elections’
(Howard, 2005, p. 5).
In addition, campaigning and party appeal was based on entrenched
social cleavages (Blumler and Kavanagh, 1999, p. 211; Farrell and Webb,
2002, p. 105). This is perhaps the most important feature of the first era.
Social cleavages were seen as the basis of the parties’ self-identification
and ideology as described in Chapter 4. Parties were linked to specific
socioeconomic classes and could rely, more or less, on getting the votes of
that class. The first era is, therefore, very much the age of the class-mass
party (see Chapter 3). Hence, according to Blumler and Kavanagh (1999,
p. 212): ‘few citizens appeared to sift the arguments concerned, tending
to vote instead on group-based loyalties’. This meant that campaigning
was less about persuading voters of the attractiveness of a party’s poli-
cies, and more about mobilising those already sympathetic to the party’s
‘product’. By ‘product’ is meant the overall image that a party is trying to
convey to voters. This includes its ideology, specific policies, and
symbols, as well as the behaviour and public utterances of members, par-
liamentarians and the party leader.
Lees-Marshment (2001, p. 28) refers to ‘product oriented’ parties and
describes such a party as arguing ‘for what it stands for and believes in. It
assumes that voters will realise that its ideas are the right ones and there-
fore vote for it’. In short, in the first era, campaigning is about the mobil-
isation of existing sympathisers rather than the persuasion of swing
voters, of which there is assumed to be very few. The campaign message
is, therefore, based on propaganda and even rabble rousing, what Wring
(1996, p. 102) refers to as ‘the sometimes manipulative appeals of polit-
ical elites’.
Second Era
The second era, as indicated by Farrell’s (1996) term the ‘television revo-
lution’, was dominated by the rise of television as a common feature in
most people’s homes in the industrialised world. The centrality of televi-
sion to this era would suggest that it commenced sometime in the late
1950s and into the 1960s. The rise of television had a number of effects.
First, it saw the early rise of specialists to help parties and individual
politicians – party leaders in particular – cope with the new demands of
television. A new style of communication was required for television-
based campaigning and professionals would now start to play an

increasing role in teaching politicians how to look good on television.
Indeed, the second era of campaigning saw a rise of public relations spe-
cialists generally – not just to cope with television, but also to help parties
utilise increasingly sophisticated opinion polling techniques.
Further, the rise of national television made a national campaign neces-
sary – meaning also the imposition of a national message on local cam-
paigns. Messages and themes were planned, tested and refined by central
campaign committees with the help of professionals well in advance of
the actual election being called. As a result of the rise of the national
campaign it became increasingly important for politicians to stick to the
agreed national campaign ‘script’. Politicians were, therefore, discouraged
from ‘speaking their mind’ unless what was on their mind had been
tested for public acceptability in advance (Blumler and Kavanagh, 1999,
p. 213).
With the rise of the national campaign came also the rise in promi-
nence of the national leadership and in particularly the national leader,
whose image had to be carefully crafted and refined. Indeed, the rise of
television saw a centralisation of campaigning and of parties more gener-
ally. We have just seen how the rise of national television led to the cen-
tralisation of campaign planning. Local branches increasingly lost
autonomy in this area. In addition, the new specialists needed to cope
with television tended to be paid for by the central party. They were,
therefore, under the control of the national leadership. In other words,
the national leadership was in control of the professionals who were best
able to predict what messages were most likely to bring electoral success.
These messages were then imposed on local branches and their cam-
paigning. It is, therefore, argued that the second era saw a centralisation
of the making of the party’s electoral message which by extension also
meant a centralisation of parties’ policy-making processes.
However, while national television necessitated a national message,
national television also made it more difficult to control what messages
potential voters were exposed to. Television presented voters with a wider
array of views than had previously been the case (Blumler and Kavanagh,
1999, pp. 212–13) – hence parties were no longer the main mediators of
political information to their traditional, often class-based, audience. They
now had to compete with other parties for ‘their’ voters’ attention and,
ultimately, for their loyalty too. Add to that the decline in many tradi-
tional class loyalties and we see the beginning of the rise of catch-all
parties competing for the attention of voters, based less on intrinsic loyal-
ties and more on the attractiveness of the product being offered.
While these changes had a major impact on political parties, it is also
argued that at this time the product was still ‘sacrosanct’ (Farrell and
Campaigning 147
Webb, 2002, p. 105). Hence, the message was more or less the same as in
the first era, but there was a new belief in the idea that people beyond the
traditional audience could be persuaded to ‘buy’ into the product. This
fits with Lees-Marshment’s idea of the sales-oriented party which:
Focuses on selling its argument to voters. It retains its pre-determined

product design, but recognises that the supporters it desires may not
automatically want it. Using market-intelligence to understand voters’
response to its behaviour, the party employs the latest advertising and
communication techniques to persuade voters that it is right. A Sales-
Oriented Party does not change its behaviour to suit what people
want, but tries to make people want what it offers. (Lees-Marshment,
2001, p. 29)
Hence, at this stage professionals are used primarily to design a campaign

that will sell the product and help present a party (leader) image that is
attractive. However, experts are not yet in charge of designing the overall
product.
Third Era
The third era is marked by a number of features. The first is the rise of
the Internet. Much has been written about the impact of the Internet on
campaigning, but the Internet is still so relatively new and constantly
changing that its impact is far from clear. Considering how fast the tools
available through the Internet are developing it is uncertain whether there
will ever be complete clarity on this issue.
In addition, the third era has seen the rise of the permanent campaign.
Parties are now not merely starting to prepare for the campaign some
time before the election is called. Rather, they are permanently viewing all
their activities through the lens of campaigning demands. Two further
features are the continuous rise of campaign professionals and the
increasing malleability of the product.
The continuing rise of campaign professionals is said to be having pro-
found effects on party organisations. In the second era external experts
were called in to deal with the challenge of new forms of media in the
form of television. This has continued with the rise of the Internet.
However, with increasingly sophisticated ways of gauging the public
mood in the forms of ever more detailed polling techniques, focus groups,
and advanced analysis of the relationship between social factors and
voting, the external expert has become ever more critical to the success of
a party’s election campaign. Indeed, the need of parties for highly
specialised experts is such that the traditional party organisation is being

sidelined (Plasser, 2001, p. 46). According to Farrell and colleagues: ‘elec-
tion campaigns appear to have outgrown the institutional limitations of
political parties, requiring a role for political consultants [...] to fill this
increasing gap’ (Farrell et al., 2001, p. 26). This is supported by Kolodny
who argues in the context of the United States that ‘if candidates and
issue groups believed that their electoral needs could be entirely served by
political parties, then there would be no market for a bevy of outside
“vendors” such as our contemporary political consultants’, however,
‘modern campaigning has indeed exceeded the institutional capacity of
political parties’ (Kolodny, 2000, pp. 110–11). In short, it appears that
traditional party organisations simply do not have the ability to cope
with modern campaigning without outside help.
Hence, the centralisation of campaigning and party organisations that
started in the second era has continued in the third era. However, with
the increasing importance of the political consultant, power is not just
moving upwards in party organisations to the leadership, but also out-
wards to these ‘for hire’ specialists (Gibson and Rommele, 2009, p. 266).
If party leaders believe that external experts have the answer to what
messages will sell, they may be inclined to give such experts a greater role
in the making of the party’s message.
This development has led to questions over who is actually in charge
of the campaign: parties or external consultants (Farrell and Webb, 2002,
p. 106). In a telling piece of research Plasser (2001) found that political
consultants, especially in the United States are increasingly seeing party
organisations as a hindrance to the effective running of the campaign and
as a growing irrelevance. There is a strong sense that the kind of cam-
paigning that the traditional party organisation was capable off – door to
door at the local level – is much less useful than was the case in earlier
eras of campaigning. What matters is the central campaign run by spe-
cialists, delivered through the mass media and centred on the top leader-
ship.
This rising centralisation is well captured by the New Labour pollster
Philip Gould who wrote that ‘Labour must replace competing existing
structures with a single chain of command leading directly to the leader
of the party’ (Gould, 1998, p. 240). As external expertise becomes
increasingly sophisticated, and expensive, the control of the national
leadership is enhanced. The central message is carefully honed and passed
down the party hierarchy. Deviation from this central line is seen as a
major problem with careful attention being given to both responding to
external critics and, as importantly, to stamping out internal dissent by
‘“rubbishing” or marginalizing internal party critics who are “off-
message”’ (Blumler and Kavanagh, 1999, p. 214).
Campaigning 149
A final point worth expanding on is the increasing malleability of the

product. While the product- and sales-oriented parties in the first and
second eras saw the product as more or less fixed, in the third era the
product is now also open to change if that is what the voters seem to be
asking for – ‘the increasing tendency is less one of selling themselves to the
voters, but rather one of designing an appropriate product to match voter
needs’ (Farrell and Webb, 2002, p. 102). In the third age, the so-called
market-oriented party: ‘designs its behaviour to provide voter satisfaction. It
uses market intelligence to identify voter demands, then designs its product
to suit them. It does not attempt to change what people think, but to deliver
what they need and want’ (Lees-Marshment, 2001, p. 30). There is, then,
an increasing tendency to see parties as businesses engaged in an attempt to
design and produce a product that will attract ‘an increasingly fickle audi-
ence of voters or consumers’ (Gibson and Rommele, 2009, p. 266). There
are obviously some limitations to how far a party can move from its orig-
inal position and still be believable. However, the argument is that parties
are increasingly detached from their original ideological origins rooted in
societal conflicts and ever more dependent for their policy directions on
input from professionals on what will ‘sell’ in the electoral marketplace.
The limits of change
The idea that parties have moved through three eras of development has
gained a considerable amount of currency. A lot of writings on cam-
paigning have at least made a nod in the direction of this view. However,
while the three eras view is popular, widely accepted and also has some
considerable empirical support, there exist a significant amount of variation
and counter trends. The most notable counter trend is the rising awareness
of the importance of local campaigning. As we saw above, one of the key
trends in the move from the first era through the second to the third is the
nationalisation of campaigning, brought about in particular by the rise of
national television. Because of this nationalisation it is the national leader-
ship that is important and the local campaign is seen as of secondary impor-
tance. However, there is increasing evidence of a continuing and indeed
growing awareness of the importance of the local campaign.
One notable contribution to this literature is Marsh’s (2004) article
‘None of that Post-modern Stuff around Here’ which, as the title sug-
gests, is a direct challenge to the themes of the third-era literature. In his
article Marsh concludes that:
Irish elections are certainly not prime examples of the post-modern,

post-fordist era of campaigning which some have claimed to see
elsewhere. Personal contact is felt to matter by politicians, and all

parties make considerable efforts to knock on as many doors as pos-
sible during the weeks of the official campaign [...] the essential style
of campaigns remains personal, with individual candidates seeking to
make an impact and doing so for the most part by the traditional
methods of meeting the folks. (Marsh, 2004, pp. 262–3)
Some of the same trends can be identified in the United Kingdom, nor-
mally a country seen as being first when it comes to follow the path of
‘Americanisation’ – the United States most often being cited as the prime
example of third-era campaigning. The counter trend in the United
Kingdom is illustrated by a comment made by the former cabinet min-
ister Hazel Blears. In a dig at the then Labour leader and prime minster
Gordon Brown’s unsuccessful attempt at embracing Internet campaigning
through a YouTube video, Blears wrote: ‘I’m not against new media.
YouTube if you want to. But it’s no substitute for knocking on doors or
setting up a stall in the town centre.’ While Blears’ comment was very
much part of her personal struggle with a party leader with whom she
did not get on, her comment is backed up by a lot of research. It has been
argued that the image of the modern electoral professional party as a cen-
trally organized ‘army’ with local activists taking orders from the
national headquarters is misleading when it comes to local campaigning.
There is considerable local reluctance to simply follow orders from the
centre. Further, it has been suggested that campaigns are more effective
when they are locally organized (Whiteley and Seyd, 2003).
Others have also stressed the importance of local campaigning, but
have also pointed to the importance of coordination from the centre
(Fisher et al., 2005). This is especially the case in the key target seats that
a party needs to win or hold in order to win the election overall. It is
argued that the national leadership is fully aware of the importance of
careful use of local activists in certain key areas and spend a great deal of
efforts to get the local campaigning right. Indeed, Fisher and colleagues
go so far as to argue that the centrally coordinated local campaign – with
activists doing traditional, that is first era, door to door campaigning – is
as, or even more, important than the national media campaign that has
received so much attention in the third era literature:
there is no doubt that constituency campaigning is now seen as more

significant than ever by the parties and possibly even more significant
than the national campaign focussed on the mass media. The planning
of constituency campaigns has become increasingly integrated into the
parties’ overall campaign strategy [...] Arguably, then, the campaigns
in the constituencies now dominate the parties’ overall campaign
Campaigning 151
strategy. The battles that matter are those in the target seats, not the
one played out in the national media. (Fisher et al., 2005, p. 18)
The importance of local campaigning has also been recognised elsewhere.

There is a longstanding literature from the United States on the role that
personal contact and local activism has on turnout at elections (Green
and Gerber, 2008). Indeed, it could be argued that the entirety of Barack
Obama’s road to the White House, from the primaries to both presiden-
tial elections, was a lesson in the importance of building grassroots
organisations and how such organisations can overcome obstacles such
as lack of financial muscle. In his book on the first Obama campaign
Plouffe writes:
Our grassroots supporters again deserve the lion’s share of credit. The
farther we got from Iowa, the more important our volunteers were.
The campaigns in the states were shorter and we had fewer staffers on
the ground. The volunteers built our campaigns in the February states
and executed incredibly well in the closing weeks. They recruited more
help, identified supporters, and made sure people knew where to vote
and caucus. (Plouffe, 2009, p. 181)
Later he writes: ‘Our secret weapon, day in and day out, was our army of
volunteers, real people who brought Obama’s message and ideas to their
neighbors, co-workers, and fellow citizens, guided by our extraordinary
staff’ (Plouffe, 2009, pp. 379–80). Plouffe is obviously writing as a firm
believer in Obama’s message and as someone intimately involved in the
campaign, but even ignoring phrases such as ‘secret army’ and ‘extraordi-
nary staff’ this seems to be an excellent example of what Fisher and col-
leagues describe in their work on the importance of centrally coordinated
local activism (‘army of volunteers’ ‘guided by our extraordinary staff’).
It is true that Obama’s campaigns became known for their innovative
use of the Internet. However, it is worth noting that it also seems that
their use of the Internet was to a large extent about facilitating the
growth of a local activists base which could then be encouraged to go out
and do the traditional ‘shoe leather’ work of knocking on doors and
getting out the vote. For example, Plouffe notes that ‘our website and its
social-networking component were heavily trafficked and becoming a
real resource for our supporters’ (Plouffe, 2009, p. 55).
An appreciation of the importance of local campaigning can also be
found in work on Australian elections. In surveys of campaign profes-
sionals it was found that such professionals certainly wanted to conduct a
modern campaign, but at the same time valued the ability of party organ-
isations to do indispensible local work: ‘The Australian example shows
that it might be possible to conduct high-tech campaigns on air while

valuing the strength of an effective party organization in closely contested
marginal seat constituencies’ (Plasser, 2001, p. 49). So, there certainly
seems to be some evidence suggesting that a number of campaign profes-
sionals in the United States see the party organisation as more of a hin-
drance. However, the examples from the United Kingdom, the Obama
campaigns and Australia suggest that the most effective form of cam-
paigning is one that combines a modern professionalised central cam-
paign of the ‘third era’ kind with centrally coordinated traditional local
‘first era’ door-to-door activism. Overall this has led Pattie and Johnston
(2009, p. 411) to conclude that ‘constituency electioneering has become
established as an important element of postmodern [that is third era]
political campaigning’.
However, there are also variations in the importance of local cam-
paigning in different countries – a variation to some extent based on the
political system, and especially the electoral system of a given country. In
this context it has been found that direct contact between parties and
voters is most notable in elections that use the single member plurality
electoral system, for example the United Kingdom and the United States
(Karp, Banducci and Bowler, 2008). Here there is far more focus on indi-
vidual candidates and, therefore, a greater perceived need for candidates
to go out and meet the voters and be seen to do so. By contrast in the
more party-focused environment of PR electoral systems (especially List
PR) there is less pressure for local activists and candidates to go door
knocking.
However, there is still some evidence that ‘shoe leather’ campaigning is
also becoming a feature in democracies using proportional representation
electoral systems. In Denmark is has been noted in that some parties,
especially on the left, have started to focus on getting volunteers out to
knock on doors and engage voters face-to-face.
Further, Plasser found that ‘third era’ campaigning techniques were
somewhat unevenly spread across the globe. Professional, marketing and
media driven campaigning is at its height in the United States, and is also a
strong feature in Europe, West and East, Latin American, Russia and East
Asian Democracies (Plasser, 2001, pp. 45–6). By contrast this type of cam-
paigning is much less widespread in Africa and India. Here we still see
much more traditional local and personal campaigning techniques domi-
nate, with little professionalization and much less media focus (Plasser,
2001, pp. 50–1). This is perhaps not so surprising. Quite apart from the
lower level of television and Internet penetration in these regions, the
‘product’ that the parties offer is also very different. As Chapter 4 showed
Africa is dominated by personal and clientelistic relationships where
access to resources through control of the levers of government is much
Campaigning 153
more important than a carefully crafted campaign message derived from

detailed study of voter demands. A lot of Indian politics has become very
local with a both a lot of small parochial parties and a heavy local orienta-
tion in the major parties. There is quite simply neither the means nor the
need for third-era style of campaigning in India or democratic states in
Africa the way there is in certain other democracies.
Drivers of change
It is now worth considering what drives change. Three driving forces will
be considered. First, as is discussed in both Chapter 3 and Chapter 4,
underlying societal changes have had a major impact on the life of
parties, and are also an important factor in understanding campaigning
change. Second, technology has played a major role. As was observed
earlier, one of the main causes of the rise of campaign professionals was
the appearance of new communications technologies which parties
needed help utilising. However, it has also been pointed out that some-
times parties operating in similar social and technological environments
will adopt modern campaign techniques to different degrees (Gibson and
Rommele, 2001). Hence, as a third issue we need to look at party-specific
factors.
Social changes
Chapter 3 and Chapter 4 show the intricate connection between political

parties and underlying social forces. Indeed, one of the main features of
the party-politics literature is the connection between society and parties
– their ideology, organisation and behaviour. Hence, if those forces
change, so will political parties – including how they campaign.
The main social changes that are of relevance to party organisational
change relate to the decline of traditional political cleavages, especially
class. As voters increasingly become divorced from traditional societal
conflicts they start voting according to a wider and more amorphous set
of concerns. This means that parties can no-longer rely on traditional
conflicts to predict voting patterns. As Hindess argued: ‘it is [...]
becoming increasingly difficult to support one of the other major party
on the basis of class loyalty – to support Labour because one is working
class, or the Conservatives because one is not’ (Hindess, 1971, p. 10).
Hence, it is no longer enough to merely mobilise already sympathetic
voters. It is necessary to also persuade them of the virtues of the party’s
position, as was seen in the move from first- to second-era campaigning.
However, ‘persuading’ requires radically different methods than

‘mobilisation’. In order to be able to persuade voters to vote for a partic-
ular party, that party needs to find out what voters will find persuasive.
That in turn leads to a massive increase in the need for information about
the demands and wishes of voters, something that requires expert help,
hence the rise of professionalization (Smith, 2009, p. 556). The argu-
ment, therefore, goes that societal changes were one of the main factors
in the rise of a new form of party employing new forms of campaigning
techniques: ‘the catch-all party emerge[d] from – and to suit – a specific
form of society’ (Smith, 2009, p. 559).
Technological change
A significant part of what has encouraged or forced changes in the way

parties campaign is related to technological change. It is true to point out
that technological change clearly is not the entire explanation. This has
been acknowledged by Farrell and Webb, who adopt what they refer to
as ‘soft technological determinism’ (Farrell and Webb, 2002, p. 103).
That is, technology is part of the explanation for change, but only one
part.
New technology has provided means by which parties can gather
increasing amounts of information, as well as ever more sophisticated
ways of analysing it.
In addition, first radio, and most significantly television and the
Internet provided parties with both a vastly increased potential for pow-
erful campaigning strategies, but also a challenge in how to best utilise
that potential. Hence, technology provided new ways of campaigning;
utilising those new ways required experts; failing to utilise these new
ways meant electoral defeat if rivals were better or faster at using new
methods. At the same time the spread of telephones and later the Internet
made questionnaire polling far easier than it had been in the past (again
with the help of experts – Twyman, 2008). The rise of ever more pow-
erful personal computers made the analysis of such data both faster and
more sophisticated. In short, technological change provided parties with
many opportunities for more effective campaigning, but at the same time
also created challenges and dangers – challenges in getting the most out
of the technology; and dangers if the technology was used badly, such as
a bad performance on television or YouTube, or by reacting slower than
rivals. As Kirchheimer pointed out, the adoption of new ways of doing
campaigning is a competitive phenomenon: ‘A party is apt to accommo-
date to its competitor’s successful style because of hope of benefits or fear
of losses on election day’ (Kirchheimer, 1966, p. 188). Hence, successful
Campaigning 155
use of a certain type of technology is likely to increase demand for that

technology, which brings us to the other side of the equation – party-spe-
cific issues affecting changes in campaigning.
Party-specific issues
There are clearly major factors that have encouraged and sometimes
forced changes in how parties operate. However, not all parties react the
same way to these outside forces. At a national level it may be possible to
explain variations between countries based on their relative technological
and social context. Hence, it is perhaps not surprising that for example
Africa and India should be using methods more associated with the first
and possibly second eras of campaigning compared to say the United
States or Europe. Considering the much lower level of technological pen-
etration in Africa and India – such as television and PC ownership and
access to the Internet – the availability of certain campaigning techniques
will simply be that much lower. However, these factors do not explain
why we sometimes find parties in the same political system adopting dif-
ferent campaigning techniques from across the three eras. In order to
understand such differences we need to look at party specific factors
(Gibson and Rommele, 2001). It has been suggested that a number of key
factors may help explain why a party becomes inclined to adopt new
approaches.
First, it is argued that a major motivating factor in adopting a new
approach to campaigning will be if the party’s primary goal is no-longer
being met (Gibson and Rommele, 2001, p. 36). If that is the case the
party will be motivated to change its behaviour to turn around its for-
tunes. Good examples of this would be the US Democrats under Bill
Clinton, the UK Labour Party in the mid-1990s when Tony Blair became
leader, and the UK Conservative Party under David Cameron. In this
context, suffering a string of heavy defeats is likely to trigger some
serious soul-searching in a party, which may lead to new campaigning
techniques being adopted. This is especially the case if such defeats lead
to a change of leadership bringing with it new approaches and ideas.
Another factor which is likely to have a significant impact on cam-
paign innovation is having a healthy bank balance (Gibson and
Rommele, 2001, p. 37). Hiring the professionals that are necessary for
utilising many modern campaign strategies is only likely to happen if
there are substantial resources in place – such help does not come cheap.
Intuitively this does make sense. If you do not have money to buy the
expensive equipment and services seen as necessary for modern cam-
paigning, then you will not get very far in that direction. However, at the
same time, as the first Obama campaign illustrated, operating a tight

budget can itself be a source of innovation. Plouffe writes that:
During the general election, on a conference call we held with labor

leaders, one of Clinton’s big labor supporters who had spent millions
on her behalf in the primary derided our belief and trust in volunteers.
He insisted that we needed to pay people to do things like door-
knocking and phone calls [...] ‘Well, our volunteers certainly did a
number on you in the primary.’ I reminded him. He was one of many
who, even after witnessing the endgame, discounted our grassroots
strategy in favor of the old dusty playbook. (Plouffe, 2009, p. 182)
Nevertheless, having said that, money does allow the use of certain
approaches that would not otherwise be possible. A major element in
Plouffe’s description of the first Obama campaign is how they built up of
a solid donor base and beat fundraising targets and records. Innovation
driven by scarce resources is fine, but having lots of resources is arguably
even better.
In addition, Gibson and Rommele (2001) argue that a party which has
‘existing norms of internal hierarchy’ – or in other words, a tradition of
strong leadership, will find it easier to adopt the top-down approach seen
as central to second- and third-era campaigning. Conversely, parties with
long histories of internal opposition to leadership control will find it
more difficult to accommodate the strategic swiftness and hierarchical
organisational structure associated with third-age modern profession-
alised campaigning. This is a theme also evident in connection with the
challenges of intra-party democracy in policy making as we saw in
Chapter 7. A good example of this issue is the UK Labour Party. One of
the reasons why Gould’s ‘single chain of command’ line (Gould, 1998, p.
240) was so controversial (‘This offended everybody’ (Gould, 1998, p.
240)) was the Labour Party’s long history of formal (if not actual) adher-
ence to internal democracy.
Finally, it is argued that a right-wing party is more likely to be sympa-
thetic to ‘the principles of marketing and use of outside consultancy firms
underpinning professionalized campaigning, than a left wing party’
(Gibson and Rommele, 2001, p. 37). Marketing, a quintessential free
market capitalist pursuit, perhaps comes easier to a party of the right
than the left.
The overall conclusion is, therefore, that the most likely adopter of
modern campaigning techniques is a ‘well-funded, mainstream, right-
wing party with significant resources and a centralized internal power
structure that has recently suffered a heavy electoral defeat and/or loss of
governing status’ (Gibson and Rommele, 2001, p. 37).
Campaigning 157
Impact of changes
From the above outline of the changes that have happened to election
campaigning, it should be fairly clear that these changes are likely to have
had significant organisational implications for the parties involved.
When looking at the changes wrought by the development in cam-
paigning it is worth noting that there seems to be a close connection
between the three eras of campaigning and party types. As Smith (2009,
pp. 557–8) points out there is a great deal of overlap between the descrip-
tion of campaigning in the first era – targeting specific social groups and
reliance on a mass membership organisation – and the literature on the
mass party. Similarly, second- and third-era campaigning overlaps with
the literature on the catch-all and electoral professional party types.
Hence, many of the changes associated with the move from first-, to
second- and to third-era campaigning are dealt with in Chapter 3 and
will not be repeated here. However, one issue will be dealt with here.
There is a potentially significant problem in combining three central
elements of third-era campaigning: 1) the increased centralisation of
the party organisation associated with campaign modernisation; 2) the
central role played by local campaigning in modern elections; and 3)
the challenge of retaining and incentivising members (see Pettitt, 2012a).
There is a significant debate over whether internal party democracy is
a good thing or not for political parties and democracy generally as we
saw in Chapter 7. However, there are reasons to believe that having influ-
ence on the direction of a party is a central motivating factor in party
member activism (Whiteley and Seyd, 2002). At the same time Lilleker
(2005) suggests that there are severe challenges in combining modern
political marketing with internal democracy. This is to some extent a
longstanding issue. As is described in Chapter 7, one of the key chal-
lenges of building internal party democracy is that while the wheels of
democracy often move slowly, the electoral market can move very
quickly indeed. There is, then, an unresolved paradox at the heart of
modern campaigning: Modern, third-era, campaigning requires a great
deal of speed, centralisation, and local participation. Yet there is a clash
between speed and centralisation on the one hand, and motivating people
to be not only members of a party, but also to be active, something which
requires giving them a sense of ownership of the organisation. How can
central control of an organisation be combined with, at the very least, a
feeling of widespread ownership?
This is a problem that does not seem to have been solved by any
modern party. The UK Labour Party saw both a significant increase in
centralisation and a rise in membership in the late 1990s. However, since
about 2001, the membership rise has turned into a massive decline. Once
the Labour Party was in government and started doing things the sense of
ownership that was part of the 1990s rise very quickly dissipated. On the
other side of the Atlantic Obama was able to build an impressive grass-
roots organisation during his first presidential primary and general elec-
tion campaigns. Arguably though this was partly because he was a
relatively unknown quantity and people were, therefore, able to fill in the
blanks as they saw fit (Pettitt, 2012a). To some extent he was able to
repeat that feat in 2012, although the earlier enthusiasm has certainly
faded. There is currently no straightforward solution to this problem.
Only a growing awareness that modern campaigning requires local vol-
unteer participation and that such participation requires incentives –
ideally, it has been argued, through giving activists a sense of ownership
of the product being sold on the door step (Pettitt, 2012a). How this can
be achieved in a centralised organisation is currently an unanswered
question.
Financing
If there is one thing that most scholars and practitioners can agree on in
the rise of modern professionalised campaigning, then it is the fact that
such activities are expensive. Hence the last issue the chapter will deal
with is the issue of party financing. This is an area which, as Fisher and
Eisendtadt (2004, p. 619) point out, is somewhat under-researched. This
may be partly explained by the fact that apart from when it sparks a
major scandal, which party financing has a tendency to do, it is also an
area which it can be difficult to learn very much about.
Often parties are required by law to open their accounts to public
scrutiny, especially if, as is increasingly the case, the state has a role to
play in the funding of parties. Even so, parties tend to be reluctant to talk
too much about their finances and their donor base. Indeed, it has been
pointed out that one of the costs of a high level of accountability and
transparency in party financing is that it can scare off potential donors
(Fisher and Eisenstadt, 2004, p. 621). Parties are not popular and being
too closely associated with them could bring bad publicity for the donors
– and possible accusations of favouritism and even corruption.
However, money is central to the life of parties: The expense of
modern campaigning has already been mentioned. In addition, parties
need money to maintain their organisations in-between elections. This
will certainly cost less than the capital intensive weeks leading up to an
election, but as we saw above, one of the characteristics of third-age cam-
paigning is its permanency. In other words, even if the permanent cam-
paign may vary in intensity it will still consume money even at its less
Campaigning 159
active stages. Further, all but the most amorphous party organisations
will maintain some form of permanent presence in the form of national
and regional offices in-between elections. This is particularly the case the
closer a party is to the mass party model with is large membership base.
Hence, parties need money and the amount they have and the way they
get it is of key importance for how democracy operates.
Hence, party financing is important for a number of reasons. First,
while having local volunteer activists is important even, or perhaps
especially, in modern campaigning, there are few things that volunteers
can offer that cannot also be bought with money (Fisher and
Eisenstadt, 2004, p. 619). That means that unequal funding streams
have the potential to result in unequal power. Unequal power is not
necessarily a problem in democracy – after all, that is a normal aspect
of politics. However, if some parties are permanently better funded
than others that could be a problem for the fair competition between
them. In other words, money can result in an uneven playing field,
which has been one of the arguments for increased state funding (van
Biezen, 2004, p. 703).
Second, where parties get their money from has a significant impact on
politics. As described in Chapter 3 the mass party was defined by its close
relationship with civil society. Unlike the cadre party and its few but
wealthy supporters, the mass party relied on thousands of small contribu-
tions (Duverger, 1964, p. 63). In return for their membership subscription
the members were given a stake in the party, its ideological development
and policy stance (even if that stake was often more imagined than real –
see Chapters 3 and 7). Hence, the mass party was a key part of civil
society and served as a bridge between citizens and politicians. However,
one of the key issues in modern party politics is the decline of the mass-
party model. In its stead has come party organisational forms that are
defined more by their relationship with the state than the people (van
Biezen and Kopecky, 2007, pp. 236–7) and this is exemplified by the
changing nature of party funding. Rather than the multitude of many
small contributions, parties are increasingly relying on fewer big dona-
tions, in the manner of the cadre party, and state funding as exemplified
by the cartel party model. These two funding streams have profound
implications for the role of parties in democratic states.
One of the implications of fewer large donations is that the sources of
these donations, be they companies, organisations or rich individuals,
may want something in return for their largesse. Hence, there is a fear
that parties may favour those who fund them over other, less well-off ele-
ments in society (Fisher and Eisenstadt, 2004, p. 620; Hopkin, 2004,
pp. 632–4). There is, then, a fear that some people may buy influence for
their particular point of view, potentially to the detriment of the majority
of the population, what West (2000) has referred to as ‘check- (or

cheque-) book’ democracy. There is then a worry about favouritism and
even outright corruption. This latter issue, outright corruption, has fre-
quently had a severe impact on the politics of democratic states. Italy saw
massive changes in its political landscape due to corruption scandals
emerging in the early 1990s. The reputation of Germany’s ‘unification’
Chancellor Helmut Kohl was largely destroyed due to funding irregulari-
ties during his time as leader of the CDU. Funding scandals involving the
French president Sarkozy dominated political news in France for several
months in 2010.
One possible solution to special interests buying influence through
donations to parties is state funding, which is indeed an increasing
feature of democratic states (see van Biesen; 2004; van Biezen and
Kopecky, 2007). If the state is the main source of funding there is some
guarantee that special interests will not buy influence and much greater
transparency in the raising and spending of financial resources can be
achieved. However, there is also a downside to state funding. One of the
main trends in party development (see Chapter 3) is the way parties are
becoming ever more absorbed by the state:
Parties have traditionally been understood in terms of their permanent

linkage with society and their temporal linkage with the state. Parties
neither depended on their state for their resources and legitimacy nor
were they managed or controlled by the state. Recent processes of
organization-building and organizational adaptation, however, have
reversed these patterns [...] parties are now perhaps best understood in
terms of their temporal linkage with society and their more permanent
linkage with the state. (van Biezen and Kopecky, 2007, p. 237)
Research has shown that state funding for parties is steadily increasing,
and in new democracies has started out at a much higher level than in
more established democracies (see van Biezen 2004; van Biezen and
Kopecky, 2007). This means that the state has a much greater control
over the internal life of parties through the creation of monitoring mecha-
nisms for ensuring the legitimate use of state funds. In addition, parties
are increasingly cut off from society as they now have an alternative
source of income, as argued in the ‘cartel party’ literature (see Chapter 3).
State and special interest funding of political parties, therefore, seem to
have important implications for the functioning of democracy.
Indeed, Hopkin argues that the mass-party way of funding political
parties is normatively more attractive than the alternatives (such as state
funding or major donors – either external to the party or internal as with
Berlusconi’s Forza Italia). However, somewhat pessimistically, he also
Campaigning 161
argues that the mass party model of funding is ‘utopian’ (Hopkin, 2004,
p. 646).
Hence, the funding of political parties is an issue that is central to the
workings of democracy in many ways, and seems to be an area that
deserves much more attention that it is currently getting.
Conclusion
Even though there is some disagreement in the literature over the exact
nature of how campaigning has changed, there is no doubt that it is
changing under the accumulated pressure of social and technological
developments. It is equally clear that this change is having a major impact
on how parties operate and how they are funded. Hence, a concern with
how election campaigns are won is not just about understanding what
works and what does not, but also about understanding the changing
relationship between parties and wider civil society.
Chapter 9
Government
One of the defining features of a democratic state is the free competition

for government power. In modern democracies that competition is
centred on political parties. When citizens vote they do so mainly to
determine which party’s or parties’ politicians will be able to enter gov-
ernment office. By and large, voters tend to be concerned more with the
party affiliation of a particular government minister, rather than his or
her individual qualities. In short, the long-term job security of any min-
ister is determined by the fate of their party, not their individual (in-)com-
petence. It is, therefore, important to understand how party politics
interacts with how governments are formed.
Getting into government is arguably the most ambitious, difficult and
costly goal a party can have. The benefits of government office in terms
of being able to shape society in a certain direction are huge. However,
being in government also carries a high price. Winning elections costs
money; will involve painful compromises; and government responsibility
inevitably means getting the blame for any number of sins. This means
that not all parties desire to enter government office, at least in the short
term. That will be the topic of the next section. Having explored the path
a party takes towards government ambition the chapter will then look at
how parties get into government with a particular focus on coalition for-
mation. The chapter will finish by looking at what parties get out of gov-
ernment participation in terms of ministerial portfolios.
Deciding to aim for government office
One of the definitions of a political party explored in Chapter 1 was as

an organisation dedicated to acquiring control of the levers of govern-
ment. This would suggest that based on some views of political parties,
their main goal is acquiring government power. However, in this book
the definition used is that a party is a body of people fielding candidates
for publically elected legislative assemblies. Why they do that is then
left open. That would suggest that gaining government power will be
only one among several possible goals. Indeed, the whole idea of
focusing the definition of a party on the electoral aspect is that the ulti-
162
Government 163
mate goal behind entering legislative assemblies may be many and

varied.
It was also argued in Chapter 1 that the idea of a party as a unified
rational actor is only a convenient shorthand, and a slightly misleading
one at that (Laver and Shepsle, 1999). A party is made up of many
internal actors, each of which may have different and contradictory
goals. This is exemplified by the idea of curvilinear disparity (see Chapter
7). Strøm (1990) argues that parties generally pursue a combination of
three goals: office (generally seen as being executive office, but could also
be legislative office); policy realisation; and votes. No goal is ever
pursued to the complete exclusion of the other two, but rather a party
will strike a balance between the three. It is worth noting that Strøm’s
model is zero-sum, which means that the pursuit of one goal will mean
that less attention can be given to the other two. Hence, the pursuit of
office will lead to losses elsewhere, and depending on a number of
factors, a party may decide that the pursuit of office is not worth the
costs, at least in the short term. Getting into government will, therefore,
only be one element in a complicated mix of goals with associated costs
and benefits attached. Indeed, goals will change over time as the result of
external and internal stimuli (Harmel and Janda, 1994). In short, when
considering the issue of government ambition in the context of party poli-
tics we will first have to consider what leads a party to pursue that goal
in the first place.
It is rare that a party will pursue one of Strøm’s three goals to the
exclusion of the others, but one may at various times weigh more heavily
than the others. Hence, government is not the only goal a party will
pursue. Virtually any party that won more than 50 per cent of the seats
in its country’s parliament would use that power to take control of the
levers of power. However, in many countries achieving more than 50 per
cent of seats is all but impossible for a single party. Hence, to enter gov-
ernment would require making deals with other parties, something that
would entail compromise and, therefore, sacrifices. Even in country
where parties routinely win 50+ per cent of seats, this electoral success
will involve making sacrifices to achieve a sufficiently broad electoral
appeal. A number of factors will influence the extent to which a party is
willing to make the sacrifices necessary to enter government.
Perhaps the most important factor in explaining a party’s government
ambitions is its origins. As we saw in Chapter 3, a party’s characteristics
depend more upon its history, especially on how the organisation origi-
nated and how it consolidated, than upon any other factor. So,
Duverger’s cadre parties (see Chapter 3) could be argued to have the
pursuit for government office written into their organisational DNA.
These parties, emerging from loosely organised individuals, where wholly
and exclusively focused on acquiring office for their own benefits and for
the benefit of their local supporters. We are here primarily talking about
parties of relatively old origins such as liberal and conservative parties in
for example the United Kingdom, United States, France and Scandinavia.
These parties started out as office-seeking groups in parliament. Only as
a result of the extension of the franchise in the late 1800s and early
1900s were they forced to add on extra-parliamentary organisations to
assist with office seeking in a changed electoral environment. Such
parties, therefore, never really made a decision to pursue government
office, but rather were ‘born’ to do so.
Something similar can be said of traditional working-class centre-left
social democratic and labour parties. These parties were created specifi-
cally with the purpose of defending the interests of the working class.
They were also founded on the basic idea that government office would
be the best way of realising that goal. This is exemplified by the British
Labour party of which Attlee wrote in 1937 that a ‘further characteristic
of the British [socialist] movement has been its practicality [...] From the
first, British Socialists have taken their share wherever possible in the
responsibility of Government’ (Attlee, 1937, p. 30). In short, these parties
were also created with government office in mind from the very begin-
ning. Duverger notes of the French Socialist Party that ‘the party aims at
the political education of the working class, at picking out from it an elite
capable of taking over the government and the administration of the
country’ (Duverger, 1964, p. 63). As with the earlier liberal and conserva-
tive parties, government ambition was a written into these parties from
the beginning.
It is true that this government ambition has at times been less strong
for some social democratic parties. Several have at various times been
marked by the presence of strong radical socialists groupings within
them. These groupings frequently had far less government oriented prior-
ities than the rest of their party, and were instead much more focused on
ideological ends. The government-leaning orientation of social demo-
cratic leaders would lead to charges of ideological betrayal from the
radical elements in the party and bouts of internal struggle. Many of
these radical socialist elements left their mother parties to form commu-
nist parties in the wake of the 1917 Russian Revolution. Nevertheless,
despite the presence of such radicals, social democratic parties have, like
the older liberal and conservative parties, still tended to be government
oriented from birth.
However, the picture is much more varied and changeable for many,
often small, but frequently critically placed, newer parties. It has been
argued that party systems since the 1960s have been marked by the rise
of a large number of new parties (Deschouwer, 2008, p. 1). These parties
Government 165
were created to champion new political cleavages (see Chapter 4), revive
old allegedly neglected issues, ‘purify’ existing ideologies from the com-
promises of the old parties (Deschouwer, 2008; Lucardie, 2000) or were
guided by somewhat more amorphous ‘anti-establishment’ rhetoric
(Bolleyer, 2008, p. 31). In short, many of these parties had far more
policy oriented stances and, at least to start with, saw government ambi-
tion as selling out to the old establishment. These parties, therefore, have
a far more complicated relationship with government participation than
the older ‘built for office’ parties of the centre right and the centre left
that preceded them.
Many factors will have an impact on if and when a party, primarily
oriented towards policy realisation rather than office, will still at some
point decide to actively pursue government participation. One such factor
is how a party weighs the costs and benefits of being in government,
which are many and varied. The benefits of being in government are
pretty straightforward and can be summarised as ‘power’ and ‘resources’.
Governments usually hold the initiative when it comes to legislation.
Even in democracies with strong traditions of negotiation between gov-
ernment and opposition, it is usually the bills proposed by the executive
that get most of the attention and have the highest likelihood of making
it onto the statute books. The opposition is often limited to amending
government initiatives at best or impotent rhetoric at worst. Being in gov-
ernment also gives access to the many resources of the state which can be
used to strengthen a party’s organisation and as patronage for supporters.
Having civil servants to help with the details of policy proposals is also a
great boon for parties. Katz and Mair’s (1995) cartel thesis is an example
of the importance of access to government resources and is based on the
idea of certain parties trying to monopolise access to those resources (see
Chapter 3). In addition, as is explored in Chapter 4, in many African
democracies, having access to the resources of the state to be used as
patronage for supporters is far more important than ideology-driven
policy realisation. Finally, there are many smaller perks of being in gov-
ernment, including chauffeur driven cars, big offices in ministry build-
ings, and the prestige of representing one’s country at international
summits.
However, there are also many costs associated with being in govern-
ment. The primary cost is perhaps that with power comes responsibility.
Not necessarily meaning that there is an expectation that power should
be wielded responsibly (even if that is normatively desirable), but rather
in the sense that when things go wrong it is usually the government
which gets the blame. Indeed, even if things are not particularly going
wrong, being responsible for allocating limited resources will always
involve disappointing someone, somewhere which may then lead to the
withdrawal of electoral support. The longer a party remains in govern-

ment the more decisions it will have to make and the more people it will
disappoint.
In addition, Strøm (1990, p. 573) argues that voters will usually value
consistency between policy pronouncements by a party and what it then
actually does. However, being in government can constrain a party’s
freedom of action with may mean that it is forced to deviate from what
they announced they would do, thus putting off voters. One example of
this is the Danish government after the 2011 election. In order for the
Social Democrats and the smaller Socialist People’s Party to form a gov-
ernment they needed the Social Liberals onboard. That involved a
number of compromises which led to accusations that the Social
Democrats and the Socialist People’s Party had broken their pledges from
before the election. Even if these accusations are not necessarily entirely
accurate they stuck, resulting in very poor opinion poll showings for the
Social Democrats and the Socialist People’s Party (see Figure 9.1). There
is then an unavoidable electoral cost to being in government. This is
further illustrated by Johnston and Pattie (2008) who show that chal-
lenger parties get more votes for their campaign spending than incumbent
parties. Put slightly simplistically, votes ‘cost’ more for incumbent parties
than for opposition parties.
Further, it has been argued that being in government requires more in
the way of internal discipline than being in opposition (Pettitt, 2012a, see
also Chapter 7). It is more difficult for a party in government to tolerate
internal dissent than it is for a party in opposition, if nothing else because
a governing party is likely to have more media attention than an opposi-
tion party. Journalists will usually see any sign of disagreement as a ‘good
story’ and present it in dramatic terms. Hence, disagreement in a gov-
erning party is much more likely to be covered in the press than disagree-
ment in an opposition party. However, clamping down on dissent usually
brings with it a backlash somewhere down the line from those who have
been affected, leading to further negative media coverage.
Also, any form of office holding by a party will generate intra-organi-
sational pressures. Any party with both parliamentary and an extra-par-
liamentary sections will have to manage the relationship between the
two, something which has often proven itself to be difficult. Entering
government means adding an additional layer (government ministers)
that has to be managed. Now it is not just a matter of having to manage
the relationship between party organisation and parliamentary party.
Instead it is a tri-party relationship between party organisation, parlia-
mentary party and government ministers. Managing this relationship has
often proven too much, especially for organisationally ‘immature’ young
parties (Bolleyer, 2008).
Government 167
25
Social democrats
20
Percentage
15
10
Socialist People’s Party

5
0
Election Oct-11 Nov-11 Dec-11 Jan-12 Feb-12 Mar-12 Apr-12 May-12
2011
Figure 9.1 Support for the Social Democrats and the Socialist People’s Party in
Denmark after the 2011 election
Source: DR (Danish Broadcasting Corporation) http://www.dr.dk/nyheder/politik/meningsmaalinger

(accessed 21 November 2013).
A final problem of incumbency worth mentioning is the danger of

becoming part of the very establishment that many young parties were
created to fight. Especially parties of the far right have been based on
more or less vague anti-establishment, anti-politician, even anti-party sen-
timents – generally speaking on an opposition to ‘normal’ politics.
Joining a government is likely to be seen by some as a sign that a given
party has become part of ‘normal’ politics, which may serve to alienate
core supporters – a group that for a fringe party is likely to be fairly small
in the first place (Bolleyer, 2008, p. 31). Arguable this is what happened
to the Freedom Party of Austria (Freiheitliche Partei Österreichs) when
they were in government 2000–05. After gaining significantly in the polls
in 2000 the party proceeded to tear itself apart over the next five years
under the pressure of being in government.
An important point to consider when analysing how parties will weigh
the pros and cons of government office is the type of democracy a party
operates in – here taken to mean where a country sits on the spectrum
between consensus and majoritarian democracy (Lijphart; 1984; 1999).
Consensus democracies tend to be dominated by multi-party systems,
coalition and/or minority governments. Hence power is shared between
several parties and the opposition is likely to have a fairly significant level
of influence, either by bargaining with minority governments or through
a powerful committee system. In majoritarian democracies on the other
hand power tends to be concentrated in single-party majority govern-

ments and the opposition is often more or less completely excluded from
legislative influence. As Sartori (2005, p. 308) writes:
In the two-party context ‘winning’ means a plurality, and whoever

does not win a plurality simply loses. In the more-than-two party
system, instead, winning means gaining votes or seats – and, further-
more, a party may be more interested in winning in terms of posi-
tioning than in terms of returns. And there is a world of difference
between the winner-takes-all and the greater-share notions of winning.
In a two-party majoritarian style democracy a party can either win and

have influence, or lose and be impotent. In a multi-party consensus style
democracy on the other hand the political game is about manoeuvring for
a share of power which can come in many forms other than government
office. Hence, the further towards the consensus end of the spectrum a
country is the more power the opposition will have and the less attractive
government will be for a largely policy oriented party. If policy goals can
to a significant extent be realised in opposition there is less reason to pay
the costs of being in government (Bolleyer, 2008; Strøm, 1990).
Therefore, opposition will be a much more attractive option in a con-
sensus democracy than a majoritarian democracy. One particularly
extreme example of this was Denmark in the 1980s where the centre-
right minority government effectively ceded control of certain areas of
foreign policy to the opposition (Damgaard and Svensson, 1989). The
government was repeatedly defeated on matters of foreign policy, but did
not see that as sufficiently important to warrant calling an election. At
the same time the opposition, which was split between the left and the
right, could not get together on a motion of no confidence in the govern-
ment. Hence, the government continued, but with the opposition effec-
tively making policy on certain matters.
Nevertheless, regardless of how much legislative influence the opposi-
tion has, the government will more often than not still be able to exert a
significantly greater control over politics in a given country than the
opposition. As was argued earlier, legislative initiative usually rests with
the government. So, while the opposition parties may have significant
power to amend legislation being proposed by the government, they are
still reacting to a government-controlled agenda, rather than taking inde-
pendent initiatives. In addition, the detailed implementation of policy is
done by the government (Bolleyer, 2008) and any negotiations on the
international scene are also usually done by government ministers.
This leads to the second reason why policy-focused parties may
decided to adopt a more government office oriented approach: failure to
Government 169
influence policy when in opposition. A consistent failure to achieve a sat-

isfactory level of policy influence would encourage a policy-oriented
party to seek greater policy influence through government office. In addi-
tion, what counts as a ‘satisfactory’ level of policy influence may well
change over time. The longer a party is in opposition the less satisfying
amending government initiatives may get. For a young party lesser
changes may be enough, but over time this may lose its appeal as frustra-
tion at the limitations of opposition sets in.
Getting into government – coalitions
Even if a party has been moving in a government-seeking direction that is

only the first step on an often long and difficult road towards actually
achieving government office. Just as the political system will have an
impact on the decision to aim for government office, so it will have an
impact on the process of getting into government.
In a majoritarian democracy the process of acquiring government
office is usually exceedingly simple, as suggested by Sartori above: typi-
cally, a party simply has to win a majority of the seats and then form a
single-party majority government. While simple, it is obviously also easier
said than done, and usually an option open only to the two main parties
in the typical two-party system of a majoritarian democracy.
However, in a great many democracies, such overall single-party
majorities are a rarely or never occurring event. This is particularly the
case in the context of increased electoral volatility and fragmentation
experienced in virtually all democratic states since about the 1960s (De
Winter and Dumont, 2006, p. 175). In the light of the relative rarity of
overall single-party majorities in many democracies, coalition building
became of crucial importance (Warwick and Druckman, 2006, p. 635).
The question then is which parties get to join a given coalition.
Understanding coalition formation is an area of significant and ongoing
debate. Much effort has been invested in trying to come up with models
that would predict what kind of coalition would be formed and which
parties would be included after a given election result. These attempts at
explaining and predicting coalition formations have become increasingly
sophisticated both theoretically and methodologically, but have also met
with widespread failure in actually explaining real world coalition forma-
tion (De Winter and Dumont, 2006, pp. 177, 184). However, the ability
to predict possible coalitions would be a very powerful tool in analysing
the politics of a country. Being able to predict the most likely governing
parties after an election would have significant value to both those
involved in politics and those writing about it in either academia or the
press. It is, therefore, an important debate in party politics, and thus

deserves attention.
Many attempts at explaining coalition formation have been highly
influenced by game theory (De Winter and Dumont, 2006; Riker, 1962;
Von Neumann and Morgenstern, 1953). The theoretical basis of much of
this literature has been that party leaders act as rational individuals
seeking to maximise their utility. Taking that as a foundation it is hoped
that it will be possible to predict the kinds of decisions party leaders will
make when negotiating coalition agreements. This game theoretical
approach has over the years produced a large number of suggestions for
what ‘rational’ coalitions will look like. Figure 9.2 shows a hypothetical
election result in a 100 seat parliament and the possible coalition combi-
nations for each proposition.
The most basic proposition is the need for a ‘winning’ coalition. This
proposition is based on the idea that only a majority government will
survive. No rational office-seeking politician would allow themselves to
be excluded from government by a minority (De Winter and Dumont,
2006, p. 176). In other words, any combination of parties that result in a
majority would be a rational possibility. This proposition has a number
of problems when encountering the real world. First, it results in a very
A B C D E
8 21 26 12 33
Left Right
Proposition Predicted ‘rational’ coalitions
Winning ABCDE; ABCD; ABC; ABE; ACE; ACDE; ADE; BCD; BCE;
BCDE; BDE; BE; CE.
Minimal winning coalition ABC; ADE; BCD; BE; CE
Minimum seats ADE
Minimum parties BE; CE
Minimal range ABC; BCD; CDE; CE
Minimal connected winning ABC; BCD; CDE
Median legislator C; BC; CD; AC; CE; ABC; ABCDE; ABCD; ACE; ACDE;
BCD; BCDE
Figure 9.2 Coalitions in a minority situation for hypothetical 100-seat legislature
Source: Modified from De Winter and Dumont (2006, p. 176).

Government 171
large number of possibilities as illustrated by Figure 9.2. Further, it is

based on the erroneous assumption that all parties are government
seeking, which would exclude the possibility of a minority government
being supported by other parties. As we saw above not all parties are
government seeking. Some parties are happy to stay in opposition and let
a minority government survive in return for policy concessions. Indeed, a
significant proportion of governments are minority governments
(Woldendorp, Keman and Budge, 2000, p. 86). Finally, the proposition is
‘policy blind’ (De Winter and Dumont, 2006, p. 176) in that it ignores
the fact that parties from opposite ends of the ideological spectrum are
unlikely to want to work together.
The problems associated with the very simple ‘winning’ proposition
led to the idea of a ‘minimal winning coalition’. This proposition is based
on the idea that as the spoils of office are limited (such as the number of
ministerial portfolios) parties will want to avoid any surplus members
that are not needed in terms of reaching a majority (Von Neumann and
Morgenstern, 1953; Riker, 1962). Again only majority coalitions are
assumed to be able to survive. As with the ‘winning’ proposition there are
a fairly large number of possible ‘rational’ options which limits the pre-
dictive value of the model. It repeats the erroneous ‘only majority govern-
ments’ and the policy blindness of the ‘winning’ proposition.
A development of the ‘minimal winning’ proposition is the ‘minimum
seats’ proposition. This proposition assumes that if a formateur (the
party leader tasked with trying to put a coalition together) has a choice
between two parties to reach a majority (which is again assumed to be
necessary) he/she will choose to work with the smaller one (Riker, 1962).
This idea derives from the so-called Gamson’s law (see below) which stip-
ulates that a party will be able to demand a share of portfolios equal to
the share of seats they bring to the coalition. Hence, a formateur will nat-
urally go for the ‘cheapest’ coalition partner. Of all the predictions pro-
posed by the various models this is the one that has produced only one
prediction based on the hypothetical election result. However, as with the
previous propositions it is policy blind and the predicted ‘rational’ coali-
tion contains both the far right and the far left, not a likely scenario in
real world coalition building.
A further development of the ‘minimum’ approach is the ‘minimum
parties’ proposition. This proposition stems from the idea that negotia-
tions will go smoother the fewer actors are involved in negotiations.
Hence, a smaller number of parties will make for easier negotiations than
a larger number of parties. This proposition and the related minimum
seats proposition have reduced the number of predictions. However,
empirical research has also shown that this reduction in the number of
predicted coalitions was also accompanied by a reduction in the ‘success
rate in predicting the actual governments’ (De Winter and Dumont,

2006, p. 177; see also Browne, 1973).
The next proposition is also a ‘minimal’ approach, but this time one
that takes into account ideology. The ‘minimal range’ proposition pro-
poses that parties forming coalitions will want to achieve the smallest ide-
ological range of parties (even if they are not necessarily next to one
another). The need for an overall majority is still assumed. While the
inclusion of ideology is an improvement, the proposition still assumes the
need for a majority. In addition, as we saw in Chapter 4, ideology is not a
straightforward issue. Two parties may sit next to each other on one ide-
ological cleavage, but far apart on another. Then there is the problem of
areas of the world where traditional Western European notions of ide-
ology do not apply. It would be difficult to apply minimal range to, say,
the African context where issues not directly related to traditional
notions of ideology separate political parties.
A development on ‘minimal range’ is the ‘minimal connected winning’
proposition. It adds the extra requirement that there will be no gaps
between the parties (Axelrod, 1970; De Winter and Dumond, 2006). This
may occasionally require a surplus party which is not needed to reach an
overall majority, but is required to plug a gap between two other parties.
This proposition makes the issue of ideology more acute. While ‘con-
nected’ parties may make sense on a one-dimensional ideological spec-
trum, it is very difficult to operationalise in a multi-dimensional
ideological space. While two parties may be close on one dimension, they
may very well differ on the other as argued earlier.
Another approach that takes a step away from the ‘minimum’ idea and
its assumption that only majority governments will survive is the ‘median
legislator’ proposition. This proposition works with the idea that the
median party can be a ‘policy dictator’ (see De Winter and Dumont,
2006, p. 177; and Laver and Schofield, 1990). The argument is that the
party in the ideological mid-point can never be faced with a hostile
majority on its right or its left. Neither the parties on the right-of-centre
nor on the left-of-centre have enough seats for an overall majority
without the party in the middle. Exactly because the centre party will
never face a majority against it, assuming that the left-of-centre and the
right-of-centre will find it impossible to co-operate, it could potentially
form a viable minority coalition. While this proposition does not neces-
sarily help very much in predicting a likely coalition it does help explain
the existence of minority governments. Unless the parties to the right and
to the left of a minority government containing the median legislator can
agree to defeat the government, which considering their inevitable ideo-
logical differences will be difficult, a minority government can survive
(Laver and Schofield, 1990). Indeed, empirical research has shown that
Government 173
the majority of coalitions contain the median party (De Winter and
Dumont, 2006, p. 185n3).
Factors impacting on coalition formation
While all these propositions have resulted in numerous testable predic-

tions and have become increasingly theoretically and methodologically
sophisticated, they have also somewhat failed in actually predicting real
world coalitions with any great success. De Winter and Dumont write
that ‘the empirical results remain modest’ (2006, p. 177) and that models
‘aggregating current knowledge still do not manage to predict correctly
more than half of the coalitions actually formed in the real world’
(p. 184).
This should perhaps not be too surprising. Coalition formations are
affected by numerous factors that are both difficult to formalise in
general models and are also sometimes unique to a particular country,
party, time or set of individuals. The problem of dealing with ideology
and especially multi-dimensional ideological spaces has already been
touched upon in Chapter 4. In addition a few other major problems are
worth mentioning.
The first issue is the impact of institutions – in the sense of regular pat-
terns of behaviour and rules and regulations of the game rather than
merely ‘bronze plaque on the door’ institutions – on coalition building.
Most countries will have more or less formalised rules for which party
leader gets to take on the role as formateur first. Obviously, the choice of
who gets to try to build a coalition first will be an important factor in
explaining which parties end up in a coalition (see, for example,
Diermeier and Merlo, 2004, p. 795). In some countries it tends to be the
leader of the largest party who is given the opportunity first (Diermeier
and Merlo, 2004, p. 795). In others it is the incumbent prime minister
who is given the chance.
Another issue is the coalition potential of a party – or the lack thereof
(see Chapter 2). It has been argued that the coalition potential of a party
should be broken down into two elements: government relevance and
government potential (Bolleyer, 2008, p. 25). Government relevance is
influenced by whether a party is ‘needed’ for a coalition in terms of its
seat share and its ‘programmatic compatibility’ with other parties. This
element is covered by several of the propositions in Figure 9.2. However,
government potential in this context is far more subtle and is not always
something a party can control on its own. Government potential in this
sense is assigned to a party by the other parties in a system. It is essen-
tially about whether others are willing to work with a given party or not.
Some parties may be regarded as being ‘unfit to govern’, often for

ideological reasons (Mattila and Raunio, 2004, p. 265). Examples
include the old Italian communist party which, despite significant elec-
toral success was kept away from government power by the other parties
(Newell, 2010). The former Danish Prime Minister, Poul Nyrup
Rasmussen, once described the far right Danish People’s Party as being
‘not house trained’ (see http://politiken.dk/politik/ECE1112637/roede-
vaelgere-enige-med-nyrup-df-er-ikke-stuerent). The idea of a cordon sani-
taire has been used to describe the process of centre-left and centre-right
parties trying to exclude radical parties on their flanks from influence
(Coffe, 2008, p. 181). Olsen (2010) describes how the Norwegian
Socialist Party had to go through a number of internal changes before it
was ready to consider joining a social democratic led coalition govern-
ment. However, just as importantly both the Social Democrats and the
other potential coalition partner, the Centre Party, had to go through a
fairly long process before they were willing to cooperate with the social-
ists. Finally, Christensen (2010) argues that the main barrier to the coali-
tion participation for the Danish Socialist People’s Party was the
members of the centrist Social Liberals who were not keen on letting the
Socialist People’s Party near government responsibility. Hence, the lack of
government potential will frequently throw out coalition options which
the models would regard as ‘rational’. Government potential is very diffi-
cult to include in abstract general models as it varies massively from
country to country and over time.
A final problem, related to the above issue of government potential is
the very human trait of personal likes and dislikes. Coalition bargaining
is heavily influenced by the personal interaction between individual
actors, something that abstract and general models will find very difficult
to deal with. This impact of the personalities of party leaders can be illus-
trated by the failure of the Danish Social Democrats to form a coalition
government after the 1990 election. Despite being the biggest party and
achieving its best result in decades, the then leader of the Social
Democrats, Svend Auken, was unable to form a coalition because the
leader of their traditional partners, the Social Liberals, did not trust
Auken and refused to work with him (Buksti, 2005). Auken was subse-
quently ousted by Poul Nyrup Rasmussen, who then went on to form a
coalition with the Social Liberals in 1993. No general model would be
able to incorporate such interpersonal matters, which is part of the expla-
nation of the limited success of abstract models in coping with real-world
coalition formation.
An additional issue is the link between gaining government power and
the actual election result. As Sartori pointed out (above) winning in a
majoritarian democracy generally means winning an overall majority of
Government 175
the seats and then acquiring exclusive control over the levers of govern-
ment. However, when it comes to minority and coalition governments the
link between electoral success and acquiring government power is much
less clear. Indeed, it has been found that electoral success is not a very
good predictor of who gets into government office (Mattila and Raunio,
2005, pp. 266–7). Gaining seats in an election does in fact not make it
more likely that a party will be able to enter government – quite the
opposite. In a study of 18 democracies between 1948 and 1979 Rose and
Mackie (1983) discovered that about 65 per cent of governing parties
had lost votes in the previous election. This was backed up by a study
conducted by Muller and Strøm (2000) who found that between 1945
and 2000 only about 36 per cent of governing parties had gained votes at
the most recent election. Finally, Budge and Keman (1990) showed that,
of the parties with the greatest electoral gains in an election, less than half
were subsequently able to join the governing coalition. Hence, being the
greatest ‘winner’ in electoral terms is not a good predictor of who would
be acquiring government office. This should not necessarily be a surprise
given what has been noted earlier about government office not neces-
sarily being a party’s primary goal and issues surrounding ‘government
potential’. The one area where gaining seats did seem to have an impact
was in becoming the formateur. It has been found that gaining seats in an
election had a positive impact on a party’s leader’s chances of becoming
formateur (Warwick, 1999).
What all this shows is that getting into government is not a straightfor-
ward process, and is dependent on much more than merely gaining seats
and making up the numbers for a majority.
Portfolio allocation
Once a group of parties are in a position to form a coalition the next step
will be to determine who get what posts in the new government – that is,
portfolio allocation. Clearly, the main point of getting into government is
getting to control government ministries – especially those covering issues
close to a party’s ideological heart. Exactly which positions go to which
parties will be a matter of negotiation and party preferences in each
country and at each election. That part is, therefore, not easy to generalise
about. What is possible to generalise about is how many ministries each
party gets to control – or in other words the share of portfolios each party
will get. One of the most important developments in this area is what has
been referred to as ‘Gamson’s law’ (Frecette, Kagel and Morelli, 2005).
Gamson wrote that: ‘any participant will expect others to demand
from a coalition a share of the payoffs proportional to the amount of
resources which they contribute to a coalition’ (Gamson, 1961, p. 376).

That is, if a party provides one-third of the seats in a coalition they will
expect to get one-third of government portfolios. The reason why this has
been labelled a ‘law’ is that this does appear to be the case with almost
complete accuracy (Browne and Franklin, 1973). That this should be the
case has been somewhat of a puzzle (Warwick and Druckman, 2006).
One of the reasons why the accuracy of Gamson’s Law is so surprising is
that it could be argued that a ‘smaller party that is pivotal to a host of
coalitions’ (Warwick and Druckman, 2006, p. 660) would be able to
demand a greater than proportional share of the portfolios in return for
supporting one possible coalition over another. In short, why should a
king maker party not demand a price for choosing one king over
another? However, this does not appear to be the case.
One possible answer to the Gamson’s law puzzle is that parties are
seeing the longer term and not just focusing on the immediate coalition
negotiations. If a king maker party demands more than its ‘fair share’ of
portfolios, measured by what share of the seats it brings to the coalition,
it may have an impact on a party’s chances next time round where it may
no longer be in a king maker position. ‘Unfair’ demands could make a
party a less desirable coalition partner in the long run. In other words,
negotiating hard after one election may reduce a party’s government
potential after the next election.
Whatever is the case, as it stands Gamson’s law ‘is an empirical rela-
tionship still deserving of its law-like status – but in acute need of a firm
theoretical foundation’ (Warwick and Druckman, 2006, p. 660) – that is,
we know it happens, but not why.
It is worth mentioning that while size may not make a big difference to
a party’s chances of getting into government, it does make a difference to
its chances of getting the post of chief executive (Prime Minister). De
Winter and Dumont (2006, p. 182) found that in about 80 per cent of
cases the largest party in a coalition will get the chief executive portfolio.
Conclusion
It should by now be clear that getting into government is a very compli-

cated process for a party. It first has to want government power, and once
in that position it has to overcome a number of hurdles to achieve that
goal. In addition, winning/losing in terms of seat shares may not be a
good predictor of a party’s chances of getting into government. However,
party size does make a difference to how much it will get out of being in
government. A greater share of the seats means greater rewards in the
shape of a greater share of portfolios. Hence even in a consensus democ-
Government 177
racy, vote and seat shares still matter, but in very different ways to a
majoritarian democracy. In short, a party’s road to government can be
long and convoluted and will be affected by numerous factors, many
beyond its control.
Chapter10
The Internationalisation of Party

Politics
This book has so far focused mainly on party politics as it has played out
on the national stage. However, to complete the picture it is worth
looking beyond the nation state. Indeed, political parties have long
endeavoured to engage with politics beyond the national. This is not just
in the sense that, to be credible all parties need to have something to say
about the foreign policy of the state in which they are operating. There is
more to it than that. Parties have in various ways tried to go beyond the
national, not just in terms of their policies, but also in how they work as
organisations. The internationalisation of party politics has a long
history. The creation of various international affiliations between parties
goes back to the 1800s. Further, in 1983 Goldman wrote (perhaps some-
what prematurely as we shall see) that there were tendencies in evidence
which ‘taken together, appear to predict the emergence of a transnational
or world party system’ (p. ix). This chapter will first look at the earliest
and longest running attempt at internationalising party politics: party
internationals. It will then consider Europarties in the context of the
European Union. Having looked at these two examples of organisational
internationalisation it will look at how parties have responded to the
issue of globalisation. Finally, it will look at the internationalisation of
campaign consultancy.
Party Internationals
There have been many attempts at building forms of permanent interna-

tional co-operation between parties. This has been particularly popular
on the left-wing where many socialist parties and, especially, communist
parties have seen themselves as part of a world-wide movement.
However, other party families have also worked to build international
structures for the pursuit of common ideals.
178
The Internationalisation of Party Politics 179
The First, Second, Fourth and Socialist Internationals
The left-wing in party politics has a long, and largely unsuccessful,

history of trying to build international cooperation in the name of the
working class solidarity. The First International (or the International
Workingmen’s Association) was created in London in 1864 and lasted
until 1876. It worked as a forum for various left-wing groups and was
supposed to lead to a common direction for left-wing parties. However, it
collapsed under the weight of internal disagreements over policy and
strategy. Karl Marx played a significant role in the First International and
was very much involved in the disagreements which eventually led to its
demise.
The idea of international socialist cooperation was revived in organisa-
tional form in 1889 with The Second International. The Second
International is noted for making 1 May the International Workers’ Day,
starting in 1889, to commemorate the so-called 1886 Haymarket
Massacre in Chicago. It also instigated 8 March as International
Women’s Day, first observed in 1910. However, like the First
International it was marked by disagreement between the national
member organisations and it finally collapsed in 1916 as socialist, social
democratic and labour parties overwhelmingly failed to oppose national
involvement in the First World War (fought, obviously, overwhelmingly
by working class soldiers on all sides). The various left-wing parties
largely supported the war effort of their respective governments.
The Third International (also known as the Communist International
or Comintern) will be dealt with separately below.
In 1938 Leon Trotsky founded the Fourth International as an alterna-
tive to the Stalinist Comintern. The Fourth International never received
much support from the rest of the left-wing and remained largely iso-
lated. The Fourth International suffered from increasing divisions and
splits starting just a few years after its creation. By the 1960s the splits
had become terminal with several independent and mutually warring
organisations claiming the Fourth International’s mantle – some of which
continue to lead a ghostly half-life on the fringes of political life even
today.
If there is one lesson to be learnt from the various left-of-centre
Internationals it is the near impossibility of achieving close cooperation
between national parties in the absence of strong incentives to overcome
disagreements. Such incentives might include access to legislative power
or funds as is the case in the European parliament which, as we shall see,
has led to increasingly institutionalised forms of transnational party for-
mations. In the absence of such incentives international cooperation on
anything more than the most superficial level has proven well-nigh
impossible, certainly for the left-of-centre which is in any case prone to

disagreements and splits.
However, the idea of international left-wing cooperation has not died
and is currently represented by the Socialist International created in 1951
(see http://www.socialistinternational.org). The Socialist International
currently serves as an umbrella organisation for 163 centre left socialist,
labour and social democratic parties. The Socialist International is to a
certain extent built like a national party which typically (see Chapter 1)
has a ‘congress’ as its highest authority to which ‘local’ fees paying
‘branches’, that is, national parties, can send delegates. In the Socialist
International’s statutes the Congress is described as ‘the supreme body’.
The Congress elects the president of the Socialist International as well as
the Vice-Presidents, the Secretary General, and is the only body that can
change the organisation’s statutes. However, apart from those very spe-
cific organisational powers there is not much to indicate what ‘supreme
body’ actually means. There is certainly no indication that the Socialist
International has any power, or indeed, desire to impose any policy or
ideological line on the national member parties. Nevertheless, it no doubt
serves as a useful forum for communication between the member parties.
The Socialist International also has ‘consultative status’ in the United
Nations. However, there is no claim that the Socialist International is
aiming to be a transnational party, but rather an umbrella group for net-
working and sharing experience and developing more or less specific
ideas for promoting social democracy.
The Liberal International
The Liberal International was created in 1947 and serves much the same
purposes as the Socialist International. It seems to share the Socialist
International’s relatively limited practical impact on the day-to-day lives
of its membership parties. It functions mostly as a networking facility
with an additional think-tank function – that is, as with the Socialist
International, but from a liberal perspective, it has the purpose of devel-
oping and promoting a certain set of standpoints. It is also an interesting
example of the standard liberal reaction to globalisation (see Chapter 4):
‘Liberals believe that the poverty of large parts of the world can be allevi-
ated through freedom to travel and to trade and to this end are com-
mitted to the further opening of “western” markets for products from the
developing world.’ (http://www.liberal-international.org/editorial.asp?ia_
id=508). That is, a theoretical commitment to the view that globalization
is a good thing and greater openness and integration are beneficial for all
nations.
The Liberal International seems even more loosely organized than the
Socialist International. While the Socialist International has a set of
statutes available on their website, vague and unarticulated as they are,
there is no such equivalent for the Liberal International. So, the Liberal
International has an important symbolic role, but it shares the limited
practical role of the Socialist International.
International Democratic Union
This association of conservative and Christian democrat parties com-

pletes the set of contemporary internationals. It was founded in 1983 by
several leading right-wing leaders including Margaret Thatcher, Helmut
Kohl and George Bush Sr. Its website (http://www.idu.org/history.aspx)
has probably the best description of the most important purpose that
these party internationals serve:
the IDU provides a forum in which Parties holding similar beliefs can
come together and exchange views on matters of policy and organisa-
tional interest, so that they can learn from each other, act together,
establish contacts and speak with one strong voice to promote democ-
racy and centre-right policies around the globe.
Although this is taken from the International Democratic Union’s website

it could serve as a good description of what the other international are
trying to achieve. The International Democratic Union does have a set of
rules (Bylaws) available on their website (http://www.idu.org/
structure.aspx), suggesting at least some form of formal structure. What
is important to note is that at the ‘Party leaders conference’ decisions
have to be taken unanimously. What is referred to as ‘recommendations’
can be passed by a three-quarters majority, but even then parties which
have expressed a ‘dissenting opinion’ are not covered by such ‘recommen-
dations’. This is indicative of the limited organisational ambitions of the
International Democratic Union, a feature it shares with the other con-
temporary internationals. These are not ‘world parties’ in the making,
but loose networks of like-minded organisations geared towards the
exchanging of ideas and experience.
Communist International (Comintern)
If the organisational ambitions of the main centre-left and centre-right

internationals have been and remain somewhat limited, the original
purpose of The Communist International was the polar opposite.

Comintern was established in 1919 and dissolved on the orders of Stalin
in 1943 in an effort to placate the concerns of his Western allies during a
critical phase of the Second World War. The idea behind the Comintern
was nothing less than the creation of a world communist party.
Comintern had it headquarters in Moscow and each of the national
parties linked to it were regarded and encouraged to behave as ‘local’
branches or sections. This was the first and only attempt at creating a
genuine world party (Thorpe, 1998, p. 637), and is thus worthy of atten-
tion. It illustrates the challenges involved in such a project, and is, there-
fore, important to our understanding of why no successful world party
exists. The extent to which Comintern was actually able to operate as a
single party – that is, its ability to truly control the behaviour of the
national member parties – has been the subject of a longstanding, and it
seems still ongoing debate. Three broad strands exist in this debate:
1 The so called ‘orthodox’ standpoint (Thorpe, 1998, pp. 637–8;

Trapeznik, 2009, p. 125) argues that the national parties were little
more than the willing and obedient tools of Moscow (see Draper,
1957; Pelling, 1958). This stance would suggest that Comintern was a
successful world party. It was a view particularly popular with anti-
communists during the Cold War (Thorpe, 1998, p. 637). Obviously
in the charged atmosphere of the Cold War if one could paint Western
communists as tools of Moscow that would undermine the legitimacy
of these parties.
2 The orthodox view was challenged by ‘revisionists’ (Thorpe, 1998;
Trapeznik, 2009). The revisionist position was the polar opposite of
the orthodox view and argued that most members of Communist
parties outside the Soviet Union were largely divorced from the machi-
nations between the leadership of Comintern and national communist
leaders (Macintyre, 1980). What mattered was not what went on
between national party leaders and Comintern, but what communists
did at the local level, far from both national leaders and Moscow.
3 Challenging both position are the ‘post-revisionists’ (Thorpe, 1998, p.
638; Trapeznik, 2009, p. 126). In essence the post-revisionist view is
placed somewhere in between the orthodox and revisionist view.
Authors in the post-revisionist vein (such as Isserman, 1982) argue
that Comintern’s influence is often exaggerated, but at the same time,
it would be wrong to assume that Comintern did not have an impact
on the day-to-day workings of Western communists.
Sadly it appears that most scholarship on Comintern has been, and to a

certain extent still is, dominated by this division with its roots in the Cold
War. Hence, any wider party-politics lessons from Comintern tend to be

lost in point scoring between left-wing and right-wing scholars. Overall,
it is probably fair to say that Comintern was a failure. There were no suc-
cessful Communist takeovers during its lifetime. Directives from Moscow
frequently put national communist parties in positions where they were
pushing enormously unsuccessful policy positions – most notably ‘class
against class’ from 1928 until 1935. The class against class policy, part of
the so called Third Period, held that anyone not a communist was an
enemy of the working class, including social democrats and labour
parties which were labelled ‘social fascists’. National communist parties
were, therefore, directed to attack centre-left parties as vigorously as
right-wing parties. Some have argued that this led the German commu-
nists to ignore the rise of the Nazi party and partially aided the collapse
of the Weimar Republic and the rise of Hitler. In addition, a number of
Western communists living in the USSR were executed during Stalin’s
purges in the 1930s. Hence, Comintern, as the only real attempt at a true
world party, not only failed in its stated purpose of promoting a world
communist revolution, it can be said to have actively undermined this
goal.
However, while a lot of the work on Comintern may to an extent be
undermined by ideological blinkers it is still worth exploring some of the
reasons behind the failure of Comintern, as it has lessons for other efforts
at building transnational parties.
Essentially, Comintern failed because of two major problems which
will face any attempt at building transnational parties: national distinc-
tiveness and maintaining discipline around a central policy line.
As hinted at above, the attempts at imposing a central policy line from
Moscow at times had the effect of damaging the efforts of local parties.
Class against class led to the isolation of communists from their potential
allies in other left-wing parties and thus managed to split the left, some-
thing which could only help the centre-right (Beckett, 1995; Jacobsen,
1996). The policy of attacking all non-communists certainly made sense
in a communist one-party state such as the Soviet Union, but in demo-
cratic countries it was counter-productive. When the policy was suddenly
reversed in 1935 after its abject failure, most catastrophically in
Germany, the damage caused by an inappropriate policy was exasperated
by the loss of credibility when national parties were instructed to turn
180 degrees – from attacking ‘social fascists’ to calling for ‘united front’
cooperation with other left-of-centre parties. Hence, Comintern illus-
trated with painful clarity the damage that a lack of consideration of
local factors can cause. Policies dreamed up in Moscow failed miserably
in states which were culturally and politically radically different from the
Soviet Union.
The second problem was how to ensure that national parties followed
the line laid down by the Comintern leadership. Thorpe (1998) argues
that Comintern found it particularly difficult to keep national parties in
democratic countries under control. As long as national communist
leaders were beyond the physical grasp of the Soviet regime, Comintern
had to rely on other means of persuasion. Money was certainly a major
factor, particularly where the local party’s membership was relatively
modest. One good example of this was the Communist Party of Great
Britain (CPGB). In 1927 the CPGB’s membership was less than 8,000.
That year it received from Comintern £21,000 for running the organisa-
tion and a further £18,000 to support a newspaper project. In comparison
the Labour Party had a membership, direct and indirect (that is, trade
union members), of 3,293,615 and an income of £55,708. In short with a
tiny fraction of the Labour Party’s membership the CPGB had finances
equal to about 70 per cent of Labour’s. The threat to withdraw that sum
from the CPGB was a major leverage in the relationship between Moscow
and the British ‘branch’. However, in, for example, France where the com-
munist party membership was vastly bigger and local resources therefore
much greater, such a threat would be much less severe.
Ultimately, a party could be excluded from the Communist
International the way an individual member or a local branch can be
excluded from a national party. For an individual member or even a
whole branch in a national context such exclusion can be devastating in
terms of electoral success. There is virtually no place in electoral politics
for non-party independents and the loss of party affiliation will almost
invariably mean the loss of any electoral office previously held once the
next election comes around. However, for a national section of a ‘world
party’ that is much less relevant. Indeed, as was shown above, anti-
communists used Comintern connections as a rod with which to beat
Communist parties. Hence, apart from the loss of Comintern income,
exclusion was probably not a very effective threat. Indeed, it could be
argued that the European far left did much better outside the Comintern
than inside it as illustrated by the notable electoral successes of the so-
called Eurocommunists in the post-war period. Further, while Comintern
could offer little more than financial support to its ‘members’, the
national parties, at least where they were electorally successful, could
offer Comintern influence on the politics of non-Communist countries. It
might, therefore, be argued that Comintern needed the national parties
more than the national parties needed Comintern.
In short, this first experiment in building a world party was largely
unsuccessful, but provided it is given more non-partisan attention might
be able to yield useful insights into the challenges of managing transna-
tional parties.
Europarties
That brings us to the most successful example of cross-national party

cooperation in existence so far: Europarties. The Europarties represent
the only example of truly transnational party organisations in the world
today. Since Comintern, they are the only party organisations that go
beyond the political context of a single country. Since 1979, there has
been a gradual, if incomplete, emergence of recognisable political party
organisations within the European Union.
That this should be so is hardly surprising. Once direct elections were
introduced in 1979, the European Parliament became the only suprana-
tional directly elected assembly in the history of democratic politics. This
meant that for political parties, defined as organisations fielding candi-
dates for publically elected assemblies, transnationalism now made sense,
especially once the EU made party funding available.
The Development of Europarties
This does not mean that their development has been straightforward or
that there are now fully fledged EU wide political parties the way we see
them at a national level. The process is still very much ongoing and still
has a considerable way to go. The emergence of Europarties can be
traced back to at least the 1970s (see Table 10.1). A notable limiting
factor in the development of Europarties was that no European treaty
made mention of parties until the Maastricht Treaty of 1993. This ought
not in and of itself be an insurmountable factor. Many national constitu-
tions do not mention parties at all, although parties do then tend to be
recognised in ‘normal’ legislation. What is now the EU did not recognise
the existence of Europarties until 1993, and while the recognition in
1993 was rather vague and aspirational it did set in motion a series of
changes that brought into existence an embryonic Europe-wide party
system.
The Maastricht Treaty contains the so-called party article (Hertner,
2011, p. 326; Johansson and Raunio, 2005, p. 515), officially called
Article 138a, which reads: ‘Political parties at the European level are
important as a factor for integration within the Union. They contribute
to forming a European awareness and to expressing the political will of
the citizens of the Union’ (see http://eur-lex.europa.eu/en/treaties/dat/
11992M/htm/11992M.html#0001000001).
While this was a major advance from previous treaties, where no
mention of political parties is made at all, it should also be immediately
evident that Article 138a is almost entirely vacuous and of very little
Table 10.1 Brief description of existing Europarties

Name Year Comments Website
Created
European 1976 The party is made up of the main centre-right conservative and Christian http://www.epp.eu/
People’s democratic parties in the EU member states. It has a strongly federal slant. As
Party of 2011 it has 75 full, associate or observer members across 39 EU and non-EU
countries.
Party of 1974 Represents the centre-left social democratic/labour tradition. PES allows members http://www.pes.org/
European of national parties to sign up as something akin to a individual membership PES,
Socialists labelled ‘PES Activists’ (http://www.pes.org/en/pes-activists). As of 2011 PES had
28 full members, 10 associate members, and 5 observer members).
European 1976 Is made up of centre-right liberal parties. The party consists of 55 national parties http://www.eldr.eu/en/index.php
Liberal from EU and non-EU countries. As with the PES individuals can join the party as
Democrats and an ‘associate member’ (http://www.eldr.eu/associate).
reform Party
European 1984 Is made up of 34 national Green parties. According to its website it is possible http://europeangreens.eu/
Green to join the European Green Party, although it is not entirely clear how.
Party
Party of the 2004 The Party of the European Left essentially consists of parties situated to the left http://www.european-left.org
European of the main centre-left social democratic and labour parties. It contains much of
Left what is left of European communism (e.g., the French Communist Party) as well
as some East European parties with their roots in the former Soviet Bloc
communist parties. It has 27 member parties and also allows for individual
party membership (http://www.european-left.org/english/about_the_el/
individual_members/)
European 2004 Is a pro-EU party situated on the centre-right. Its MEP co-operate with the http://www.pde-edp.net/main/_
Democratic MEPs from the Liberal Democrats and Reform Party, but the two Europarties pde/index.jsp
Party remain separate entities. It is a relatively small party with 12 national party
members (2011).
European 1994 The European Free Alliance is made up of 40 regionalist and separatist parties, http://www.e-f-a.org/home/
Free very much representing ‘the periphery’ in the ‘national revolution’ political
Alliance cleavage (see Chapter [ideology]).
EU 2004 This is a very small Euro-sceptic (or ‘Euro-realist’ according to their website) http://www.eudemocrats.org/

Democrats group. Several of its national members are unaffiliated to any national party. eud/index.php
Alliance of 2009 Represents Euro-sceptic conservative parties, most notably the British http://aecr.eu/
European Conservative Party. It is made up mainly of East European parties with a very
Conservatives right wing and highly euro-sceptic profile. It is one of the smallest groups with
and Reformists only 11 national members.
European 2009 Is a Christian right wing and Euro-sceptic group to the right of the European http://www.ecpm.info/en/
Christian People’s Party. Unusually it is not associated with any of the groups in the
Political European parliament and according to their website seems to be made up
Movement largely of NGO (including one from the USA). Indeed, it does not even appear
to have any MEPs.
practical value. It did however open the door for further developments,
by constituting what Johansson and Raunio (2005) call an ‘incomplete
contract’. Europarties were able to capitalise on the opening provided by
the party article in the Maastricht Treaty to develop an increasingly sig-
nificant role in the European Union system. Over the next 10 years after
the Maastricht Treaty the existing Europarties successfully argued for a
more prominent role. A major result of this was that the Nice Treaty,
signed in 2001 and coming into force in 2003, made provisions for
European funding of Europarties via the European Parliament’s budget
(see Siglas et al., 2010, p. 6). Because funding was now going to be made
available to Europarties it was necessary to define what a Europarty actu-
ally looks like in practice. According to a regulation issued in 2003 a
Europarty was defined as follows:
A political party at European level shall satisfy the following condi-

tions:
a) it must have legal personality in the Member State in which its seat
is located;
b) it must be represented, in at least one quarter of Member States, by
Members of the European Parliament or in the national
Parliaments or regional Parliaments or in the regional assemblies,
or it must have received, in at least one quarter of the Member
States, at least three per cent of the votes cast in each of those
Member States at the most recent European Parliament elections;
c) it must observe, in particular in its programme and in its activities,
the principles on which the European Union is founded, namely
the principles of liberty, democracy, respect for human rights and
fundamental freedoms, and the rule of law;
d) it must have participated in elections to the European Parliament,
or have expressed the intention to do so. (see http://eur-lex.europa.
eu/LexUriServ/LexUriServ.do?uri=CELEX:32003R2004:EN:NOT)
At first European funding could not be used for election campaigning,

but that was changed in time for the 2009 European Parliament elections
(Hertner, 2011, p. 327). The availability of funding had a major impact
on the development of Europarties, formalising those already in existence
and encouraging the creation of new ones. By the 2009 elections there
were a total of 10 Europarties (see Table 10.1 and Gagatek, 2009, p. 16).
It is important to note that Europarties are not the same as the formal
groups in the European parliament even though they are linked (see
Figure 10.1 and Gagatek, 2009, p. 13; and Hertner, 2011, p. 323). For
example the Party of European Socialists is linked to the ‘Group of the
Poiltical group
Europarty Cooperation in the European
Parliament
Send Send
delegates MEPs
National
political parties
Figure 10.1 Relations between Europarties and European Parliament Groups
Source: Modified from Gagatek (2009, p. 13).
Progressive Alliance of Socialists and Democrats’. Where a Europarty

does have links with a European Parliament group – which tends to be
the norm – the two do work very closely together.
Hence, in the ‘three faces of parties’ framework (see Chapter 1) the
European Parliament group is the party in public office, the national
parties are the party on the ground and the Europarties are the party in
central office. However, this picture is somewhat confused by the fact
that some Europarties have started adding a separate individual member-
ship organisation as mentioned in Table 10.1 – hence some Europarties
have two forms of party on the ground. However, the ‘individual mem-
bership’ organisations are very weakly developed.
Limits to the development of Europarties
It would, therefore, be fair to say that Europarties have made consider-

able strides since the introduction of direct elections to the European
Parliament in 1979. From being little more than very loose associations
of national parties, they have now developed into formal organisations
with legal recognition from the European Union, independent legal ‘per-
sonalities’ in the countries where they are registered and an independent
budget.
There are however, also notable limitations to their institutionalisa-
tion. Table 10.2 outline some of the key features of the parties whose
statutes are available on their websites. As we can see the power struc-
tures of all five parties in the table are not very articulated. In some of the
party statutes it is specified that the Congress is the highest authority, but
there is little indication of what that means in practice. Sanctions in the
Table 10.2 Key organisational features of Europarty statutes
Name Key features of party statutes
Highest decision-making body Sanctions against members Nature of decision making
Party of The congress – little detail of Membership can be suspended by ‘Efforts shall be made to establish broadest
European what that means in practices. the Presidency. Requests for lifting possible measure of agreement’. Administrative
Socialists of suspension can be made to the and organisational matters can be made by a
Presidency and rejections of such simple majority. Political decisions require a
requests appeal to the Congress. qualified majority of 75 per cent. A party is not
Members can be excluded by a bound by a decision made by qualified majority
qualified (75 per cent) majority of if it declares itself unable to do so before the
Congress. vote is taken.
European Not specified Members can be expelled by the Decisions at conference can be made with a
Liberal Council with ²⁄³ majority. simple majority and are binding on all
Democrats members, including those absent and
and Reform dissenting.
Party
European Not specified Membership of a national party Decisions in the Council requires ²⁄³ majority.
Green Party can be decided by ²⁄³ majority in
the Council. Expulsion requires
¾ majority.
European ‘The congress has the most Exclusion of a member party can Congress decisions are taken by a simple
Democratic extended powers for the happen by a ²⁄³ majority of the majority and a binding on all members
Party attainment of the purpose Council. including those who voted against or abstained.
of the association’ Changes to the statutes require ¾ majority.
European ‘The general assembly is the Exclusion possible but process Not specified.
Christian general meeting by law. All not specified.
Political powers in the association not
Movement conferred on the board by law
or in this charter shall vest in

the General Assembly.’
Sources: European People’s Party http://www.epp.eu/; The Party of European Socialists (PES) http://www.pes.org/; Alliance of Liberals and Democrats for Europe
Party http://www.eldr.eu/en/index.php; European Green Party http://europeangreens.eu/; European Left http://www.european-left.org/; European Democratic Party
http://www.pde-edp.net/main/_pde/index.jsp; European Free Alliance http://www.e-f-a.org/home/; EUD http://www.eudemocrats.org/eud/index.php; AECR (Alliance
of European Conservatives and Reformists) http://aecr.eu/;European Christian Political Movement http://www.ecpm.info/en/ (accessed 21 November 2013).
form of a suspension of membership and expulsion exist in all parties.

However, in several of the parties there are requirements for qualified
majorities and in the Party of European Socialists decisions do not apply
to members who have declared before a vote that they cannot abide by
them. Notably the Liberals and the European Democratic Party both
have clauses stating that all members are bound by decisions regardless of
whether they agree with them or not. This makes those two parties the
most institutionalised of the five. However, across the statutes of the five
parties, institutionalisation is relatively low and power structures are left
fairly vague. Hence, judging by their statutes Europarties still fall short of
the institutionalisation of many national parties. Despite the fact that
some Europarties now allow for direct individual membership, rather
than merely indirect via national parties, the important players are over-
whelmingly the leaders of national parties and there is very little direct
contact between Europarties and European voters.
Further, there may be doubts about the extent to which Europarties
can be regarded as ‘proper parties’ based on the definition of a political
party adopted in this book (that is, as organisations fielding candidates).
As of the 2009 European Parliament election none of the parties fielded
candidates under their own name. All candidates in the European elec-
tions in 2009 stood under the banner of national parties, with the excep-
tion of the short-lived and largely unsuccessful Libertas party (Gagatek,
2009, p. 10). In short, European elections are still national elections with
a European outcome (Gagatek, 2009, p. 9). The result obviously affects
the composition of the European Parliament, but it is fought on national
issues and by national parties. One prime example of this is the adoption
of Europarty election manifestos. All the Europarties adopted manifestos
in the lead up to the 2009 European elections. Indeed, some parties went
through quite extensive consultation processes in making their manifesto.
However, while some of these manifestos were adopted in their entirety
by a few national parties or partly incorporated in national manifestos,
they were to a very large extent ignored. So, despite the fact that
Europarties now have their own campaign funds and can therefore
provide election materials to national parties, this is only rarely utilised
by the national parties. Campaigns remain overwhelmingly national
affairs focused on issues related to national rather than EU politics
(Gagatek, 2009; Hertner, 2011; Siglas et al., 2010).
There are many reasons for the limited development of Europarties.
However, two can be singled out as being of particular importance: the
difference between the nature of European Union politics and the nature
of national politics; and the problem of national peculiarities.
It is true that the European Union political system can be said to have
a separation of powers similar to that found to varying degrees at the
national level. So, the European Union has a legislature, an executive and
a judiciary. However, unlike most parliamentary democracies there is
very little link between legislative elections and the formation of the exec-
utive (Day, 2005, p. 4; Gagatek, 2009, pp. 20–1; Lightfoot, 2006, p.
304). The executive office in the European Union, in the form of the
European Commission and the Council of Ministers, is only open to
national parties, not Europarties. The composition of the Commission
and the Council of Ministers is determined by the outcomes of national
elections, not election to the European Parliament. Europarties have very
little impact on the makeup of the Commission and none at all on the
Council of Ministers. European Commissioners are appointed by
national governments; and the Council of Ministers is made up of minis-
ters from national governments. It is true that the European Parliament
has a role in approving commissioners, but that is very far from the influ-
ence that a national parliament has on the composition of a national gov-
ernment. As a result, and in contrast to the national politics of European
states, there is very little in the way of a government/opposition dynamics
in the European Parliament. After a European Parliament election it may
be possible to say that either the left or the right has won or lost based on
how many seats the two wings won. However, ‘winning’ in this sense
does not give the winner the right to put their people on the Commission
or the Council of Ministers. Indeed, the Commission is supposed to be
non-partisan and is very little affected by European Parliament election
outcomes. Hence, unlike elections fought at the national level, European
Elections are not fought on the record of the incumbent executive. That
means that by default European Elections tend to be fought on the basis
of the performance of national governments, and therefore on national
issues, and in the context of national party politics. Europarties, and
indeed European Union issues, are, therefore, often largely ignored in
European Parliament elections. There is then little reason for national
parties to follow the lead of Europarties when it comes to deciding the
themes of the campaign. If Europarties cannot effectively control the
campaign themes of their national ‘branches’ they are never going to be
fully fledged parties.
Added to that is the problem of dealing with many different national
contexts. As was explored in the context of Comintern, the imposition of
a supranational set of policies from some distant centre, which is likely to
be ignorant or uncaring about local circumstances, is unlikely to be elec-
torally beneficial. This problem has affected the formation of Europarty
election manifestos. They have to avoid being too specific in order to not
render themselves irrelevant to local circumstances. However, this has
often meant that Europarty manifestos have been driven by the lowest
common denominator and have ended up being very vague and lacking
in substance. Europarties have struggled to find a balance between being

too specific and too vague and have so far not managed to find a solu-
tion. Other, that is, than leaving policy development to the national
parties, thus yet again illustrating the problem of creating meaningful
supranational parties.
In short, while Europarties have certainly come a long way since 1979,
and have seen an acceleration in their development since the early 1990s
they are still a long way off being anything near as important as national
parties. Indeed, it is perhaps unlikely that they will ever reach that stage.
Having said that, Europarties are still developing and their future trajec-
tory is far from certain.
Parties and globalization
It is clear from the above that party politics has a significant, if also very
under-developed international element. This under-development of the
international aspect of party politics is particularly noteworthy consid-
ering how much the study of so many other areas of politics has been
dominated by issues surrounding globalisation. It would be fair to say
that in most areas of politics the term globalisation has become a major,
if often also overhyped, area of concern and study. In both teaching and
research the idea that politics is becoming ever more interconnected
across national borders has gained a big following and has had a huge
impact on what is researched and taught. Degree courses with ‘interna-
tional’ and/or ‘global’ in the title have been attracting students in ever
greater numbers, which in turn has led many university departments to
focus greater and greater attention on this area in terms of teaching and
new hires.
It is obviously the study of international relations, and related fields,
such as national security and political economy, which have been the
most affected by this trend. However, it has also affected the study of
national politics. The national is increasingly seen as declining in impor-
tance compared to the global. Many civil society organisations, especially
those dealing with civil rights and environmental concerns, have
embraced globalisation in a major way. Political parties, however, have
remained relatively unaffected by this growing infatuation with globalisa-
tion. Both in their activities and in the academic study of them, parties
have remained steadfastly national: ‘Hardly anyone is talking about glob-
alization in the context of political parties. Political parties have, to a
large extent, remained limited to the national spheres’ (Sehm-Patomäki
and Ulvila, 2007, p. 1). This is not to say that parties have remained
completely ignorant of globalization. In terms of their policy, parties have
had to react. Broadly speaking the reaction of political parties to global-

ization can be divided into three main trends:
First, many free-market liberal parties have embraced globalization as a
more or less unreserved good. The free movement of everything – finance,
goods, services and labour – across borders has, at least officially been
welcomed and encouraged in the formal policy stance of free market sup-
porting parties. Many conservative parties have also embraced this trend.
Indeed it was Margaret Thatcher of the British Conservative Party who
popularised the term TINA (There Is No Alternative) in the context of eco-
nomic liberalisation. It is true that most conservative parties have limited
their enthusiasm for globalisation to certain aspects of ‘low’ politics (social
and economic issues) and have resisted it in ‘high’ politics (most issues
related to traditional sovereignty – such as national security and border
control). Likewise, the free movement of labour has been noticeably less
popular among conservative, and indeed many liberal, parties than the free
movement of goods, finance and services. Nevertheless, in policy terms,
parties of the right with strong laissez faire elements have by and large
embraced globalization as both inevitable and beneficial.
Second, the social democratic centre-left parties have, on the whole,
displayed noticeably less enthusiasm for globalization in their policy
stances than the free-market right. However, most social democratic and
labour parties have to a large extent bought into the idea of TINA, and
have approached the issue with the view that if one cannot beat it one
might as well make the most of it. Hence, social democratic parties have
mostly adopted a moderate pro-globalisation stance. This approach has
gone the furthest in the ‘Third way’ stance of Tony Blair and Gerhard
Schroder – even though the latter was somewhat less eager than the
former. Indeed, Blair and Schroder’s attempt at a common approach to
modernising social democracy can be seen as an example of the interna-
tionalisation of party politics. However, the fact that their project never
went very far is illustrative of the limits of this internationalisation.
Third, outright opposition to globalization tends to be limited to the
fringes of most party systems. Most far-right and far-left parties have
taken a very national, even nationalistic, approach, of opposing global-
ization in all its forms – usually by opposing the lowering of taxes and
cutting of social services and the free movement of labour (the far right
fears the dilution of the national spirit by foreigners; the far left fears
foreign, and often un-unionized, workers will undercut wages). This links
back to the discussion in Chapter 4 where we saw that some parties have
taken on the mantle of protectors of those who seem to be losing out
under globalisation.
However, while policies have been affected by globalization, it has had
virtually no impact on parties organisationally. There are no global
parties and few transnational parties, with the partial exception of the
Europarties.
According to some, this failure to react organisationally to globalisa-
tion represents a failure on behalf of parties to move with the times:
‘political parties have for the most part retained a now obsolete statist-
territorialist-nationalist modus operandi’ (Scholte, 2007, p. 12). This is in
sharp contrast with both state institutions and non-party civil society
organisations, both of which have made considerable strides in engaging
with globalised politics. According to some authors this is a deficiency
that is in urgent need of being rectified. What we need, according to
Scholte (2007, p. 13) is ‘more globally oriented political parties’. Scholte
suggest a number of things parties could do to rectify the lack of party
politics engagement with globalisation (2007, pp. 23–30):
• Educate the general public about the importance and impact of glob-
alisation.
• Use government power to improve the democratic elements in global
governance.
• Scrutinise state policies on global governance issues.
• Engage directly with global governance bodies, such as the WTO, the
World Bank and the IMF, the way many NGOs with a global slant
have done.
• Promote greater global equality on the basis that only people with a
certain minimum of resources can engage meaningfully in democratic
processes.
• Move beyond the idea that ‘the people’ to be represented are only the
people of a single territorial state. Instead parties should start to cater
to supraterritorial types of demos.
Leaving aside the issue of the pure practicality of doing these things, there
are two major issues with Scholt’s proposals for a globalisation of party
politics. The first problem is that some of the issues he raises are clearly of
a left-wing slant. It would not be too much to say that a significant section
of the political spectrum has no interest in a more equal redistribution of
global resources. Most parties of liberal, conservative or nationalist per-
suasion would see no benefit in such a project. His remedies, therefore,
speak to a particular normative position and not to ‘parties’ as a group. It
might well be argued that left-wing parties ought to think more about
global democracy and equality and have failed in properly promoting this
agenda by focusing on national rather than global governance. However,
many parties have as their main founding goals the promotion of national
and not global interests. Hence, his criticism of ‘parties’ as a whole are in
fact much more relevant for parties of the left than parties of the right.
Scholte, therefore, neglects to recognise that parties have a plethora of

goals, only some of which may be undermined by parties’ lack of engage-
ment with global governance. Indeed, many party goals would run com-
pletely counter to his remedies. Hence, the problems he identifies and the
remedies he proposes speak to some parties of a particular ideological
makeup. He does not speak of all parties.
The second problem is linked to the very nature of what a political
party is. This is related to a question raised, but never answered, by
Patomäki and Teivainen (2007, p. 95): ‘the fact that there are no global
elections or parliaments may indicate that building party-like transna-
tional organs is a relatively unattractive idea. What is the point if we
cannot win elections?’ In the Introduction to this book it was argued that
the definition which best captured what a party is, is as an organisation
which fields candidates under a common name for publically elected
assemblies. This is what separates parties from other groups in civil
society. Parties do many other things than field candidates – seek power,
pursue ideological goals, strive for access to wealth or the personal
aggrandisement for its leader(s). As was explored in Chapter 1 most such
goals are shared by everything from guerrilla groups and terrorist organi-
sations to NGOs. The fielding of candidates is what differentiates parties
from the many other organisations around them that they may share
traits with. There are, as Patomäki and Teivainen point out, no global
elected assemblies. Hence, a key problem with Scholte’s argument, that
there is a need for more globally oriented parties, is that there is nothing
for parties, as parties, to be oriented towards. For some parties, as
seekers of certain policy goals, there may well be a call for a more global
mode of operating. However, that very much depends on the specific
goals being pursued by a party – goals which are as varied as parties are
numerous.
Hence, truly global parties are only relevant in the context of global
publicly elected assemblies, of which there are none; or in the context of
specific policy goals which will only speak to some parties. The very defi-
nition of a party would, therefore, explain why both the organisational
work of parties and the study of them have remained focused on the
national. The lack of globalisation of party organisations cannot, there-
fore, be seen as a sign of parties failing. At most it can be seen as a failure
of some parties to effectively pursue specific policy goals.
The internationalisation of campaign consultancy
While parties as organisations have remained overwhelmingly national in

their operation, there is one area where there has been a significant inter-
nationalisation – the use of campaign consultants. Since the late 1960s,

campaign consultants have increasingly worked abroad (Farrell, 1998;
Plasser, 2000). This has been particularly the case for consultants nor-
mally based in the United States, a country often seen as being at the
cutting edge of ‘modern’ campaigning. The International Association of
Political Consultants had 110 members on its 2010 list of members. Of
these 36 (about 33 per cent) claim to have some level of international
experience. Of those 36, 22 come from the United States.
There are many reasons why consultants might have an incentive to go
abroad. One driving force, particularly relevant for US-based consultants,
is that there is a significant oversupply of political consultants in the
American market. According to Plasser (2000, see also Farrell, 1998)
there is a glut of campaign consultants operating in the domestic consul-
tancy market in the United States, which means that the competition for
contracts is fierce. Many consultants may, therefore, be tempted to try to
tap new markets outside the United States.
In addition there seems to be a growing demand for consultants who
are perceived as being able to provide the latest in campaigning know-
how. Techniques found to be successful in one country, often first in the
United States, attract attention in other countries where parties seek to
gain an advantage over domestic opponents. This in turn will drive a
domestic arms races further pushing up demand for expertise. As Duverger
argued in connection with the mass-party and Kirchheimer for the catch-
all party, the reason why these models spread from their specific origins to
the rest of the party system was because they were successful – and success
breeds demand. The same is the case for successful campaigning methods
and consultants deemed able to teach such methods to their clients.
Further, many long-established democracies have implemented demo-
cratic assistance programmes to new democracies (Farrell, 1998, p. 173;
Plasser, 2000, pp. 36–7), especially since the end of the Cold War. This
has often been in the form of financial aid, which has allowed parties in
newly democratic countries to buy in expert help.
There is some disagreement about what exactly the effects are of this
internationalisation of campaign consultancy. In essence the disagreement
is focused on whether what we are seeing is a process of Americanisation
or modernisation (Swanson and Mancini, 1996; see Figure 10.2). In the
Americanization view other countries are converging on the American
model. By contrast the modernisation view is that all countries are simply
moving towards ever more ‘modern’ campaign techniques, albeit with the
United States in the lead.
Which of these views is more accurate is not clear. As was explored in
Chapter 8 the debate over the exact nature of modern campaigning and
what direction it is going in is very much an ongoing affair.
‘Cutting
USA edge’
B USA
C B
A A
Figure 10.2 Americanisation vs. Modernisation
There are also several factors that limit the spread of American/new
techniques. One issue is that not all techniques ‘travel’ very well.
Especially techniques which rely on a certain level of technological pene-
tration, that is, access to the Internet or television ownership, will be vir-
tually useless in many developing countries. Hence, many of the lessons
from Barack Obama’s first presidential campaign, which saw heavy use
of web sites to coordinate local volunteers, will not be relevant in many
new democracies in, for example, Africa. In addition, in countries where
the focus of campaigning is on parties rather than individual candidates,
the demand for consultants will be lower. Where individual candidates
are responsible for their own campaigning as it the case in the United
States, there will be a much larger demand for consultants than in coun-
tries where a single-party campaign headquarter hires consultants to
work on a fairly centralised national campaign. Imagine a country of 100
parliamentary seats and five parties. That will mean 500 candidates. In a
party-oriented system each party’s head office would be in charge of
hiring consultants. There would, therefore, be five clients. In a candidate-
focused system there would be 500 potential clients. Clearly, the political
system will make a difference to how much demand there is for the work
of consultants.
However, regardless of these limitations there can be no doubt that the
internationalisation of campaign consultancy is a trend that is likely to
continue. In ever-more-competitive electoral markets, parties will want to
make use of every trick to beat their opponents. Inspiration for such
tricks is likely to increasingly come from abroad.
Conclusion
The internationalisation of party politics is not an area that has received

very much attention. This is partly because it is an area that parties have
not shown a great interest in organisationally. The one serious attempt to

create a true world party, Comintern, failed in no little part due to the
inherent problems of building international party organisations. Indeed,
there are good reasons to believe that the reason parties have not interna-
tionalised organisationally is that they simply do not need to. There are
no international assemblies elected under a universal franchise except for
the European Parliament. Unsurprisingly, it is in connection with the
European Parliament that we have seen the most advanced development
of functioning international parties. Even here it is only on a regional
scale, and the result has so far been very far from fully formed party
organisations seen at the national level. Nevertheless, parties have still
been impacted by internationalisation, especially when it comes to policy
stances and campaigning. The internationalisation of campaign consul-
tancy is an area that is likely to continue to attract attention, even as
other aspects of party organisations remain resolutely national.
Chapter 11
The Future of Party Politics
This final chapter will look at the future of the academic study of party
politics and consider what the big areas of research are. It will also try to
evaluate one of the big questions in party politics: are parties as a collec-
tive in crisis?
Future direction in the study of political parties
The study of party politics is a very active and growing field. This section
will look briefly at what is going on in research in each of the areas
covered by this book.
There is little to suggest that Sartori’s model of party systems is about
to be challenged. It has, as could be expected, been criticised, but no one
has come up with a scheme that has seriously looked to undermine
Sartori’s dominance. The discipline seems relatively content with his
scheme and mostly uses it as the starting point for research, if not explic-
itly, at least implicitly. As Bardi and Mair write, there has been ‘almost no
substantial innovation since the publication of Sartori’s classic work in
1976’ (2008, p. 148). So, the main concern in the field of party systems is
not defining them. Instead the focus is on: a) what drives party system
formation and change; and b) developing better ways of calculating party
system fragmentation. As was explored in Chapter 2 what drives the for-
mation of one party system rather than another is not clear, and this lack
of clarity has had the effect of spawning a growing literature. This has
also been driven by the emergence of new democracies in Eastern Europe,
Africa and East and South East Asia since the end of the Cold War. These
areas have provided rich soil for examining how party systems emerge
and stabilise (or fail to stabilise). So, for example Croissant and Volkel
(2012) consider the formation of party systems in East and South East
Asia; Rose and Mishler (2010) explore the lack of stability in Russia’s
party system; Mozaffar and Scarritt (2005) examine the reason behind the
low fragmentation, yet high volatility in African party systems. Another
area that has received attention is calculating the fragmentation of party
systems. Laakso and Taagepera’s (1979) approach to calculating the effec-
tive number of parties is popular and widely accepted, but has also
201
received considerable critical attention. Both the emergence and stability

of party systems and approaches to counting parties and measuring frag-
mentation are likely to continue to be the subject of scholarly concern.
The issue of how parties are developing is certainly an area that could
be said to be the foundational concern of the party politics field.
However, more specifically the ‘party types’ literature is an area of con-
tinuous growth. As was shown in Chapter 3 there is a steady stream of
suggestions for new party types. While the cadre to catch-all development
is more or less accepted, there seems to be no consensus on what has hap-
pened since. The competition to be Kirchhemimer’s successor is certainly
fierce. Most new suggestions do not get much attention, but some such as
Katz and Mair’s cartel party generate considerable debate.
The reasons for this proliferation of new types compared to the rela-
tively straightforward cadre to catch all development are not entirely
clear. As was discussed in Chapter 3, the proliferation of new party types
since Kirchheimer’s work could be down to one of two explanations.
First, it may be that democracy, and therefore party organisations have
become more complex – thus giving rise to many more new party types
than was previously the case. Second, there may just be that many more
people doing work on party organisations and, therefore, there is much
more focus on the small differences between parties as new scholars find
their individual niche. Whichever is the case, the development of new
party types is likely to continue.
The issue of ideology is an area that is dominated by the efforts of
trying to understand what the end of the Cold War and the seeming
victory of liberalism and the rise of globalisation are doing to the distinc-
tiveness of political parties. There is a strong sense that many parties
seem to be converging on a liberal free-market standpoint. However,
there is also a considerable amount of work being done on the rise of
parties challenging the status quo, especially on the far right. The far left
is still struggling to re-define itself in the wake of the collapse of
Communism, but the far right seems to have found rich soil among the
‘losers’ from globalisation, most prominently illustrated by the strong, if
also uneven, rise of anti-immigration parties in many countries. However,
one area which perhaps deserves far more attention than it has been
receiving is what ideology means in party political terms outside Europe.
As we saw in Chapter 4 many of the models developed to understand
party ideology have their origins in Western Europe. They do not always
travel very well outside Europe and there seems to be a need to develop
models that can cope better with the ideological foundations of political
parties more widely. The case of African parties suggests that the very
idea of ideology may not be easily applied to parties in certain political
contexts. This would seem to be a research agenda with a rich potential.
The Future of Party Politics 203
Party membership is not only a topic receiving considerable attention

in the academic literature, it is also the subject of a great deal of soul-
searching and agonising both in academia and among political parties.
The cause of this soul-searching and agonising is the widespread decline
in party membership numbers experienced by a great many parties. The
idea that the mass-membership party is, or perhaps increasingly was, the
‘normal’ and also the most desirable form of party organisation means
that the decline of party membership organisations is viewed with partic-
ular trepidation. There, then, seems to be three main areas active in
party-membership research.
The first is research into the trends in membership numbers. That this
area of research should continue despite the generally agreed conclusion
that numbers are going down is driven by two issues. The first is that
there are still parties where numbers are not going down, but up, or at
least merely fluctuating. Identifying which parties are resisting the general
trend will allow scholars to study them to identify what allows them to
resist decline. The second is that, as was pointed out in Chapter 5, the
extent to which mass-membership organisations were ‘normal’, as
opposed to a temporary phase, may have been exaggerated. The more
data is available the easier it is to understand the long-term developments
of membership organisations.
The second main area of research into party members is to understand
what drives decline. This is exemplified by Whiteley and Seyd’s (2002),
work based on membership surveys. Using that data they have tried to
establish whether the decline is driven by long-term societal change, or
shorter-term party-specific issues (that is, party leaders no longer seeing
membership recruitment and retention as a priority).
The third area of research in relation to party members is focused on
better understanding the members that are left. Who are this dwindling
minority of people who insist on holding on to what some regard as an
anachronistic form of political engagement? What are their beliefs; their
social, educational and employment backgrounds; what are their ideolog-
ical beliefs; and to what extent do they differ from non-member citizens?
Candidate selection procedures are a very active area of research. It is
an area that is broadly speaking of interest to two groups of scholars.
One group consist of those who want to know something about the
balance of power between the three faces of parties. As was argued in
Chapter 6, candidate selection is an enormously important area of party
life and understanding who controls it will tell us an awful lot about who
has the power in a party. In addition, candidate recruitment is of great
interest to those who want to understand the under-representation of
certain groups in society, most notably women. It is probably in the area
of gender studies that we are seeing the most active research agenda on
this topic at the moment. Women make up more than half of the world’s
population, yet are massively under-represented in almost all democrati-
cally elected parliaments. This is obviously a major problem for the dem-
ocratic world and more than warrants the attention devoted to this area
of research. This research agenda promises to continue to grow.
The issue of who controls policy making in political parties is a some-
what under-researched area. In 2002 Scarrow et al. (2002, p. 144) wrote
that ‘for all the importance of party programmes to democratic theory
and practice, we still know very little about where these programmes
come from, and particularly about the role of intra-party democracy in
the process’. Even now this still broadly holds true. There have been
several recent studies of intra-party democracy, including of Denmark
(Pedersen, 2010), Sweden (Loxbo, 2013) and comparing Denmark and
the United Kingdom (Pettitt, 2012b). However, there is still no major
comparative study to draw general conclusion across several cases about
wider patterns of who controls policy making. This is an area that holds
a lot of potential for further research. As was explored in Chapter 7, the
normative side of the argument, that is, should parties be internally dem-
ocratic, has been relatively settled on the ‘yes’ side. It seems that, gener-
ally speaking, democracy is seen as a ‘good thing’ and something to be
encouraged including within political parties.
One area that has received a lot of attention and that continues to be
very active is election campaigning and how this is affecting party organi-
sations. That this should receive as much attention as it does is hardly
surprising considering how important elections are to the very founda-
tions of what a party is about: the presenting of candidates in elections.
One area that has seen a lot of interest is the re-emergence of local cam-
paigning and with it the return in importance of having a local volunteer
base to rely on at election time. It is very likely that this will be a topic
that will continue to receive more attention as data becomes available
from future elections.
As we saw in Chapter 9 the research on coalition formation has a long
history and a steadily expanding literature. It has been argued that signif-
icant success in predicting coalition formation remains somewhat elusive,
but this will no doubt be an area where people will continue to refine the
predictive tools. One research agenda which has seen considerable
growth is recent years is how parties usually regarded as being on the
fringes of the political system cope with entering government for the first
time. A lot of attention has been focused on understanding the price that
the parties of the far right and the far left have to pay for government
participation. This is a research agenda that will no doubt continue for as
long as parties previously regarded as being beyond the pale are given the
opportunity to enter government.
The internationalisation of party politics is probably the least devel-

oped area in terms of research that has been covered in this book. This is
partly because party politics are overwhelmingly focused on domestic
politics. This is especially so when looking at the organisational side of
party politics. There are no world parties, and the only really interna-
tional parties are the Europarties linked to the European parliament.
However, Europarties have only emerged fairly recently. Hence, research
on them is also at an early stage. Nevertheless, Europarties will no doubt
provide a longstanding rich vein of research as they continue to develop.
Another area that will also continue to provide rich research pickings will
be the increasing internationalisation of campaign consultancy. Parties
have been seeking inspiration from likeminded partners internationally
and this is only likely to intensify. The ways in which this intensification
manifests itself and the consequences will almost certainly be one of the
major research agendas in the coming decades. What is not likely to
happen is the appearance of truly international parties. Without the exis-
tence of popularly elected international assemblies outside the European
Union, parties simply have no incentive to develop their organisation in
an international direction. International cooperation between parties is
likely to be limited to knowledge exchange, especially in the area of elec-
tion campaigning.
Are parties in crisis?
It is possible to argue that political parties have never been more signifi-
cant than they are now. As a collective they constitute the most important
and powerful type of political organisation in the world. No democratic
country has been able to do without them, and with the rise of new
democracies in Eastern Europe, Africa and Asia they have continued their
dominance. Modern democracy is and continues to be party democracy,
a fact confirmed in every single post-1989 democracy. Throughout the
democratic world the path to political power is exclusively through a
political party. Despite considerable anti-party feelings in most demo-
cratic countries, independents have nowhere made significant gains. For
anyone with political ambitions in democratic societies the means to real-
ising those ambitions was, is and is likely to continue to be a political
party. There is currently no viable alternative to party democracy, and
thus Schattschneider’s claim that democracy is unthinkable save in term
of political parties, cited at the very beginning of this book, holds as true
today as it did then.
However, there are also serious claims of party weakness both in the
academic literature and in the media. Broadly speaking these claims can
be summarised under three headings: unpopularity; shrinking member-

ship organisations; and disengagement.
It certainly is true that political parties are unpopular. That was
evident from the data presented in the Introduction. However, it is also
true that parties have always been viewed with suspicion. Their unpopu-
larity is, therefore, not something particularly new. It is also true that
political parties are losing members, sometimes at a significant rate. This
has been cited as one piece of evidence for the idea that political parties
are now in crisis. It is important though to keep in mind the argument
made in Chapter 5 that mass-membership organisations should not nec-
essarily be viewed as the normal state of affairs for political parties.
Indeed, the high point of the mass-membership party was a fairly brief
period immediately after the Second World War.
What may well be true is that parties are struggling to maintain their
engagement with voters. Lawson’s view of the failure of linkage and the
cartel party argument that political parties are withdrawing into the state
(see Chapter 3) do seem to have some merit. Mair’s (2006) claim that
contemporary democracy is witnessing a mutual withdrawal, of citizens
into private life or short-term forms of political engagement and of politi-
cians into the institutions of the state is backed up by considerable evi-
dence.
The question then is: are parties in crisis? At the most fundamental
level, based on the definition of what a political party is laid out in the
introduction to this book, the answer must be no. If we see a party as
being defined as an organisation that presents candidates to publically
elected assemblies then parties are, as was argued above, stronger than
ever. There is no evidence to suggest that parties are being threatened by
a rise in non-party independents. In addition there is little evidence to
make us believe that modern representative democracy is likely to turn in
a more direct democracy direction, bypassing political parties. Parties
make the most sense in a representative democracy where they are the
main and even only suppliers of representatives. If a more direct form of
democracy was to emerge, parties would potentially have much less justi-
fication for their existence. However, such a change is not in evidence.
The extent to which parties are in crisis must, therefore, be related to
the other tasks that they perform in modern democracy. Ignazi (1996, p.
550) argues that the idea of a crisis of party is mainly related to the crisis
of a certain type of party, that is, the Duvergerian mass party. As should
be clear from both Chapter 3 on the development of parties through
several stages/types and Chapter 5 on party members, the mass-party
model is not ‘normal’. It is, or was, depending on how far one thinks the
decline has gone, one temporary stage in the continual development of
political parties. Modern democracy, and indeed society as such, is not
static, but continually evolving, and that includes party organisations.

The leaders of the old cadre parties may well have regarded the rise of the
mass party as a sign of a crisis of party and even of democracy. That cer-
tainly seems to have been Ostrogorski’s stance. Ostrogorski did not have
a problem with organised groups in parliament, but was notoriously sus-
picious of permanent organisations outside parliament. To him the rise of
the mass party was a sign of crisis and decline of the old, and better,
order based on cadre parties. We now seem to be seeing a repeat of that
view: that new forms and types of party organisations are evolving that
are seen as inferior to what came before, that is, the mass party.
So, yes: parties are changing. They always have and they always will.
Just as human society is not static so parties are continually evolving as
they react to and influence changes in their environment. There is no
‘normal’ or ‘standard’ model for what a party looks like and against
which changes can be evaluated. They are, however, many normative
ideas of what parties ought to look like and what is currently happening
to parties may well deviate from some of those norms. Especially if they
rest on ideas derived from the mass-party model of party organisation.
In short, the answer to the question of whether parties are in crisis
relies on normative ideas of what a party should look like – normative
ideas that are inherently and unavoidably contested. Because these ideas
are contested the question of whether parties are in crisis or not cannot
really be settled objectively. Indeed the question has been with us for as
long as political parties have been the subject of academic study and
public debate. Parties as such show no signs of disappearing or reaching
some stable form and consequently will, therefore, continue to generate
research and disagreement. The one thing that can be said then, is that
party politics as an academic discipline is in rude good health, and that
most other questions regarding parties, including their supposed crisis, is
a matter for continuing debate.
Bibliography
African National Congress (2012) African National Congress Constitution,

Johannesburg, African National Congress.
Aldrich, John Herbert (2011) Why Parties: A Second Look, Chicago, University of
Chicago Press.
American Political Science Association Committee on Political Parties (1950) Towards
a More Responsible Two-Party System, New York, Rinehart.
Amery, L. S. (1947) Thoughts on the Constitution, Oxford, Oxford University Press.
Amery, L. S. (1964) Thoughts on the Constitution, London, Oxford University Press.
Ashworth, T. R. and H. P. C. Ashworth (1901) Proportional Representation Applied
to Party Government: A New Electoral System, London, Swan Sonnenschein.
Attlee, Clement (1937) The Labour Party in Perspective, London, Victor
Gollancz.
Axelrod, Robert (1970) Conflict of Interest, Chicago, Markham.
Bacharatz, Peter and M. S. Baratz (1963) ‘Decisions and non-decisions: an analytical
framework’, American Political Science Review, 57, 632–42.
Bardi, L. and P. Mair (2008) ‘The parameters of party systems’, Party Politics, 14, 2,
147–66.
Basedau, Matthias and Alexander Stroh (2009) ‘Ethnicity and party systems in
Francophone Sub-Saharan Africa’, GIGA Working Paper, 100, May 2009,
Hamburg, GIGA.
Beckett, Francis (1995) Enemy Within: The Rise and Fall of the British Communist
Party, London, John Murray.
Beer, Samuel H. (1965) Modern British Politics: A Study of Parties and Pressure
Groups, London, Faber and Faber.
Benoit, Kenneth and Michael Laver (2006) Party Policy in Modern Democracies,
London, Routledge.
Berntson, Lennart (1974) Politiska partier och sociala klasser: En analys av
partiteoren i den moderna statskunskapen och marxismen, Lund, Cavefors
Bokforlag.
Bille, Lars (1997) Partier i forandring, Odense, Odense Universitetsforlag.
Bille, Lars (2001) ‘Democratizing a democratic procedure: myth or reality? Candidate
selection in Western European parties’, Party Politics, 7, 3, 363–80.
Blau, Adrian (2008) ‘The effective number of parties at four scales: votes, seats,
legislative power, executive power’, Party Politics, 14, 2, 167–87.
Blondel, Jean (1968) ‘Party systems and patterns of government’, Canadian Journal
of Political Science, 1, 2, 180–203.
Blumler, J. and D. Kavanagh (1999) ‘The third age of political communication:
influence and features’, Political Communication, 16, 3, 209–30.
Bochel, John and David Denver (1983) ‘Candidate in the Labour Party: what the
selectors seek’, British Journal of Political Science, 13, 1, 45–69.
Bolleyer, Nicole (2008) ‘The organizational costs of public office’, in K. Deschouwer
(ed.), New Parties in Government, London, Routledge, 17–41.
Brass, Paul (1966) Factional Politics in an Indian State, Oxford, Oxford University
Press.
208
Bibliography 209
Browne, Eric (1973) Coalition Theories: A Logical and Empirical Technique, London,
Sage.
Browne, Eric and Mark Franklin (1973) ‘Aspects of coalition payoffs in European
Parliamentary democracies’, The American Political Science Review, 67, 2,
453–69.
Budge, Ian and Hans Keman (1990) Parties and Democracy: Coalition Formation
and Government Function in Twenty States, Oxford, Oxford University Press.
Buksti, Jacob (2005) Ånden fra finderup: formandsopgøret i Socialdemokratiet 1992,
Copenhagen, Aschehoug.
Burke, Edmund (1981) ‘Thoughts on the cause of the present discontents’, in Paul
Langford (ed.), The Writings and Speeches of Edmund Burke: Volume 2: Party
Parliament and the American Crisis, 1766–1774, Oxford, Clarendon Press,
251–322.
Burnell, P. (2001) ‘The party system and party politics in Zambia: continuities past,
present and future’, African Affairs, 100, 399, 239–63.
Carty, R. Kenneth (2004) ‘Parties as franchise systems: the stratarchical organizational
imperative’, Party Politics, 10, 1, 5–24.
Cecil, Hugh (1912) Conservatism, London, Williams and Norgate.
Christensen, Dag (2010) ‘The Danish Socialist People’s Party: still waiting after all
these years’, in Jonathan Olsen, Michael Koß and Dan Hough (eds), Left Parties
in National Governments, Basingstoke, Palgrave Macmillan, 121–37.
Clapham, Christopher (1996) Africa in the International System: The Politics of State
Survival, Cambridge, Cambridge University Press.
Clark, Peter and James Wilson (1961) ‘Incentive systems: a theory of organizations’,
Administrative Science Quarterly, 6, 2, 129–66.
Coffe, H. (2008) ‘Social Democratic parties as buffers against the extreme right: the
case of Belgium’, Contemporary Politics, 14, 2, 179–95.
Colomer, Josep (2005) ‘It’s parties that choose electoral systems (or Duverger’s law
upside down),’Political Studies, 53, 1, 1–21.
Conservative Party (2012) Guide to Becoming a Conservative MP, London, The
Conservative Party.
Croissant, Aurel and Philip Vokel (2012) ‘Party system types and party system
institutionalization: comparing new democracies in East and Southeast Asia’, Party
Politics, 18, 2, 235–65.
Cowley, Philip (2005) The Rebels: How Blair Mislaid His Majority, London,
Politico’s.
Cutts, David, Sarah Childs and Edward Fieldhouse (2005) ‘“This is what happens
when you don’t listen”’, Party Politics, 14, 5, 575–95.
Daalder, Hans (1992) ‘A crisis of party?’, Scandinavian Political Studies, 15, 4,
269–88.
Dahlerup, Drude (2001) Men kvinderne vil jo ikke selv: diskurs omkring betydningen
af køn ved danske kommunevalg, Aalborg, Aalborg Universitet GEP-Tekstserie
No. 2.
Damgaard, Erik and Palle Svensson (1989) ‘Who governs? Parties and policies in
Denmark’, European Journal of Political Research, 17, 6, 731–45.
Day, Stephen (2005) ‘Developing a conceptual understanding of Europe’s
transnational political parties (with a specific focus on the Party of European
Socialists)’, Journal of Contemporary European Studies, 13, 1, 59–77.
De Winter, Lieven and Patrick Dumont (2006) ‘Parties into government: still many
puzzles’, in Richard S. Katz and Willian Crotty (eds), Handbook of Party Politics,
London, Sage, 175–88.
210 Bibliography
Deschouwer, Kris (2008) ‘Comparing newly governing parties’, in K. Deschouwer

(ed.), New Parties in Government, London, Routledge, 1–16.
Diermeier, Daniel and Antonio Merlo (2004) ‘An empirical investigation of coalitional
bargaining procedures’, Journal of Public Economics, 88, 3–4, 783–97.
Dominguez, Casey (2005) ‘Does a house divided really fall? A test of the consequences
of party unity around Primary candidates’, Paper presented at the annual meeting
of the Midwest Political Science Association, Palmer House Hilton, Chicago,
Illinois, 7 April.
Downs, Anthony (1957) An Economic Theory of Democracy, New York, Harper and
Brothers .
Draper, Theodore (1957) The Roots of American Communism, New York, Viking
Press.
Dunleavy, Patrick (1991) Democracy, Bureaucracy and Public Choice, New York,
Harvester.
Duverger, Maurice (1964) Political Parties: Their Organization and Activity in the
Modern State, London, Methuen.
EIU (2011) The Democracy Index 2011: Democracy under Stress, London, Economist
Intelligence Unit.
Eldersveld, Samuel (1964) Political Parties: A Behavioural Analysis, Chicago, Rand
McNally.
Epstein, Leon (1967) Political Parties in Western Democracies, New York, Praeger.
Epstein, Leon (1980) Political Parties in Western Democracies, New Brunswick,
Transaction Publishers.
Evans, Elizabeth (2008) ‘Supply and demand: women candidates and the Liberal
Democats’, British Journal of Politics and International Relations, 4, 10,
590–606.
Farrell, David (1996) ‘Campaign strategies and tactics’, in Lawrence Leduc, Richard
Niemi and Pippa Norris (eds), Comparing Democracies: Elections and Voting in
Global Perspective, Thousand Oak, Sage, 156–81.
Farrell, David (1998) ‘Political consultancy overseas: the internationalization of
campaign consultancy’, PS: Political Science and Politics, 31, 2, 171–6.
Farrell, David and Paul Webb (2002) ‘Political parties as campaign organizations’, in
Russel Falton and Martin Wattenberg (eds), Parties Without Partisans: Political
Change in Advanced Industrial Democracies, Oxford, Oxford University Press,
102–28.
Farrell, David, Robin Kolodny and Stephen Medvic (2001) ‘Parties and campaign
professionals in a digital age: political consultants in the United States and their
counterparts overseas’, The Harvard International Journal of Press/Politics, 6, 4,
11–30.
Fisher, J. and T. Eisenstadt (2004) ‘Introduction: comparative party finance’, Party
Politics, 10, 6, 619–26.
Fisher, Justin, David Denver, Edward Fieldhouse, David Cutts and Andrew Russell
(2005) Constituency Campaigning in the 2005 British General Election, Paper
prepared for the Annual Conference of the PSA Specialist Group on Elections,
Public Opinion and Parties, Essex, September.
Franceschet, Susan (2005) Women and Politics in Chile, Boulder, Lynne Reinner.
Frechette, G., J. Kagel and M. Morelli (2005) ‘Nominal bargaining power, selection
protocol, and discounting in legislative bargaining’, Journal of Public Economics,
89, 8, 1497–517.
Freire, A (2008) ‘Party polarization and citizens’ left-right orientation’, Party Politics,
14, 2, 189–209.
Bibliography 211
Gagatek, Wojciech (2009) European Political Parties as Campaign Organisations:

Toward a Greater Politicisation of the European Parliament Elections, Brussels,
Centre for European Studies.
Gallagher, Michael (1985) Political Parties in the Irish Republic, Manchester,
Manchester University Press.
Gallagher, Michael (1988) ‘Introduction’, in Michael Gallagher and Michael Marsh
(eds) Candidate Selection in Comparative Perspective: The Secret Garden of
Politics, London, Sage, 1–19.
Gallagher, Michael and Michael Marsh (eds) (1988) Candidate Selection in
Comparative Perspective: The Secret Garden of Politics, London, Sage.
Gallagher, Michael and Paul Mitchell (2008) ‘Appendix B: Indices of fragmentation
and disproporitonality’, in Michael Gallagher and Paul Mitchell (eds), The Politics
of Electoral Systems, Oxford, Oxford University Press, 598–606.
Gallagher, Michael, Michael Laver and Peter Mair (2011) Representative
Government in Modern Europe, Maidenhead, McGraw Hill.
Gamson, William (1961) ‘A theory of coalition formation’, American Sociological
Review, 26, 3, 373–82.
Giddens, Anthony (1998) The Third Way: The Renewal of Social Democracy,
Cambridge, Polity.
Giddens, Anthony (2000) The Third Way and Its Critics, Cambridge, Polity.
Gerring, John (1998) Party Ideologies in America, 1828–1996, Cambridge,
Cambridge University Press.
Gibson, Rachel and Andrea Rommele (2001) ‘Changing campaign communications:
A party centered theory of professionalized campaigning’, Harvard International
Journal of Press/Politics, 6, 4, 31–44.
Gibson, Rachel and Andrea Rommele (2009) ‘Measuring the professionalization of
political campaigning’, Party Politics, 15, 3, 265–93.
Goldman, Ralph (1983) Transnational pparties: Organizing the World’s Precincts,
New York, University Press of America.
Gould, Philip (1998) The Unfinished Revolution: How the Modernisers Saved the
Labour Party, London, Abacus.
Gray, John (1997) Is Conservatism Dead?, London, Profile Books.
Green, Donald and Alan Gerber (2008) Get out the Vote: How to Increase Voter
Turnout, Washington, DC, The Brookings Institute.
Grum, John G. (1958) ‘Theories of electoral systems’, Midwest Journal of Political
Science, 2, 4, 357–76.
Gutmann, Amy and Dennis Thompson (1996) Democracy and Disagreement,
Cambridge, MA, The Belknap Press of Harvard University.
Hansen, Bernhard and Karina Pedersen (2003) ‘Medlemsrollen og det interne
partidemokrati’, in Lars Bille and Jørgen Elklit (eds), Partiernes Medlemmer,
Aarhus, Aarhus Universitets Forlag, 103–31.
Hardgrave, Robert and Stanley Kochanek (2008) India: Government and Politics in
a Developing Nation, Boston, Thomson Wadsworth.
Harmel, Robert and Kenneth Janda (1994) ‘An integrated theory of party goals and
party change’, Journal of Theoretical Politics, 6, 3, 259–87.
Haupt, Andrea B. (2010) ‘Parties’ responses to economic globalization. What is left
for the Left and right for the Right?’, Party Politics, 16, 1, 5–27.
Hazan, Reuven and Gideon Rahat (2010) Democracy within Parties: Candidate Selection
Methods and their Political Consequences, Oxford, Oxford University Press.
Heidar, Knut (2006) ‘Party membership and participation’, in Richards S. Katz and
William Crotty (eds), Handbook of Party Politics, London, Sage, 301–15.
212 Bibliography
Heidar, Knut and Jo Saglie (2002) Hva skjer med Partierne?, Oslo, Gyldendal.
Hertner, Isabelle (2011) ‘Are European election campaigns Europeanized? The case
of the Party of European Socialists in 2009’, Government and Opposition, 46, 3,
321–44.
Hindess, Barry (1971) The Decline of Working Class Politics, London, MacGibbon
and Kee.
Hinnfors, Jonas (2006) Reinterpreting Social Democracy: A History of Stability in
the British Labour Party and the Swedish Social Democratic Party, Manchester,
Manchester University Press.
Hoffman, David (1961) ‘Intra-party democracy: a case study’, The Canadian
Journal of Economics and Political Science, 27, 2, 223–35.
Hofstadter, Richard (1948) The American Political Tradition and the Men Who Made
It, New York, A. A. Knopf.
Hopkin, J. (2001) ‘Bringing the members back in? Democratizing candidate selection
in Britain and Spain’, Party Politics, 7, 3, 343–61.
Hopkin, J. and Paolucci, C. (1999) ‘The business firm model of party organization:
cases from Spain and Italy’, European Journal of Political Research, 35, 3,
307–39.
Hopkin, K. (2004) ‘The problem with party finance: theoretical perspectives on the
funding of party politics’, Party Politics, 10, 6, 627–51.
Hough, Dan, Michael Koß and Jonathan Olsen (2007) The Left Party in
Contemporary German Politics, London, Palgrave Macmillan.
Howard, Norman (2005) A New Dawn: The General Election of 1945, London,
Politico’s.
Ignazi, Piero (1996) ‘The crisis of parties and the rise of new political parties’, Party
Politics, 2, 4, 549–66.
Inglehart, Ronald (1977) Silent Revolution: Changing Values and Political Styles
among Western Politics, Princeton, Princeton University Press.
Isserman, Maurice (1982) Which Side Were You On? The American Communist Party
During the Second World War, Middletown, Wesleyan University Press.
Jacobs, Herbert (1962) ‘Internal recruitment of elected officials in the US – a model’,
The Journal of Politics, 24, 4, 703–16.
Jacobsen, Kurt (1996) Aksel Larsen – en politisk biografi, Copenhagen, Vinderose.
Johansson, Karl and Tapio Raunio (2005) ‘Regulating Europarties: cross-party
coalitions capitalizing on incomplete contracts’, Party Politics, 11, 5, 515–34.
Johnston, Ron, and Charles Pattie (2008) ‘How much does a vote cost? Incumbency
and the impact of campaign spending at English general elections’, Journal of
Elections, Public Opinion and Parties, 18, 2, 129–52.
Karp, Jeffery, Susan Banducci and Shaun Bowler (2008) ‘Getting out the vote: party
mobilization in a comparative perspective’, British Journal of Political Science, 38,
1, 91–112.
Katz, Richard S. (1986) ‘Party government: a rationalistic conception’, in Francis G.
Castles and Rudolf Wildenmann (eds), Visions and Realities of Party Government,
New York, Walter de Gruyter, 31–50.
Katz, Richard S. (1997) Democracy and Elections, London, Oxford University Press.
Katz, Richard S. and Robin Kolodny (1994) ‘Party organizations and an empty vessel:
parties in American politics’, in Richard Katz and Peter Mair (eds), How Parties
Organize: Change and Adaptation in Party Organizations in Western
Democracies, London, Sage, 23–51.
Katz, Richard S. and Peter Mair (eds) (1992) Party Organizations: A Data Handbook,
London, Sage.
Bibliography 213
Katz, Richard and Peter Mair (1993) The evolution of party organizations in Europe:
the three faces of party organization’, The American Review of Politics, 14,
593–617.
Katz, Richard and Peter Mair (1995) Changing models of party organization and
party democracy: the emergence of the cartel party’, Party Politics, 1, 1, 5–28.
Katz, Richard S. and Peter Mair (2002) The ascendency of the party in public office:
party organisational changes in twentieth-century democracies’, in Richard
Gunther, Jose Ramon Montero and Juan J. Linz (eds), Political Parties: Old
Concepts and New Challenges, Oxford, Oxford University Press, 113–36.
Kennedy, Fiachra, Pat Lyons and Peter Fitzgerald (2006) ‘Pragmatists, ideologues and
the general law of curvilinear disparity: the case of the Irish Labour Party’, Political
Studies, 54, 4, 786–805.
Kirchheimer, Otto (1966) ‘The transformation of the Western European party’, in
Joseph LaPalombara and Myron Weiner (eds), Political Parties and Political
Development, Princeton, Princeton University Press, 177–200.
Kitschelt, Herbert (1989) ‘The internal politics of parties: the laws of curvilinear
disparity revisited’, Political Studies, 37, 3, 400–21.
Kittilson, M, and S. Scarrow (2003) ‘Political parties and the rhetoric and realities of
democratcization’, in B. E. Cain, R. J. Dalton and S. Scarrow (eds), Democracy
Transformed? Expanding Political Opportunities in Advanced Industrial
Democracies, Oxford, Oxford University Press, 59–80.
Kolodny, Robin (2000) ‘Electoral partnerships: political consultants and political
parties’, in A. Thuber and Candice Nelson (eds), Political Consultants in Elections,
Washington DC, Brookings Institute Press, 110–32.
Krook, Mona Lena (2010a) ‘Beyond supply and demand: a feminist-institutionalist
theory of candidate selection’, Political Research Quarterly, 63, 4, 707–20.
Krook, Mona Lena (2010b) ‘Why are fewer women than men elected: gender and the
dynamics of candidate selection’, Political Studies Review, 8, 2, 155–68.
Kriesi, Hanspeter (1998) ‘The transformation of cleavage politics: the 1997 Stein
Rokkan lecture’, European Journal of Political Research, 33, 2, 165–85.
Kriesi, Hanspeter, Edgar Grande, Romain Lachat, Martin Dolezal, Simon Bornschier
and Timotheos Frey (2006) ‘Globalization and the transformation of the national
political space: six European countries compared’ European Journal of Political
Research, 45, 6, 921–56.
Krishna, Gopal (1967) ‘One party dominance: development and trends’, in Rajni
Kothari (ed.), Party System and Election Studies, Mumbai, Allied Publisher,
133–56.
Laakso, M. and R. Taagepera (1979) ‘“Effective” number of parties: a measure with
application to Western Europe’, Comparative Political Studies, 12, 1, 3–27.
Labour Party (2013) Labour Party Rule Book 2013, London, The Labour Party.
Laffin, Martin, Eric Shaw and Gerald Taylor (2005) ‘The Labour Party and
devolution’, Economic and Social Research Council’s Devolution and
Constitutional Change Programme Briefing No. 14, January.
Laponce, J. A. (1981) The Topography of Political Perceptions, Toronto, University
of Toronto Press.
Laver, Michael and Norman Schofield (1990) Multiparty Government: The Politics
of Coalition in Europe, Oxford, Oxford University Press.
Laver, Michael and Kenneth Shepsle (1990) ‘Government coalitions and intraparty
politics’, British Journal of Political Science, 20, 4, 489–507.
Lawless, Jennifer and Richard Fox (2005) It Takes a Candidate: Why Women Don’t
Run For Office, Cambridge, Cambridge University Press.
214 Bibliography
Lawless, Jennifer and Richard Fox (2010) It Still Takes a Candidate: Why Women
Don’t Run For Office, Cambridge, Cambridge University Press.
Lawson, Kay (1988) ‘When linkage fails’, in Kay Lawson and Peter Merkl (eds), When
Parties Fail: Emerging Alternative Organizations, Princeton, Princeton University
Press, 3–12.
Laybourn, Keith (2001) A Century of Labour: A History of the Labour Party, Stroud,
Sutton.
LeDuc, Lawrence (2001) ‘Democratizing party leadership selection’, Party Politics,
7, 3, 323–42.
Lees-Marshment, Jennifer (2001) Political Marketing and British Political Parties:
The Party’s Just Begun, Manchester, Manchester University Press.
Lightfoot, Simon (2006) ‘The consolidation of Europarties? The “party regulation”
and the development of political parties in the European Union’, Representation,
42, 4, 303–14.
Lijphart, Arend (1984) Democracies: Patterns of Majoritarian and Consensus
Government in Twenty-One Countries, New Haven, Yale University Press.
Lijphart, Arend (1999) Patterns of Democracy: Government Forms and Performance
in Thirty-Six Countries, New Haven, Yale University Press.
Lilleker, Darren (2005) ‘The impact of political marketing on internal party
democracy’, Parliamentary Affairs, 58, 3, 570–84.
Lipset, Seymour M. and Martin Rokkan (1967) Party Systems and Voter Alignments,
New York, Free Press.
Lipson, Leslie (1964) The Democratic Civilization, Oxford, Oxford University
Press.
Loxbo, Karl (2013) ‘The face of intra-party democracy: leadership autonomy and
activist influence in the mass party and the cartel party’, Party Politics, 19, 4,
537–54.
Lucardie, P (2000) ‘Prophets, purifiers and prolocutors: towards a theory on the
emergence of new parties’, Party Politics, 6, 2, 175–85.
Lundell, Krister (2004) ‘Determinants of candidate selection: the degree of
centralization in comparative perspective’, Party Politics, 10, 1, 25–47.
MacDonald, J. R. (1909) Socialism and Government, London, Independent Labour
Party.
Macintyre, Stuart (1980) A Proletarian Science: Marxism in Britain, 1917–1933,
Cambridge, Cambridge University Press.
Mair, Peter (2003), ‘Political parties and democracy: what sort of future?’, Central
European Political Science Review, 4, 13, 7–15.
Mair, Peter (2006) ‘Ruling the void: the hollowing of Western democracy’, New Left
Review, 42, 25–51.
Mair, Peter and Cas Mudde (1998) ‘The party family and its study’, Annual Review
of Political Science, 1, 211–29.
Manning, Carrie (2005) ‘Assessing African party systems after the Third Wave’, Party
Politics, 11, 6, 707–27.
Margetts, Helen (2001) ‘The cyber party’, paper presented at the ECPR Joint Sessions
of Workshops, Grenoble, 6–11 April.
Mattila, Mikko and Tapio Raunio (2004) ‘Does winning pay? Electoral success and
government formation in 15 West European countries’, European Journal of
Political Research, 43, 2, 263–85.
Marsh, Michael (2004) ‘None of that post-modern stuff around here: grassroots
campaigning in the 2002 Irish General Election’, British Elections and Parties
Review, 14, 245–67.
Bibliography 215
May, John D. (1973) ‘Opinion structure of political parties: the special law of
curvilinear disparity’, Political Studies, 21, 2, 135–51.
McKenzie, Robert (1963) British Political Parties, London, Mercury Books.
McKenzie, Robert (1982) ‘Power in the Labour Party: the issue of intra-party
democracy’, in Dennis Kavanagh (ed.), The Politics of the Labour Party, London,
George Allan and Unwin, 191–201.
Michels, Robert (1915) Political Parties: A Sociological Study of the Oligarchical
Tendencies of Modern Democracy, New York, Hearst’s International Library
Company.
Miliband, Ralph (1958) ‘Party democracy and parliamentary government’, Political
Studies, 6, 2, 170–4.
Mill, John Stuart (1994) Principles of Political Economy, Oxford, Oxford University
Press.
Mozaffar, S. and J. Scarritt (2005) ‘The puzzle of African party systems’, Party
Politics, 11, 4, 399–421.
Muller, Wolfgang and Kaare Strøm (2000) ‘Conclusion: cabinet governance in
Western Europe’, in Wolfgang Muller and Kaare Strøm (eds), Coalition
Government in Western Europe, Oxford, Oxford University Press, 403–30.
Murray, Rainbow (2010) Parties, Gender Quotas and Candidate Selection in France,
Basingstoke, Palgrave Macmillan.
Murray, Rainbow, Mona Lena Krook and Katherine Opello (2012) ‘Why are gender
quotas adopted? Party pragmatism and parity in France’, Political Research
Quarterly, 65, 3, 529–43.
New Patriotic Party of Ghana (2009) Constitution of the New Patriotic Party,
Accra North, New Patriotic Party of Ghana.
Newell, James L. (2010), The Politics of Italy: Governance in a Normal Country,
Cambridge, Cambridge University Press.
NIMD (2004) Internal Party Democracy: The State of Affairs and the Road Ahead –
Results of the Mini-Survey Commissioned by NIMD Knowledge Centre, The
Hague, NIMD.
Niven, Favid (1998) ‘Party elites and women candidates: the shape of bias’, Women
and Politics, 19, 2, 57–80.
Norris, Pippa (1995) ‘May’s law of curvilinear disparity revisited: leaders,
officers, members and voters in British political parties’, Party Politics, 1, 1,
29–47.
Norris, Pippa (1997) ‘Introduction: theories of recruitment’, in Pippa Norris (ed.),
Passages to Power: Legislative Recruitment in Advanced Democracies, Cambridge,
Cambridge University Press, 1–14.
Norris, Pippa (2002) ‘Campaign Communications’, in Lawrence Le Duc, Richard G.
Niemi and Pippa Norris (eds), Comparing Democracies, 2nd edn., London, Sage,
127–47.
Norris, Pippa (2006) ‘Recruitment’, in Richard S. Katz and William Crotty (eds),
Handbook of Party Politics, London, 89–109.
Norris, Pippa and Joni Lovenduski (1993) Gender and Party Politics, London, Sage.
Norris, Pippa and R. Mattes (2003) ‘Does ethnicity determine support for the
incumbent party?’, KSG Working Papers Series No. RWP03-009.
Norris, Pippa, Elizabeth Vallence and Joni Lovenduski (1992) ‘Do candidates make
a difference? Gender, race, ideology and incumbancy’, Parliamentary Affairs, 45,
4, 496–517.
Obama, Barack (2006) The Audacity of Hope: Thoughts on Reclaiming the American
Dream, Edinburgh, Canon Gate Books.
216 Bibliography
Obler, Jeffery (1974) ‘Intraparty democracy and the selection of parliamentary

candidates: the Belgian case’, British Journal of Political Science, 4, 2, 157–84.
Olsen, Jonathan (2010) ‘The Norwegian Socialist Left Party: office-seekers in the
service of policy?’, in Jonathan Olsen, Michael Koß and Dan Hough (eds), Left
Parties in National Governments, Basingstoke, Palgrave Macmillan, 16–32.
Olson, Mancur (1965) The Logic of Collective Action: Public Good and the Theory
of Groups, Cambridge, Harvard University Press.
O’Neill, Michael (2000) ‘Preparing for power: the German Greens and the challenge
of party politics’, Contemporary Politics, 6, 2, 165–84.
Ostrogorski, M. (1902) Democracy and the Organization of Political Parties, London,
Macmillan & Co.
Panebianco, Angelo (1988) Political Parties: Organization and Power, Cambridge,
Cambridge University Press.
Patomäki, Heikki and Teivo Teivainen (2007) ‘Researching global political parties’,
in Katarina Sehm-Patomäki and Marko Ulvila (eds), Global Political Parties,
London, Zed Books, 92–113.
Pattie, Charles and Ron Johnston (2009) ‘Still talking, but is anyone listening? The
changing face of constituency campaigning in Britain, 1997–2005’ Party Politics,
15, 4, 411–34.
Pedersen, Helene Hedboe (2010) ‘Differences and changes in Danish party
organisations: central party organisations versus parliamentary party group
power’, The Journal of Legislative Studies, 16, 2, 233–50.
Pedersen, Karina and Bernhard Hansen (2003) ‘Partimedlemmernes aktivitet’, in Lars
Bille and Jørgen Elklit (eds), Partiernes Medlemmer, Aarhus, Aarhus
Universitetsforlag, 73–102.
Pedersen, Mogens N. (1989) ‘En kortfattet oversight over det danske partisystems
udvikling’, Politica, 21, 3, 265–77.
Pettitt, Robin T. (2007) ‘Challenging the leadership: the party conference as a platform
for dissenting membership voice in British and Danish parties of the left’,
Scandinavian Political Studies, 30, 2, 229–48.
Pettitt, Robin T. (2012a) ‘Internal party political relationship marketing: encouraging
activism amongst local party members’, in Jennifer Lees-Marshment (ed.),
Routledge Handbook of Political Marketing, Abingdon, Routledge, 137–50.
Pettitt, Robin T. (2012b) ‘Exploring variations in intra-party democracy: a
comparative study of the British Labour Party and then Danish Centre-Left’,
British Journal of Politics and International Relations, 14, 4, 630–50.
Pelling, Henry (1958) The British Communist Party: A Historical Profile, London,
A. and C. Black.
Plasser, F. (2000) ‘American campaign techniques worldwide’, The Harvard
International Journal of Press/Politics, 5, 4, 33–54.
Plasser, F. (2001) ‘Parties’ diminishing relevance for campaign professionals’, Harvard
International Journal of Press/Politics 6, 4, 44–59.
Plouffe, David (2009) The Audacity to Win: The Inside Story and Lessons of Barack
Obama’s Historic Victory, London, Viking Penguin.
Poguntke, Thomas (2001) ‘From nuclear building sites to cabinet: the career of the
German Green Party’, Keele European Research Unit Working Paper 6.
Popkin, Samuel (1991) The Reasoning Voter, Chicago, University of Chicago Press.
Puhle, Hans-Jürgen (2002) ‘Still the age of catch-allism?: Volksparteien and
Parteienstaat in crisis re-equilibration’, in Richard Gunther, Jose Ramon Montero
and Juan J. Linz (eds), Political Parties: Old Concepts and New Challenges,
Oxford, Oxford University Press, 58–83.
Bibliography 217
Ranney, Austin (1962) The Doctrine of Responsible Party Government: Its Origins
and Present State, Urbana, University of Illinois Press.
Ranney, Austin (1981) ‘Candidate selection’, in David Butler, Howard Penniman and
Austin Ranney (eds), Democracy at the Polls, Washington DC, American
Enterprise Institute, 75–106.
Reinhardt, Gina and Jennifer Victor (2008) Competing for the Platform, the
Politics of Interest Group Influence on Political Platforms, paper presented
to the 105th Annual Meetings of the American Political Science Association,
Toronto.
Richards, Paul (2000) Is the Party Over? New Labour and the Politics of
Participation, London, Fabian Society.
Riker, William (1962) The Theory of Political Coalitions, New Haven, Yale University
Press.
Riker, William (1982) ‘The two-party system and Duverger’s Law: an essay on the
history of political science’, 76, 4, 753–66.
Rommele, Andrea, David M. Farrell and Piero Ignazi (2005) ‘Preface’, in Andrea
Rommele, David M. Farrell and Piero Ignazi (eds), Political Parties and Party
Systems: The Concept of Linkage Revisited, London, Praeger, vii–x.
Rose, Richard and Thomas Mackie (1983) ‘Incumbency in government: asset or
liability?’, in Hans. Daalder and Peter Mair (eds), Western European Party
Systems: Continuity and Change, London, Sage, 115–37.
Rose, Richard and William Mishler (2010) ‘A supply-demand model of party-system
institutionalization: the Russian case’, Party Politics, 16, 6, 801–21.
Rousseau, Jean Jacques (1966) The Social Contract and Discourses, London,
Everyman’s Library.
Russell, Meg (2005) Must Politics Disappoint?, London, Fabian Society Pamphlet no.
614.
Sartori, Giovanni (2005) Parties and Party Systems: A Framework for Analysis,
Colchester, ECPR Press.
Scarrow, Susan E. (1996) Parties and Their Members: Organizing for Victory in
Britain and Germany, Oxford, Oxford University Press.
Scarrow, Susan E. (2002) Perspectives on Political Parties: Classic Readings,
Basingstoke, Palgrave Macmillan.
Scarrow, Susan E (2005) Implementing Intra-Party Democracy, Washington, DC, The
National Democratic Institute for International Affairs.
Scarrow, Susan E. (2006) ‘The nineteenth-century origins of modern political parties:
the unwanted emergence of party-based politics’, in Richard S. Katz and William
Crotty (eds), Handbook of Party Politics, London, Sage, 16–24.
Scarrow, Susan E. and Burcu Gezgor (2010) ‘Declining memberships, changing
members? European political party members in a new era’, Party Politics, 16, 6,
823–43.
Scarrow, Susan E., Paul Webb and David M. Farrell (2002) ‘From social integration
to electoral contestation: the changing distribution of power within political
parties’, in Russell J. Dalton and Martin P. Wattenberg (eds), Parties without
Partisans: Political Change in Advanced Industrial Democracies, Oxford, Oxford
University Press, 129–56.
Schattschneider, E. E. (1942) Party Government, New Brunswick, Transaction
Publishers.
Schattschneider, E. E. (1960) The Semi-Sovereign People: A Realist View of
Democracy in America, New York, Holt, Rinehart and Winston.
Scholte, Jan (2007) ‘Political parties and global democracy’, in Katarina Sehm-
218 Bibliography
Patomäki and Marko Ulvila (eds), Global Political Parties, London, Zed Books,
12–38.
Schumpeter, Joseph A. (1976) Capitalism, Socialism and Democracy, London,
Routledge.
Sehm-Patomäki, Katarina and Marko Ulvila (2007) ‘Introduction’, in Katarina Sehm-
Patomäki and Marko Ulvila (eds), Global Political Parties, London, Zed Books,
1–11.
Seyd, Patrick (1987) The Rise and Fall of the Labour Left, London, Macmillan.
Seyd, Patrick (1999) ‘New parties/new politics?: A case study of the British Labour
Party’, Party Politics, 5, 3, 383–405.
Seyd, Patrick and Paul Whiteley (2002) New Labour’s Grassroots: The Transformation
of the Labour Party Membership, Basingstoke, Palgrave Macmillan.
Seyd, Patrick and Paul Whiteley (2004) ‘British party members: an overview’, Party
Politics, 10, 4, 355–66.
Shugart, M. S. (2008) ‘Comparative electoral systems research: the maturation of a
field and new challenges ahead’, in Michael Gallagher and Paul Mitchell (eds), The
Politics of Electoral Systems, Oxford, Oxford University Press, 25–56.
Siglas, Emmanuel, Monika Mokre, Johannes Pollak, Peter Slominski and Jozef Batora
(2010) Democracy Models and Parties at the EU Level: Empirical Evidence from
the Adoption of the 2009 European Election Manifestos, Recon Online Working
Paper 2010/13.
Smith, Jennifer (2009) ‘Campaigning and the catch-all party: the process of party
transformation in Britain’, Party Politics, 15, 5, 555–72.
Spirer, Herbert F. and Louise Spirer (1991) ‘The use of statistics: Bismarck, sausages
and policy’, American Journal of Economics and Sociology, 50, 3, 347–50.
Stanyer, James (2001 The Creation of Political News: Television and British Party
Political Conferences, Brighton, Sussex Academic Press.
Strøm, Kaare (1990) ‘A behavioural theory of competitive political parties’, American
Journal of Political Science, 34, 2, 565–98.
Studlar, Donlev T. and Ian McAllister (1991) ‘Political recruitment to the Australian
legislature: towards an explanation of women’s electoral disadvantages’, The
Western Political Quarterly, 44, 2, 467–85.
Svåsand, L. (2008) Internal Party Democracy: The Case of the United Democratic
Front in Malawi, paper presented to Nordic Political Science Association
Convention, Tromsø.
Swanson, David and Paolo Mancini (1996) Politics, Media and Modern Democracy:
An International Study of Innovations in Electoral Campaigning and their
Consequences, Westport, Greenwood Publishing Group.
Taagepera, R. (2003) ‘Arend Lijphart’s dimensions of democracy: logical connections
and institutional design’, Political Studies, 51, 1, 1–19.
Taagepera, R. and B. Grofman (1985) ‘Rethinking Duverger’s Law: predicting the
effective number of parties in plurality and PR systems – parties minus issues equal
one’, European Journal of Political Research, 13, 341–52.
Teorell, Jan (1999) ‘A deliberative defence of intra-party democracy’, Party Politics,
5, 3, 363–82.
Thorpe, Andrew (1998) ‘Comintern “control” of the Communist Party of Great
Britain’, The English Historical Review, 113, 452, 637–62.
Trapeznik, Alexander (2009) ‘The Comintern and the Communist Party of New
Zealand’, Journal of Cold War Studies, 11, 1, 124–49.
Twyman, Joe (2008) ‘Getting it right: YouGov and online survey research in Britain’,
Journal of Elections, Public Opinion and Parties, 18, 4, 343–54.
Bibliography 219
van Biezen, Ingrid (2004) ’Political parties as public utilities’, Party Politics, 10, 6,
701–22.
van Biezen, Ingrid and Petr Kopecky (2007) ‘The state and the parties: public funding,
public regulation and party patronage in contemporary democracies’, Party
Politics, 13, 2, 235–54.
van Biezen, Ingrid, Peter Mair, and Thomas Poguntke (2011) ‘Going, going, ... gone?
The decline of party membership in contemporary Europe’, European Journal of
Political Science, 51, 1, 24–56.
Verney, Douglas (2004) ‘How has the proliferation of parties affected the Indian
Federation: a comparative approach’, in Soya Hasan, E. Sridharaan and R
Sudarshan (eds), India’s Living Consitution, Delhi, Permanent Black, 134–58.
von Beyme, Klaus (1985) Political Parties in Western Democracies, Aldershot, Gower.
Von Neumann, John and Oskar Morgenstern (1953) Theory of Games and Economic
Behaviour, Princeton, Princeton University Press.
Ware, Alan (1979) The Logic of Party Democracy, London, Macmillan.
Ware, Alan (1996) Political Parties and Party Systems, Oxford, Oxford University
Press.
Warwick, Paul (1999) ‘Ministerial autonomy or ministerial accommodation?
Contested bases of government survival in parliamentary democracies’, British
Journal of Political Science, 29, 2, 369–94.
Warwick, Paul (2002) ‘Toward a common dimensionality in West European policy
spaces’, Party Politics, 8, 1, 101–22.
Warwick, Paul V. and James N. Druckman (2006) ‘The portfolio allocation paradox:
an investigation into the nature of a very strong but puzzling relationship’,
European Journal of Political Research, 45, 4, 635–65.
Weber, Max (1990) ‘The advent of plebiscitarian democracy’, in Peter Mair (ed.), The
West European Party System, Oxford, Oxford University Press, 31–7.
Weinberg, Martha (1977) ‘Writing the Republican platform’, Political Science
Quarterly, 92, 4, 655–62.
West, Darrell (2000) Checkbook Democracy: How Money Corrupts Political
Campaigns, Boston, Northeastern University Press.
White, John K. (2006) ‘What is a political party?’, in Richard S. Katz and William
Crotty (eds), Handbook of Party Politics, London, Sage, 5–15.
Whiteley, Paul (2011) ‘Is the party over? The decline of party activism and
membership across the democratic world’, Party Politics, 17, 1, 21–44.
Whiteley, Paul and Patrick Seyd (2002) High-Intensity Participation: The Dynamics
of Party Activism in Britain, Ann Arbor, University of Michigan Press.
Whiteley, Paul and Patrick Seyd (2003) ‘How to win a landslide by really trying: the
effects of local campaigning on voting in the 1997 British general election’,
Electoral Studies, 22, 2, 301–24.
Whitfield, Lindsay (2009) ‘“Change for a better Ghana” – party competition,
institutionalization and alternation in Ghana’s 2008 elections’, African Affairs,
108, 433, 621–41.
Widtfeldt, Anders (1997) Linking Parties with People?: Party Membership in Sweden
1960–1994, Göteborg, Göteborg University.
Williams, Michelle (2009) ‘Kirchheimer’s French Twist: a moden of the catch-all thesis
applied to the French case’, Party Politics, 15, 5, 592–614.
Witcover, Jules (2003) Party of the People: A History of the Democrats, New York,
Random House.
Woldendorp, Jaap, Hans Keman and Ian Budge (2000) Party Govcernment in 48
Democracies (1945–1998), Dordrecht, Kluwer Academic Publishers.
220 Bibliography
Wolf, Frieder Otto (2003) ‘Whatever happened to the German Greens?’, Red Pepper,
August 2003, online at: http://www.redpepper.org.uk/whatever-happened-to-the-
german/.
Wolinetz, Steven B. (2002) ‘Beyond the catch-all party: approaches to the study of
parties and party organizations in contemporary democracies’, in Richard Gunther,
Jose Ramon Montero and Juan J. Linz (eds), Political Parties: Old Concepts and
New Challenges, Oxford, Oxford University Press, 136–65.
Wring, Dominic (1996) ‘Political marketing and party development in Britain: a
“secret” history’, European Journal of Marketing, 30, 10/11, 92–103.
Yishai, Yael (2002) ‘Bringing society back in: post-cartel parties in Israel’, Party
Politics, 7, 6, 667–87.
Young, Lisa and William Cross (2002) ‘The rise of plebiscitary democracy in
Canadian political parties’, Party Politics, 8, 6, 673–99.
Index
activists, 12, 15, 38, 87, 103, 128, 130, conservative parties, 12, 44, 45, 47, 61, 62,
131–2, 143, 144, 150, 151, 152, 158, 63, 66, 68, 69, 73, 77, 164, 181, 195,
159 196
Africa, 30, 65, 109, 130, 140, 152, 165, Communist Party of Great Britain, 184
172, 202 Communist Party of the Soviet Union, 13,
ideology, 81–2 69
cleavage, political
von Beyme, Klaus, 67–8 centre-perihpery, 65–6
Bharatiya Janata Party (BJP), India, 11, 80 church-state, 66
blackmail potential, 27–8, 33 cultural, 72, 73–4
Blondel, Jean, 26–7 class, 66, 73
Botswana, 30–1 socio-economic, 72
Burke, Edmund, 5 Urban-rural, 67
clientilism, Africa, 82
cadre party, 50–2, 57, 138, 159, 163, 202, cordon sanitaire, 174
207 curvilinear disparity, law of, 131–2
campaign consultants, 144, 145–6, 147–8,
197–9 Democratic Party, USA, 11, 23, 24, 31,
Canada, 60, 90, 98, 106, 136 76–8, 140
candidate selection Denmark, 22, 23, 24, 32, 46, 48, 66, 85, 87,
context, 106 98, 114, 136, 152, 167
electoral systems, impact of, 106 Downs, Anthony, 6, 62, 92
funnel of, 105 Duverger, Maurice, 10, 12, 13, 25–6, 36–7,
gender, 118–123 44–7, 50–2, 88, 96, 107, 134, 136,
ideology, impact of, 107 159, 198, 206
political culture, impact of, 108
quotas in, 112 electoral-professional party, 53–6, 157
cartel party, 57–8, 128, 137, 159, 160, 165, Epstein, Leon, 6
202
catch-all party, 52–3, 75, 121, 137, 146, flawed democracies, 4
154, 157, 202 France, 7, 17, 28, 31, 32, 37, 41, 64, 75–6,
Chile, 66, 90, 122 90, 106, 112, 160, 164, 184
Christian Democratic parties, 66, 69, 67, 72, Freedom Party, Austria, 167
181 full democracies, 3
coalitions funding, 96–7, 120, 158–61, 185, 188
median legislator, 172–3 Fukuyama, Francis, 2
minimal connected winning, 172
minimal winning, 171
minimal range, 172 Gamson’s Law, 171, 175–6
minimum parties, 171–2 Germany, 30, 31, 32, 38, 41, 46, 66, 69, 70,
minimum seats, 171 90, 107, 137, 160, 183
winning, 170–1 Ghana, 11, 31, 81, 117
coalition potential, 27–8, 173 government potential, 173–4, 176
Conservative Party, Britain, 12, 13, 31, 113, government relevance, 173
139, 195 globalisation, 73, 74, 178, 180, 194–7, 202
221
222 Index
golden age of mass membership, 90–1 liberal, 68

regional, 70
India, 30, 33, 106, 152, 153, 155 right wing extremist, 70
ideology, 79–80 socialist and social democratic, 68–9
Indian National Congress, 79–80 party in central office, 8, 54, 189
Israel, 31, 33, 56, 66, 106, 112, 137 party in public office, 7–8, 44, 46, 54,
Italy, 30, 41, 56, 70, 108, 160 137–8, 189
party on the ground, 8, 44, 46, 54, 85, 87–8,
Japan, 30, 90 124, 137–8, 189
party system
Kirchheimer, Otto, 52–3, 154, 198 Blondel’s six types, 26–7
Duverger’s three types, 25–6
Labour Party, Britain, 11, 12, 31, 37, 46, 86, Sartori’s typology, 27–34
89, 139, 156, 164, 184 atomised, 33
liberal parties, 47, 68, 77, 195 hegemonic, 30
Lijphart, Arend, 41–2, 167 moderate pluralism, 31, 32–3
Linkage, 14–5, 18 one party, 30
local campaigning, 144, 149–152 polarised pluralism, 31–33
left–right ideological spectrum, 61–4 predominant, 30–31
two party, 31
market-oriented party, 149 streams of interaction in, 21–3
mass party, 50–2, 54–5, 57–8, 84, 90, 137,
145, 157, 160–1, 206, 207 post-materialism, 70–1, 72
Maastricht Treaty, 185, 188 product oriented party, 145
May, John D. 131–2
Mexico, 66, 90 religion, 5, 63, 66, 67, 72, 80, 81, 99
Michels, Robert, 2–3, 133–4, 136, 138 representative bureaucracy, 54
Mill, John Stuart, 128, 129 Republican Party, USA, 11, 22, 23, 31,
Miliband, Ralph, 129 76–8, 140
Rousseau, J. J., 16
National Democratic Congress, Ghana,
81–2 sales-oriented party, 147
New Patriotic Party, Ghana, 11, 81–2 Sartori, Giovanni, 27–33, 128, 168, 201
New Zealand, effective number of parties, Schattschneider, E. E., 1, 125–6, 205
34–5 Schumpeter, Joseph A., 125
Nice Treaty, 188 Social Democrats, Denmark, 11, 46, 87,
166, 174
Obama, Barack, 56, 96, 97, 109, 151, 156, Social Democrats, Norway, 55, 174
158, 199 Social Democrats, Sweden, 55, 139
organisational articulation, 12–3 Social Democratic Party, Germany, 46
Ostrogorski, M., 16–7, 104, 130, 207 socialist parties, 5, 12, 45, 47, 86, 87, 178
South Korea, 90
Panebianco, Angelo, 43, 47–50, 54–6 Spain, 31, 40, 65, 106, 108
party, definition of, 7 state funding, 160
party conference, 11
parties as empty vessels, 10 third way, 74–5
party family de Tocqueville, Alexis, 16
agrarian, 69–70
Christian democratic, 69 United Kingdom, effective number of parties,
conservative, 68 34
communist, 69 USA, 5, 9, 10, 22, 23, 26, 27, 31
ecology, 70, 72 ideology, 76–9
ethnic, see regional
green, see ecology Weimar Republic, 29, 32, 41, 183

Contemporary Party Politics

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Contemporary Party Politics

Uploaded by

Copyright:

Available Formats

Contemporary Party Politics

This page intentionally left blank

All rights reserved. No reproduction, copy or transmission of this

No portion of this publication may be reproduced, copied or transmitted

First published 2014 by

Palgrave Macmillan in the UK is an imprint of Macmillan Publishers Limited,

Palgrave Macmillan in the US is a division of St Martin’s Press LLC,

Palgrave Macmillan is the global academic imprint of the above companies

Palgrave® and Macmillan® are registered trademarks in the United States,

ISBN: 978-0-230-23779-7 hardback

3 Theories of Party Development 43

Ideology in contemporary European party politics 70

5 Party Members, Activists and Supporters 84

6 Candidate Selection 103

7 Policy Making 124

10 The Internationalisation of Party Politics 178

11 The Future of Party Politics 201

1.1 Confidence in political parties: Not very much/None at

6.1 Illustration of registration requirements for candidates

Political parties are the most important organisations in modern demo-

In other words if people who do not have significant personal resources

They define a flawed democracy as countries that:

Defining ‘political party’

Because parties are to be found in every democratic country, and in a

classic left–right view of ideology, these groups are defined as ‘classes’,

A political party is a group of people who, under a common label,

The three faces of party organisations

is the party on the ground. This is basically the extra-parliamentary mem-

The structure of parties

be at a minimum when there is not an election on, so as avoid local inter-

There are a number of constituent bodies that can be found in most

• The British Labour Party has its conference every year.

nature of the national committee, its general role is to provide leadership

The role of parties in democratic states

• Participatory linkage: is focused on giving citizens a way of partici-

A significant part of the debate in the context of linkage is about the

• Mobilisation and socialisation: parties mobilise citizens, not just to

The crisis of parties

Source: World Value survey 2005–8.

Outline of the book

One of the main concerns of political parties in their day-to-day activities

The importance of understanding party systems

The idea of party systems is a key element in understanding a country’s

Labour Party Conservative

Figure 2.1 Streams of interaction

Figure 2.2 Election results in Denmark (2011) and the UK (2010)

Sources: Statistics Denmark (2011) www.dst.dk/da/Statistik/emner/valg/folketingsvalg.aspx BBC

Kingdom) politics will be marked by confrontation and a lack of com-

Finally, and as an extension of what was said above, in comparative

Early approaches to party systems

1 The two-party system such as that found in the United States.

However, the problem with Blondel’s is that it is based very specifically

Party system typologies

Both Duverger and Blondel are important models to mention because

or coalition majority. On the other hand, a party, regardless of how

Party Party Party Party Party

Party Party Party Party Party

Figure 2.3 Segmentation and polarisation

However, focusing purely on the number of parties in a system is not

A ‘one party system’ is characterised by all power being concentrated in

held open, free and fair elections continuously since independence in

1 The existence of relevant anti-system parties, that is, parties that

2 Bi-lateral opposites, meaning the presence of extremist parties which

the party politics of that country more complicated. By contrast, it